1595 lines
91 KiB
Python
1595 lines
91 KiB
Python
|
import sys, yaml
|
|||
|
|
from easydict import EasyDict as edict
|
|||
|
|
from concurrent.futures import ThreadPoolExecutor
|
|||
|
|
|
|||
|
|
sys.path.extend(['..','../AIlib2' ])
|
|||
|
|
|
|||
|
|
from AI import AI_process,AI_process_forest,get_postProcess_para,get_postProcess_para_dic,ocr_process,AI_det_track,AI_det_track_batch,get_images_videos,AI_process_N,default_mix
|
|||
|
|
from stdc import stdcModel
|
|||
|
|
from yolov5 import yolov5Model
|
|||
|
|
|
|||
|
|
import cv2,os,time
|
|||
|
|
from segutils.segmodel import SegModel
|
|||
|
|
|
|||
|
|
from segutils.trafficUtils import tracfficAccidentMixFunction
|
|||
|
|
from models.experimental import attempt_load
|
|||
|
|
from utils.torch_utils import select_device
|
|||
|
|
from utils.plots import plot_one_box_PIL
|
|||
|
|
from utilsK.queRiver import get_labelnames,get_label_arrays,save_problem_images,riverDetSegMixProcess,draw_painting_joint
|
|||
|
|
from utilsK.drownUtils import mixDrowing_water_postprocess
|
|||
|
|
from ocrUtils.ocrUtils import CTCLabelConverter,AlignCollate
|
|||
|
|
from trackUtils.sort import Sort,track_draw_boxAndTrace,track_draw_trace_boxes,moving_average_wang,drawBoxTraceSimplied
|
|||
|
|
from obbUtils.load_obb_model import load_model_decoder_OBB
|
|||
|
|
from obbUtils.shipUtils import OBB_infer,draw_obb
|
|||
|
|
import numpy as np
|
|||
|
|
import torch,glob
|
|||
|
|
import tensorrt as trt
|
|||
|
|
from utilsK.masterUtils import get_needed_objectsIndex
|
|||
|
|
from utilsK.noParkingUtils import mixNoParking_road_postprocess
|
|||
|
|
from copy import deepcopy
|
|||
|
|
from scipy import interpolate
|
|||
|
|
#import warnings
|
|||
|
|
#warnings.filterwarnings("error")
|
|||
|
|
import inspect
|
|||
|
|
import psutil
|
|||
|
|
|
|||
|
|
def check_cpu(current_line_number):
|
|||
|
|
cpu_use = psutil.cpu_percent()
|
|||
|
|
cpu_mem = psutil.virtual_memory().percent
|
|||
|
|
cpu_swap = psutil.swap_memory().percent
|
|||
|
|
mem = psutil.virtual_memory()
|
|||
|
|
|
|||
|
|
# 已经使用的内存量（包括缓存和缓冲区）
|
|||
|
|
used_mem = mem.used/(1024**2)
|
|||
|
|
#current_line_number = inspect.currentframe().f_lineno
|
|||
|
|
print( '---line:{} ,CPUe使用率:{}, 内存使用:{},{:4.0f}M, SWAP内存使用率:{}'.format(current_line_number,cpu_use, cpu_mem,used_mem,cpu_swap) )
|
|||
|
|
|
|||
|
|
|
|||
|
|
|
|||
|
|
rainbows=[ [0,0,255],[0,255,0],[255,0,0],[255,0,255],[255,255,0],[255,129,0],[255,0,127],[127,255,0],[0,255,127],[0,127,255],[127,0,255],[255,127,255],[255,255,127],[127,255,255],[0,255,255],[255,127,255],[127,255,255], [0,127,0],[0,0,127],[0,255,255]]
|
|||
|
|
def view_bar(num, total,time1,prefix='prefix'):
|
|||
|
|
rate = num / total
|
|||
|
|
time_n=time.time()
|
|||
|
|
rate_num = int(rate * 30)
|
|||
|
|
rate_nums = np.round(rate * 100)
|
|||
|
|
r = '\r %s %d / %d [%s%s] %.2f s'%(prefix,num,total, ">" * rate_num, " " * (30 - rate_num), time_n-time1 )
|
|||
|
|
sys.stdout.write(r)
|
|||
|
|
sys.stdout.flush()
|
|||
|
|
|
|||
|
|
|
|||
|
|
'''
|
|||
|
|
多线程
|
|||
|
|
'''
|
|||
|
|
|
|||
|
|
def drawAllBox(preds,imgDraw,label_arraylist,rainbows,font):
|
|||
|
|
for box in preds:
|
|||
|
|
#cls,conf,xyxy = box[0],box[5], box[1:5]
|
|||
|
|
#print('#'*20,'line47',box)
|
|||
|
|
cls,conf,xyxy = box[5],box[4], box[0:4] ##2023.08.03,修改了格式
|
|||
|
|
#print('#####line46 demo.py:', cls,conf,xyxy, len(label_arraylist),len(rainbows) )
|
|||
|
|
imgDraw = draw_painting_joint(xyxy,imgDraw,label_arraylist[int(cls)],score=conf,color=rainbows[int(cls)%20],font=font,socre_location="leftTop")
|
|||
|
|
return imgDraw
|
|||
|
|
|
|||
|
|
def process_v1(frame):
|
|||
|
|
#try:
|
|||
|
|
|
|||
|
|
time00 = time.time()
|
|||
|
|
H,W,C = frame[0][0].shape
|
|||
|
|
#frmess---- (im0s,model,segmodel,names,label_arraylist,rainbows,objectPar,digitFont,os.path.basename(imgpath),segPar,mode,postPar)
|
|||
|
|
#p_result[1] = draw_painting_joint(xyxy,p_result[1],label_arraylist[int(cls)],score=conf,color=rainbows[int(cls)%20],font=font,socre_location="leftBottom")
|
|||
|
|
|
|||
|
|
p_result,timeOut = AI_process(frame[0],frame[1],frame[2],frame[3],frame[4],frame[5],objectPar=frame[6],font=frame[7],segPar=frame[9],mode=frame[10],postPar=frame[11])
|
|||
|
|
p_result[1] = drawAllBox(p_result[2],p_result[1],frame[4],frame[5],frame[7])
|
|||
|
|
|
|||
|
|
time11 = time.time()
|
|||
|
|
image_array = p_result[1]
|
|||
|
|
|
|||
|
|
cv2.imwrite(os.path.join('images/results/',frame[8] ) ,image_array)
|
|||
|
|
bname = frame[8].split('.')[0]
|
|||
|
|
if frame[2]:
|
|||
|
|
if len(p_result)==5:
|
|||
|
|
image_mask = p_result[4]
|
|||
|
|
if isinstance(image_mask,np.ndarray) and image_mask.shape[0]>0:
|
|||
|
|
|
|||
|
|
cv2.imwrite(os.path.join('images/results/',bname+'_mask.png' ) , (image_mask).astype(np.uint8))
|
|||
|
|
|
|||
|
|
boxes=p_result[2]
|
|||
|
|
with open( os.path.join('images/results/',bname+'.txt' ),'w' ) as fp:
|
|||
|
|
for box in boxes:
|
|||
|
|
box_str=[str(x) for x in box]
|
|||
|
|
out_str=','.join(box_str)+'\n'
|
|||
|
|
fp.write(out_str)
|
|||
|
|
time22 = time.time()
|
|||
|
|
print('%s,%d*%d,AI-process: %.1f,image save:%.1f , %s'%(frame[8],H,W, (time11 - time00) * 1000.0, (time22-time11)*1000.0,timeOut))
|
|||
|
|
return 'success'
|
|||
|
|
#except Exception as e:
|
|||
|
|
# return 'failed:'+str(e)
|
|||
|
|
def get_video_para(cap):
|
|||
|
|
fps = int(cap.get(cv2.CAP_PROP_FPS)+0.5)
|
|||
|
|
width = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH )+0.5)
|
|||
|
|
height = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT)+0.5)
|
|||
|
|
framecnt=int(cap.get(7)+0.5)
|
|||
|
|
return fps,width,height,framecnt
|
|||
|
|
|
|||
|
|
def process_video(video,par0,mode='detSeg'):
|
|||
|
|
cap=cv2.VideoCapture(video)
|
|||
|
|
if not cap.isOpened():
|
|||
|
|
print('#####error url:',video)
|
|||
|
|
return False
|
|||
|
|
#check_cpu(inspect.currentframe().f_lineno)
|
|||
|
|
bname=os.path.basename(video).split('.')[0]
|
|||
|
|
fps = int(cap.get(cv2.CAP_PROP_FPS)+0.5)
|
|||
|
|
width = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH )+0.5)
|
|||
|
|
height = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT)+0.5)
|
|||
|
|
framecnt=int(cap.get(7)+0.5)
|
|||
|
|
cap.set(cv2.CAP_PROP_BUFFERSIZE, 1)
|
|||
|
|
|
|||
|
|
save_path_AI = os.path.join(par0['outpth'],os.path.basename(video))
|
|||
|
|
problem_image_dir= os.path.join( par0['outpth'], 'probleImages' )
|
|||
|
|
os.makedirs(problem_image_dir,exist_ok=True)
|
|||
|
|
vid_writer_AI = cv2.VideoWriter(save_path_AI, cv2.VideoWriter_fourcc(*'mp4v'), fps, (width,height))
|
|||
|
|
num=0
|
|||
|
|
iframe=0;post_results=[];fpsample=30*10
|
|||
|
|
#check_cpu(inspect.currentframe().f_lineno)
|
|||
|
|
imgarray_list = []; iframe_list = []
|
|||
|
|
|
|||
|
|
#patch_cnt = par0['trackPar']['patchCnt']
|
|||
|
|
##windowsize 对逐帧插值后的结果做平滑，windowsize为平滑的长度,没隔det_cnt帧做一次跟踪。
|
|||
|
|
trackPar={'det_cnt':10,'windowsize':29 }
|
|||
|
|
|
|||
|
|
|
|||
|
|
##track_det_result_update= np.empty((0,8)) ###每100帧跑出来的结果，放在track_det_result_update，只保留当前100帧里有的tracker Id.
|
|||
|
|
#while cap.isOpened():
|
|||
|
|
while True:
|
|||
|
|
ret, imgarray = cap.read() #读取摄像头画面
|
|||
|
|
if not ret: break
|
|||
|
|
iframe +=1
|
|||
|
|
if not ret:break
|
|||
|
|
if mode=='detSeg':
|
|||
|
|
p_result,timeOut = AI_process([imgarray],par0['model'],par0['segmodel'],par0['names'],par0['label_arraylist'],par0['rainbows'],objectPar=par0['objectPar'],font=par0['digitFont'],segPar=par0['segPar'])
|
|||
|
|
|
|||
|
|
else:
|
|||
|
|
p_result,timeOut = AI_process_forest([imgarray],par0['model'],par0['segmodel'],par0['names'],par0['label_arraylist'],par0['rainbows'],par0['half'],par0['device'],par0['conf_thres'], par0['iou_thres'],par0['allowedList'],font=par0['digitFont'],trtFlag_det=par0['trtFlag_det'])
|
|||
|
|
|
|||
|
|
p_result[1] = drawAllBox(p_result[2],p_result[1],par0['label_arraylist'],par0['rainbows'],par0['digitFont'])
|
|||
|
|
#check_cpu(inspect.currentframe().f_lineno)
|
|||
|
|
if mode != 'track':
|
|||
|
|
image_array = p_result[1];num+=1
|
|||
|
|
ret = vid_writer_AI.write(image_array)
|
|||
|
|
view_bar(num, framecnt,time.time(),prefix=os.path.basename(video))
|
|||
|
|
##每隔 fpsample帧处理一次，如果有问题就保存图片
|
|||
|
|
if (iframe % fpsample == 0) and (len(post_results)>0) :
|
|||
|
|
parImage=save_problem_images(post_results,iframe,par0['names'],streamName=bname,outImaDir=problem_image_dir,imageTxtFile=False)
|
|||
|
|
post_results=[]
|
|||
|
|
|
|||
|
|
if len(p_result[2] )>0:
|
|||
|
|
post_results.append(p_result)
|
|||
|
|
#imgarray.release()
|
|||
|
|
|
|||
|
|
vid_writer_AI.release();
|
|||
|
|
def detSeg_demo(opt):
|
|||
|
|
|
|||
|
|
if opt['business'] == 'river':
|
|||
|
|
par={
|
|||
|
|
'device':'0', ###显卡号,如果用TRT模型，只支持0（单显卡）
|
|||
|
|
'labelnames':"../AIlib2/conf/river/labelnames.json", ###检测类别对照表
|
|||
|
|
'max_workers':1, ###并行线程数
|
|||
|
|
'Detweights':"../weights/%s/AIlib2/%s/yolov5_%s_fp16.engine"%(opt['gpu'], opt['business'] ,opt['gpu'] ),###检测模型路径
|
|||
|
|
'detModelpara':[{"id":str(x),"config":{"k1":"v1","k2":"v2"}} for x in [0,1,2,3,4,5,6,7,8,9] ],###控制哪些检测类别显示，输出
|
|||
|
|
'seg_nclass':2,###分割模型类别数目，默认2类
|
|||
|
|
'segRegionCnt':1,###分割模型结果需要保留的等值线数目
|
|||
|
|
'segPar':{'modelSize':(640,360),'mean':(0.485, 0.456, 0.406),'std' :(0.229, 0.224, 0.225),'numpy':False, 'RGB_convert_first':True,#分割模型预处理参数
|
|||
|
|
'mixFunction':{'function':riverDetSegMixProcess,'pars':{'slopeIndex':[1,3,4,7], 'riverIou':0.1}} #分割和检测混合处理的函数
|
|||
|
|
},
|
|||
|
|
'Segweights' : "../weights/%s/AIlib2/%s/stdc_360X640_%s_fp16.engine"%(opt['gpu'], opt['business'] ,opt['gpu'] ),###分割模型权重位置
|
|||
|
|
'postFile': '../AIlib2/conf/river/para.json',###后处理参数文件
|
|||
|
|
'txtFontSize':40,###文本字符的大小
|
|||
|
|
'digitFont': { 'line_thickness':2,'boxLine_thickness':1, 'fontSize':1.0,'waterLineColor':(0,255,255),'segLineShow':True,'waterLineWidth':3},###显示框、线、数字设置
|
|||
|
|
'testImgPath':'/mnt/thsw2/DSP2/videos/river/',
|
|||
|
|
'testOutPath':'images/results/',###输出测试图像位置
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
if opt['business'] == 'river2':
|
|||
|
|
par={
|
|||
|
|
'device':'0', ###显卡号,如果用TRT模型，只支持0（单显卡）
|
|||
|
|
'labelnames':"../AIlib2/conf/river2/labelnames.json", ###检测类别对照表
|
|||
|
|
'max_workers':1, ###并行线程数
|
|||
|
|
'Detweights':"../weights/%s/AIlib2/%s/yolov5_%s_fp16.engine"%(opt['gpu'], opt['business'] ,opt['gpu'] ),###检测模型路径
|
|||
|
|
'detModelpara':[{"id":str(x),"config":{"k1":"v1","k2":"v2"}} for x in [0,1,2,3,4,5,6,7,8,9] ],###控制哪些检测类别显示，输出
|
|||
|
|
'seg_nclass':2,###分割模型类别数目，默认2类
|
|||
|
|
'segRegionCnt':1,###分割模型结果需要保留的等值线数目
|
|||
|
|
'segPar':{'modelSize':(640,360),'mean':(0.485, 0.456, 0.406),'std' :(0.229, 0.224, 0.225),'numpy':False, 'RGB_convert_first':True,#分割模型预处理参数
|
|||
|
|
'mixFunction':{'function':riverDetSegMixProcess,'pars':{'slopeIndex':[1,3,4,7], 'riverIou':0.1}} #分割和检测混合处理的函数
|
|||
|
|
},
|
|||
|
|
'Segweights' : "../weights/%s/AIlib2/%s/stdc_360X640_%s_fp16.engine"%(opt['gpu'], opt['business'] ,opt['gpu'] ),###分割模型权重位置
|
|||
|
|
'postFile': '../AIlib2/conf/river2/para.json',###后处理参数文件
|
|||
|
|
'txtFontSize':40,###文本字符的大小
|
|||
|
|
'digitFont': { 'line_thickness':2,'boxLine_thickness':1, 'fontSize':1.0,'waterLineColor':(0,255,255),'segLineShow':True,'waterLineWidth':3},###显示框、线、数字设置
|
|||
|
|
'testImgPath':'images/river2/',
|
|||
|
|
'testOutPath':'images/results/',###输出测试图像位置
|
|||
|
|
}
|
|||
|
|
if opt['business'] == 'riverT':
|
|||
|
|
par={
|
|||
|
|
'device':'0', ###显卡号,如果用TRT模型，只支持0（单显卡）
|
|||
|
|
'labelnames':"../AIlib2/conf/riverT/labelnames.json", ###检测类别对照表
|
|||
|
|
'max_workers':1, ###并行线程数
|
|||
|
|
'Detweights':"../weights/%s/AIlib2/%s/yolov5_%s_fp16.engine"%(opt['gpu'], opt['business'] ,opt['gpu'] ),###检测模型路径
|
|||
|
|
'detModelpara':[{"id":str(x),"config":{"k1":"v1","k2":"v2"}} for x in [0,1,2,3,4,5,6,7,8,9] ],###控制哪些检测类别显示，输出
|
|||
|
|
'seg_nclass':2,###分割模型类别数目，默认2类
|
|||
|
|
'segRegionCnt':1,###分割模型结果需要保留的等值线数目
|
|||
|
|
'segPar':{'modelSize':(640,360),'mean':(0.485, 0.456, 0.406),'std' :(0.229, 0.224, 0.225),'numpy':False, 'RGB_convert_first':True,#分割模型预处理参数
|
|||
|
|
'mixFunction':{'function':riverDetSegMixProcess,'pars':{'slopeIndex':[1,3,4,7], 'riverIou':0.1}} #分割和检测混合处理的函数
|
|||
|
|
},
|
|||
|
|
'Segweights' : "../weights/%s/AIlib2/%s/stdc_360X640_%s_fp16.engine"%(opt['gpu'], opt['business'] ,opt['gpu'] ),###分割模型权重位置
|
|||
|
|
'postFile': '../AIlib2/conf/riverT/para.json',###后处理参数文件
|
|||
|
|
'txtFontSize':40,###文本字符的大小
|
|||
|
|
'digitFont': { 'line_thickness':2,'boxLine_thickness':1, 'fontSize':1.0,'waterLineColor':(0,255,255),'segLineShow':True,'waterLineWidth':3},###显示框、线、数字设置
|
|||
|
|
'testImgPath':'images/riverT/',
|
|||
|
|
'testOutPath':'images/results/',###输出测试图像位置
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
|
|||
|
|
|
|||
|
|
if opt['business'] == 'highWay2':
|
|||
|
|
par={
|
|||
|
|
'device':'0', ###显卡号,如果用TRT模型，只支持0（单显卡）
|
|||
|
|
'labelnames':"../AIlib2/conf/%s/labelnames.json"%(opt['business']), ###检测类别对照表
|
|||
|
|
'max_workers':1, ###并行线程数
|
|||
|
|
'Detweights':"../weights/%s/AIlib2/%s/yolov5_%s_fp16.engine"%(opt['gpu'], opt['business'] ,opt['gpu'] ),###检测模型路径
|
|||
|
|
'detModelpara':[{"id":str(x),"config":{"k1":"v1","k2":"v2"}} for x in [0,1,2,3,5,6,7,8,9] ],###控制哪些检测类别显示、输出
|
|||
|
|
|
|||
|
|
'seg_nclass':3,###分割模型类别数目，默认2类
|
|||
|
|
'segRegionCnt':2,###分割模型结果需要保留的等值线数目
|
|||
|
|
'segPar':{
|
|||
|
|
'modelSize':(640,360),
|
|||
|
|
#'modelSize':(1920,1080),
|
|||
|
|
'mean':(0.485, 0.456, 0.406),'std' :(0.229, 0.224, 0.225),'predResize':True,'numpy':False, 'RGB_convert_first':True,###分割模型预处理参数
|
|||
|
|
'mixFunction':{'function':tracfficAccidentMixFunction,
|
|||
|
|
'pars':{ 'RoadArea': 16000, 'roadVehicleAngle': 15, 'speedRoadVehicleAngleMax': 75, 'roundness': 1.0, 'cls': 9, 'vehicleFactor': 0.1, 'confThres':0.25,'roadIou':0.6,'radius': 50 ,'vehicleFlag':False,'distanceFlag': False}
|
|||
|
|
|
|||
|
|
}
|
|||
|
|
},
|
|||
|
|
|
|||
|
|
'Segweights' : "../weights/%s/AIlib2/%s/stdc_360X640_%s_fp16.engine"%(opt['gpu'], opt['business'] ,opt['gpu'] ),###分割模型权重位置
|
|||
|
|
#'Segweights' : "../weights/%s/AIlib2/%s/stdc_1080X1920_%s_fp16.engine"%(opt['gpu'], opt['business'] ,opt['gpu'] ),###分割模型权重位置
|
|||
|
|
#'Segweights' :'/mnt/thsw2/DSP2/weights/highWay2/stdc_360X640.pth',
|
|||
|
|
'postFile': '../AIlib2/conf/%s/para.json'%(opt['business'] ),###后处理参数文件
|
|||
|
|
'txtFontSize':20,###文本字符的大小
|
|||
|
|
'digitFont': { 'line_thickness':2,'boxLine_thickness':1, 'fontSize':1.0,'waterLineColor':(0,255,255),'segLineShow':True,'waterLineWidth':2},###显示框、线设置
|
|||
|
|
#'testImgPath':'images/highWayTest/',###测试图像的位置
|
|||
|
|
'testImgPath':'images/tt',
|
|||
|
|
#'testImgPath':'/mnt/thsw2/DSP2/highWay2/videos/',
|
|||
|
|
'testOutPath':'images/results/',###输出测试图像位置
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
par['segPar']['mixFunction']['pars']['modelSize'] = par['segPar']['modelSize']
|
|||
|
|
if opt['business'] == 'drowning':
|
|||
|
|
par={
|
|||
|
|
'device':'0', ###显卡号,如果用TRT模型，只支持0（单显卡）
|
|||
|
|
'labelnames':"../AIlib2/conf/%s/labelnames.json"%(opt['business']), ###检测类别对照表
|
|||
|
|
'max_workers':1, ###并行线程数
|
|||
|
|
'Detweights':"../weights/%s/AIlib2/%s/yolov5_%s_fp16.engine"%(opt['gpu'], opt['business'] ,opt['gpu'] ),###检测模型路径
|
|||
|
|
#'Detweights':"/mnt/thsw2/DSP2/weights/drowning/yolov5.pt",
|
|||
|
|
'detModelpara':[{"id":str(x),"config":{"k1":"v1","k2":"v2"}} for x in [0,1,2,3,5,6,7,8,9] ],###控制哪些检测类别显示、输出
|
|||
|
|
'seg_nclass':2,###分割模型类别数目，默认2类
|
|||
|
|
'segRegionCnt':2,###分割模型结果需要保留的等值线数目
|
|||
|
|
'segPar':{'modelSize':(640,360),'mean':(0.485, 0.456, 0.406),'std' :(0.229, 0.224, 0.225),'predResize':True,'numpy':False, 'RGB_convert_first':True,###分割模型预处理参数
|
|||
|
|
'mixFunction':{'function':mixDrowing_water_postprocess,
|
|||
|
|
'pars':{ }
|
|||
|
|
}
|
|||
|
|
},
|
|||
|
|
|
|||
|
|
|
|||
|
|
'Segweights' : "../weights/%s/AIlib2/%s/stdc_360X640_%s_fp16.engine"%(opt['gpu'], opt['business'] ,opt['gpu'] ),###分割模型权重位置
|
|||
|
|
|
|||
|
|
#'Segweights' : "/mnt/thsw2/DSP2/weights/drowning/stdc_360X640_2080Ti_fp16.engine",
|
|||
|
|
|
|||
|
|
'postFile': '../AIlib2/conf/%s/para.json'%(opt['business'] ),###后处理参数文件
|
|||
|
|
'txtFontSize':20,###文本字符的大小
|
|||
|
|
'digitFont': { 'line_thickness':2,'boxLine_thickness':1, 'fontSize':1.0,'waterLineColor':(0,255,255),'segLineShow':True,'waterLineWidth':2},###显示框、线设置
|
|||
|
|
#'testImgPath':'/mnt/thsw2/DSP2/videos/drowning/',###测试图像的位置
|
|||
|
|
'testImgPath':'images/drowning/',
|
|||
|
|
'testOutPath':'images/results/',###输出测试图像位置
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
par['segPar']['mixFunction']['pars']['modelSize'] = par['segPar']['modelSize']
|
|||
|
|
if opt['business'] == 'noParking':
|
|||
|
|
par={
|
|||
|
|
'device':'0', ###显卡号,如果用TRT模型，只支持0（单显卡）
|
|||
|
|
'labelnames':"../AIlib2/conf/%s/labelnames.json"%(opt['business']), ###检测类别对照表
|
|||
|
|
'max_workers':1, ###并行线程数
|
|||
|
|
'Detweights':"../weights/%s/AIlib2/%s/yolov5_%s_fp16.engine"%(opt['gpu'], opt['business'] ,opt['gpu'] ),###检测模型路径
|
|||
|
|
'detModelpara':[{"id":str(x),"config":{"k1":"v1","k2":"v2"}} for x in [0,1,2,3,5,6,7,8,9] ],###控制哪些检测类别显示、输出
|
|||
|
|
'seg_nclass':4,###分割模型类别数目，默认2类
|
|||
|
|
'segRegionCnt':2,###分割模型结果需要保留的等值线数目
|
|||
|
|
|
|||
|
|
'segPar':{'modelSize':(640,360),'mean':(0.485, 0.456, 0.406),'std' :(0.229, 0.224, 0.225),'predResize':True,'numpy':False, 'RGB_convert_first':True,###分割模型预处理参数
|
|||
|
|
'mixFunction':{'function':mixNoParking_road_postprocess,
|
|||
|
|
'pars':{ 'roundness': 0.3, 'cls': 9, 'laneArea': 10, 'laneAngleCha': 5 ,'RoadArea': 16000,'fitOrder':2}
|
|||
|
|
|
|||
|
|
|
|||
|
|
}
|
|||
|
|
},
|
|||
|
|
|
|||
|
|
'Segweights' : "../weights/%s/AIlib2/%s/stdc_360X640_%s_fp16.engine"%(opt['gpu'], opt['business'] ,opt['gpu'] ),###分割模型权重位置
|
|||
|
|
'postFile': '../AIlib2/conf/%s/para.json'%(opt['business'] ),###后处理参数文件
|
|||
|
|
'txtFontSize':20,###文本字符的大小
|
|||
|
|
'digitFont': { 'line_thickness':2,'boxLine_thickness':1, 'fontSize':1.0,'waterLineColor':(0,255,255),'segLineShow':True,'waterLineWidth':2},###显示框、线设置
|
|||
|
|
'testImgPath':'images/noParking/',###测试图像的位置
|
|||
|
|
'testOutPath':'images/results/',###输出测试图像位置
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
par['segPar']['mixFunction']['pars']['modelSize'] = par['segPar']['modelSize']
|
|||
|
|
if opt['business'] == 'illParking':
|
|||
|
|
from utilsK.illParkingUtils import illParking_postprocess
|
|||
|
|
par={
|
|||
|
|
'device':'0', ###显卡号,如果用TRT模型，只支持0（单显卡）
|
|||
|
|
'labelnames':"../AIlib2/conf/%s/labelnames.json"%(opt['business']), ###检测类别对照表
|
|||
|
|
'max_workers':1, ###并行线程数
|
|||
|
|
'Detweights':"../weights/%s/AIlib2/%s/yolov5_%s_fp16.engine"%(opt['gpu'], opt['business'] ,opt['gpu'] ),###检测模型路径
|
|||
|
|
'detModelpara':[{"id":str(x),"config":{"k1":"v1","k2":"v2"}} for x in [0,1,2,3,5,6,7,8,9] ],###控制哪些检测类别显示、输出
|
|||
|
|
'seg_nclass':4,###分割模型类别数目，默认2类
|
|||
|
|
'segRegionCnt':2,###分割模型结果需要保留的等值线数目
|
|||
|
|
|
|||
|
|
'segPar':{
|
|||
|
|
'mixFunction':{'function':illParking_postprocess,
|
|||
|
|
'pars':{ }
|
|||
|
|
}
|
|||
|
|
},
|
|||
|
|
|
|||
|
|
'Segweights' : None,###分割模型权重位置
|
|||
|
|
'postFile': '../AIlib2/conf/%s/para.json'%(opt['business'] ),###后处理参数文件
|
|||
|
|
'txtFontSize':20,###文本字符的大小
|
|||
|
|
'digitFont': { 'line_thickness':2,'boxLine_thickness':1, 'fontSize':1.0,'waterLineColor':(0,255,255),'segLineShow':True,'waterLineWidth':2},###显示框、线设置
|
|||
|
|
'testImgPath':'images/cityMangement',###测试图像的位置
|
|||
|
|
'testOutPath':'images/results/',###输出测试图像位置
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
if opt['business'] == 'cityMangement2':
|
|||
|
|
from DMPR import DMPRModel
|
|||
|
|
from DMPRUtils.jointUtil import dmpr_yolo
|
|||
|
|
|
|||
|
|
|
|||
|
|
par={
|
|||
|
|
'device':'0', ###显卡号,如果用TRT模型，只支持0（单显卡）
|
|||
|
|
'labelnames':"../AIlib2/conf/%s/labelnames.json"%(opt['business']), ###检测类别对照表
|
|||
|
|
'max_workers':1, ###并行线程数
|
|||
|
|
'Detweights':"../weights/%s/AIlib2/%s/yolov5_%s_fp16.engine"%(opt['gpu'], opt['business'] ,opt['gpu'] ),###检测模型路径
|
|||
|
|
'detModelpara':[{"id":str(x),"config":{"k1":"v1","k2":"v2"}} for x in [0,1,2,3,5,6,7,8,9] ],###控制哪些检测类别显示、输出
|
|||
|
|
'seg_nclass':4,###分割模型类别数目，默认2类
|
|||
|
|
'segRegionCnt':2,###分割模型结果需要保留的等值线数目
|
|||
|
|
|
|||
|
|
'segPar':{ 'depth_factor':32,'NUM_FEATURE_MAP_CHANNEL':6,'dmpr_thresh':0.3, 'dmprimg_size':640,
|
|||
|
|
'mixFunction':{'function':dmpr_yolo,
|
|||
|
|
'pars':{'carCls':0 ,'illCls':3,'scaleRatio':0.5,'border':80}
|
|||
|
|
}
|
|||
|
|
},
|
|||
|
|
'Segweights':"../weights/%s/AIlib2/%s/dmpr_%s.engine"%(opt['gpu'], opt['business'] ,opt['gpu'] ),###检测模型路径
|
|||
|
|
#'Segweights':"../AIlib2/conf/cityMangement2/dmpr.pth",###检测模型路径
|
|||
|
|
'postFile': '../AIlib2/conf/%s/para.json'%(opt['business'] ),###后处理参数文件
|
|||
|
|
'txtFontSize':20,###文本字符的大小
|
|||
|
|
'digitFont': { 'line_thickness':2,'boxLine_thickness':1, 'fontSize':1.0,'waterLineColor':(0,255,255),'segLineShow':True,'waterLineWidth':2},###显示框、线设置
|
|||
|
|
'testImgPath':'images/cityMangement2/',
|
|||
|
|
'testOutPath':'images/results/',###输出测试图像位置
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
|
|||
|
|
|
|||
|
|
if par['Segweights']:
|
|||
|
|
par['trtFlag_seg']=True if par['Segweights'].endswith('.engine') else False
|
|||
|
|
else:
|
|||
|
|
par['trtFlag_seg']=False
|
|||
|
|
|
|||
|
|
|
|||
|
|
par['trtFlag_det']=True if par['Detweights'].endswith('.engine') else False
|
|||
|
|
|
|||
|
|
mode = par['mode'] if 'mode' in par.keys() else 'others'
|
|||
|
|
postPar = par['postPar'] if 'postPar' in par.keys() else None
|
|||
|
|
|
|||
|
|
device_=par['device']
|
|||
|
|
labelnames = par['labelnames'] ##对应类别表
|
|||
|
|
|
|||
|
|
max_workers=par['max_workers'];
|
|||
|
|
|
|||
|
|
trtFlag_det=par['trtFlag_det'];trtFlag_seg=par['trtFlag_seg'];segRegionCnt=par['segRegionCnt']
|
|||
|
|
device = select_device(device_)
|
|||
|
|
names=get_labelnames(labelnames)
|
|||
|
|
|
|||
|
|
half = device.type != 'cpu' # half precision only supported on CUDA
|
|||
|
|
if trtFlag_det:
|
|||
|
|
Detweights = par['Detweights']##升级后的检测模型
|
|||
|
|
logger = trt.Logger(trt.Logger.ERROR)
|
|||
|
|
with open(Detweights, "rb") as f, trt.Runtime(logger) as runtime:
|
|||
|
|
model=runtime.deserialize_cuda_engine(f.read())# 输入trt本地文件，返回ICudaEngine对象
|
|||
|
|
print('############locad det model trtsuccess:',Detweights)
|
|||
|
|
else:
|
|||
|
|
Detweights = par['Detweights']
|
|||
|
|
model = attempt_load(Detweights, map_location=device) # load FP32 model
|
|||
|
|
print('############locad det model pth success:',Detweights)
|
|||
|
|
if half: model.half()
|
|||
|
|
|
|||
|
|
#check_cpu(inspect.currentframe().f_lineno)
|
|||
|
|
par['segPar']['seg_nclass'] = par['seg_nclass']
|
|||
|
|
segPar=par['segPar']
|
|||
|
|
if par['Segweights']:
|
|||
|
|
if opt['business'] == 'cityMangement2':
|
|||
|
|
segmodel = DMPRModel(weights=par['Segweights'], par = par['segPar'])
|
|||
|
|
else:
|
|||
|
|
segmodel = stdcModel(weights=par['Segweights'], par = par['segPar'])
|
|||
|
|
else:
|
|||
|
|
segmodel= None
|
|||
|
|
print('############None seg model is loaded###########:' )
|
|||
|
|
#check_cpu(inspect.currentframe().f_lineno)
|
|||
|
|
postFile= par['postFile']
|
|||
|
|
digitFont= par['digitFont']
|
|||
|
|
#conf_thres,iou_thres,classes,rainbows=get_postProcess_para(postFile)
|
|||
|
|
detPostPar = get_postProcess_para_dic(postFile)
|
|||
|
|
conf_thres,iou_thres,classes,rainbows = detPostPar["conf_thres"],detPostPar["iou_thres"],detPostPar["classes"],detPostPar["rainbows"]
|
|||
|
|
if 'ovlap_thres_crossCategory' in detPostPar.keys(): ovlap_thres_crossCategory=detPostPar['ovlap_thres_crossCategory']
|
|||
|
|
else:ovlap_thres_crossCategory = None
|
|||
|
|
|
|||
|
|
if 'score_byClass' in detPostPar.keys(): score_byClass=detPostPar['score_byClass']
|
|||
|
|
else: score_byClass = None
|
|||
|
|
#check_cpu(inspect.currentframe().f_lineno)
|
|||
|
|
####模型选择参数用如下：
|
|||
|
|
mode_paras=par['detModelpara']
|
|||
|
|
allowedList,allowedList_string=get_needed_objectsIndex(mode_paras)
|
|||
|
|
#allowedList=[0,1,2,3]
|
|||
|
|
##加载模型，准备好显示字符
|
|||
|
|
label_arraylist = get_label_arrays(names,rainbows,outfontsize=par['txtFontSize'],fontpath="../AIlib2/conf/platech.ttf")
|
|||
|
|
|
|||
|
|
|
|||
|
|
#check_cpu(inspect.currentframe().f_lineno)
|
|||
|
|
##图像测试
|
|||
|
|
#impth = 'images/slope/'
|
|||
|
|
impth = par['testImgPath']
|
|||
|
|
outpth = par['testOutPath']
|
|||
|
|
imgpaths=[]###获取文件里所有的图像
|
|||
|
|
for postfix in ['.jpg','.JPG','.PNG','.png']:
|
|||
|
|
imgpaths.extend(glob.glob('%s/*%s'%(impth,postfix )) )
|
|||
|
|
videopaths=[]###获取文件里所有的视频
|
|||
|
|
for postfix in ['.MP4','.mp4','.avi']:
|
|||
|
|
videopaths.extend(glob.glob('%s/*%s'%(impth,postfix )) )
|
|||
|
|
|
|||
|
|
#check_cpu(inspect.currentframe().f_lineno)
|
|||
|
|
###先处理图像
|
|||
|
|
frames=[]
|
|||
|
|
for imgpath in imgpaths:
|
|||
|
|
im0s=[cv2.imread(imgpath)]
|
|||
|
|
objectPar={ 'half':half,'device':device,'conf_thres':conf_thres,'ovlap_thres_crossCategory':ovlap_thres_crossCategory,'iou_thres':iou_thres,'allowedList':allowedList,'segRegionCnt':segRegionCnt, 'trtFlag_det':trtFlag_det,'trtFlag_seg':trtFlag_seg ,'score_byClass':score_byClass}
|
|||
|
|
|
|||
|
|
#p_result[1] = draw_painting_joint(xyxy,p_result[1],label_arraylist[int(cls)],score=conf,color=rainbows[int(cls)%20],font=font,socre_location="leftBottom")
|
|||
|
|
|
|||
|
|
frame=(im0s,model,segmodel,names,label_arraylist,rainbows,objectPar,digitFont,os.path.basename(imgpath),segPar,mode,postPar)
|
|||
|
|
frames.append(frame)
|
|||
|
|
t1=time.time()
|
|||
|
|
if max_workers==1:
|
|||
|
|
for i in range(len(imgpaths)):
|
|||
|
|
print('-'*20,imgpaths[i],'-'*20)
|
|||
|
|
t5=time.time()
|
|||
|
|
process_v1(frames[i])
|
|||
|
|
t6=time.time()
|
|||
|
|
#print('#######%s, ms:%.1f , accumetate time:%.1f, avage:%1.f '%(os.path.basename(imgpaths[i]), (t6-t5)*1000.0,(t6-t1)*1000.0, (t6-t1)*1000.0/(i+1)))
|
|||
|
|
else:
|
|||
|
|
with ThreadPoolExecutor(max_workers=max_workers) as t:
|
|||
|
|
for result in t.map(process_v1, frames):
|
|||
|
|
#print(result)
|
|||
|
|
t=result
|
|||
|
|
|
|||
|
|
t2=time.time()
|
|||
|
|
if len(imgpaths)>0:
|
|||
|
|
print('All %d images time:%.1f ms ,each:%.1f ms, with %d threads'%(len(imgpaths),(t2-t1)*1000, (t2-t1)*1000.0/len(imgpaths) , max_workers) )
|
|||
|
|
|
|||
|
|
|
|||
|
|
#check_cpu(inspect.currentframe().f_lineno)
|
|||
|
|
objectPar={ 'half':half,'device':device,'conf_thres':conf_thres,'iou_thres':iou_thres,'allowedList':allowedList,'segRegionCnt':segRegionCnt, 'trtFlag_det':trtFlag_det,'trtFlag_seg':trtFlag_seg }
|
|||
|
|
|
|||
|
|
par0={ 'model':model,'segmodel':segmodel,'names':names,'label_arraylist':label_arraylist,'rainbows':rainbows,
|
|||
|
|
'objectPar':objectPar,'digitFont':digitFont,'segPar':segPar,'outpth':outpth
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
|
|||
|
|
|
|||
|
|
|
|||
|
|
###如果是视频文件
|
|||
|
|
for video in videopaths:
|
|||
|
|
process_video(video,par0)
|
|||
|
|
print(' ')
|
|||
|
|
def detSeg_demo2(opt):
|
|||
|
|
|
|||
|
|
if opt['business'] == 'cityMangement3':
|
|||
|
|
from DMPR import DMPRModel
|
|||
|
|
from DMPRUtils.jointUtil import dmpr_yolo_stdc
|
|||
|
|
par={
|
|||
|
|
'device':'0', ###显卡号,如果用TRT模型，只支持0（单显卡）
|
|||
|
|
'labelnames':"../AIlib2/conf/%s/labelnames.json"%(opt['business']), ###检测类别对照表
|
|||
|
|
'max_workers':1, ###并行线程数
|
|||
|
|
'postProcess':{
|
|||
|
|
'function':dmpr_yolo_stdc,
|
|||
|
|
'pars':{'carCls':0 ,'illCls':5,'scaleRatio':0.5,'border':80,
|
|||
|
|
#车辆","垃圾","商贩","裸土","占道经营","违停"--->
|
|||
|
|
#"车辆","垃圾","商贩","违停","占道经营","裸土"
|
|||
|
|
'classReindex':{ 0:0,1:1,2:2,3:5,4:4,5:3 }
|
|||
|
|
}
|
|||
|
|
},
|
|||
|
|
'models':[
|
|||
|
|
{
|
|||
|
|
# 'weight':"../weights/%s/AIlib2/%s/yolov5_%s_fp16.engine"%(opt['gpu'], opt['business'] ,opt['gpu'] ),###检测模型路径
|
|||
|
|
'weight':'../weights/pth/AIlib2/%s/yolov5.pt'%(opt['business'] ),
|
|||
|
|
#"weight":'/mnt/thsw2/DSP2/cityMangement3/weights/yolov5.pt',
|
|||
|
|
'name':'yolov5',
|
|||
|
|
'model':yolov5Model,
|
|||
|
|
'par':{ 'half':True,'device':'cuda:0' ,'conf_thres':0.55,'iou_thres':0.45,'allowedList':[0,1,2,3,4,5],'segRegionCnt':1, 'trtFlag_det':False,'trtFlag_seg':False, "score_byClass":{"0":0.25,"1":0.3,"2":0.3,"3":0.5 ,"4":0.5,"5":0.5 } }
|
|||
|
|
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
#'weight':"../weights/%s/AIlib2/%s/dmpr_%s.engine"%(opt['gpu'], opt['business'] ,opt['gpu'] ),###检测模型路径
|
|||
|
|
#'weight':'../weights/pth/AIlib2/%s/dmpr.pth'%(opt['business'] ),
|
|||
|
|
#'weight':'/mnt/thsw2/DSP2/cityMangement3/weights/dmpr_20231202.pth',
|
|||
|
|
'weight':'/mnt/thsw2/DSP2/cityMangement3/weights/dmpr.pth',
|
|||
|
|
'par':{
|
|||
|
|
'depth_factor':32,'NUM_FEATURE_MAP_CHANNEL':6,'dmpr_thresh':0.1, 'dmprimg_size':640,
|
|||
|
|
'name':'dmpr'
|
|||
|
|
},
|
|||
|
|
'model':DMPRModel,
|
|||
|
|
'name':'dmpr'
|
|||
|
|
},
|
|||
|
|
{
|
|||
|
|
#'weight':"../weights/%s/AIlib2/%s/dmpr_%s.engine"%(opt['gpu'], opt['business'] ,opt['gpu'] ),###检测模型路径
|
|||
|
|
'weight':'../weights/pth/AIlib2/%s/stdc_360X640.pth'%(opt['business'] ),
|
|||
|
|
'par':{
|
|||
|
|
'modelSize':(640,360),'mean':(0.485, 0.456, 0.406),'std' :(0.229, 0.224, 0.225),'predResize':True,'numpy':False, 'RGB_convert_first':True,'seg_nclass':2},###分割模型预处理参数
|
|||
|
|
'model':stdcModel,
|
|||
|
|
'name':'stdc'
|
|||
|
|
|
|||
|
|
}
|
|||
|
|
],
|
|||
|
|
|
|||
|
|
'detModelpara':[{"id":str(x),"config":{"k1":"v1","k2":"v2"}} for x in [0,1,2,3,5,6,7,8,9] ],###控制哪些检测类别显示、输出
|
|||
|
|
'segRegionCnt':2,###分割模型结果需要保留的等值线数目
|
|||
|
|
'postFile': '../AIlib2/conf/%s/para.json'%(opt['business'] ),###后处理参数文件
|
|||
|
|
'txtFontSize':20,###文本字符的大小
|
|||
|
|
'digitFont': { 'line_thickness':2,'boxLine_thickness':1, 'fontSize':1.0,'waterLineColor':(0,255,255),'segLineShow':True,'waterLineWidth':2},###显示框、线设置
|
|||
|
|
#'testImgPath':'images/%s/'%(opt['business']),
|
|||
|
|
#'testImgPath':'images/tt',
|
|||
|
|
'testImgPath':'/mnt/thsw2/DSP2/cityMangement3/images',
|
|||
|
|
'testOutPath':'images/results/',###输出测试图像位置
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
if opt['business'] == 'highWaySpill':
|
|||
|
|
from utilsK.spillUtils import mixSpillage_postprocess
|
|||
|
|
|
|||
|
|
par={
|
|||
|
|
'device':'0', ###显卡号,如果用TRT模型，只支持0（单显卡）
|
|||
|
|
'labelnames':"../AIlib2/conf/%s/labelnames.json"%(opt['business']), ###检测类别对照表
|
|||
|
|
'max_workers':1, ###并行线程数
|
|||
|
|
'postProcess':{
|
|||
|
|
'function':mixSpillage_postprocess,
|
|||
|
|
'pars':{}
|
|||
|
|
},
|
|||
|
|
'models':[
|
|||
|
|
{
|
|||
|
|
# 'weight':"../weights/%s/AIlib2/%s/yolov5_%s_fp16.engine"%(opt['gpu'], opt['business'] ,opt['gpu'] ),###检测模型路径
|
|||
|
|
#'weight':'../weights/pth/AIlib2/%s/yolov5.pt'%(opt['business'] ),
|
|||
|
|
"weight":'../weights/pth/AIlib2/%s/yolov5.pt'%( opt['business'] ),
|
|||
|
|
'name':'yolov5',
|
|||
|
|
'model':yolov5Model,
|
|||
|
|
'par':{ 'half':True,'device':'cuda:0' ,'conf_thres':0.55,'iou_thres':0.45,'allowedList':[0,1,2,3,4,5],'segRegionCnt':1, 'trtFlag_det':False,'trtFlag_seg':False, "score_byClass":{"0":0.25,"1":0.3,"2":0.3,"3":0.5 ,"4":0.5,"5":0.5 } }
|
|||
|
|
|
|||
|
|
},
|
|||
|
|
|
|||
|
|
{
|
|||
|
|
|
|||
|
|
#'weight':'../weights/pth/AIlib2/highWay2/stdc_360X640.pth',
|
|||
|
|
'weight':'../weights/%s/AIlib2/highWay2/stdc_360X640_%s_fp16.engine'%(opt['gpu'],opt['gpu']),
|
|||
|
|
'par':{
|
|||
|
|
'modelSize':(640,360),'mean':(0.485, 0.456, 0.406),'std' :(0.229, 0.224, 0.225),'predResize':True,'numpy':False, 'RGB_convert_first':True,'seg_nclass':3},###分割模型预处理参数
|
|||
|
|
'model':stdcModel,
|
|||
|
|
'name':'stdc'
|
|||
|
|
|
|||
|
|
}
|
|||
|
|
],
|
|||
|
|
|
|||
|
|
'detModelpara':[{"id":str(x),"config":{"k1":"v1","k2":"v2"}} for x in [0,1,2,3,5,6,7,8,9] ],###控制哪些检测类别显示、输出
|
|||
|
|
'segRegionCnt':2,###分割模型结果需要保留的等值线数目
|
|||
|
|
'postFile': '../AIlib2/conf/%s/para.json'%(opt['business'] ),###后处理参数文件
|
|||
|
|
'txtFontSize':20,###文本字符的大小
|
|||
|
|
'digitFont': { 'line_thickness':2,'boxLine_thickness':1, 'fontSize':1.0,'waterLineColor':(0,255,255),'segLineShow':True,'waterLineWidth':2},###显示框、线设置
|
|||
|
|
#'testImgPath':'images/%s/'%(opt['business']),
|
|||
|
|
#'testImgPath':'images/tt',
|
|||
|
|
'testImgPath':'/mnt/thsw2/DSP2/%s/images'%(opt['business']),
|
|||
|
|
'testOutPath':'images/results/',###输出测试图像位置
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
|
|||
|
|
|
|||
|
|
if opt['business'] == 'forest2':
|
|||
|
|
from utilsK.crowdGather import gather_post_process
|
|||
|
|
par={
|
|||
|
|
'device':'0', ###显卡号,如果用TRT模型，只支持0（单显卡）
|
|||
|
|
'labelnames':"../AIlib2/conf/forest2/labelnames.json", ###检测类别对照表
|
|||
|
|
'max_workers':1, ###并行线程数
|
|||
|
|
|
|||
|
|
'postProcess':{'function':default_mix,'pars':{}},
|
|||
|
|
'models':
|
|||
|
|
[
|
|||
|
|
{
|
|||
|
|
#'weight':"../weights/%s/AIlib2/forest2/yolov5_%s_fp16.engine"%(opt['gpu'], opt['gpu'] ),###检测模型路径
|
|||
|
|
'weight':'../weights/pth/AIlib2/forestCrowd/yolov5.pt',###检测模型路径
|
|||
|
|
#'weight':'../weights/pth/AIlib2/%s/yolov5.pt'%(opt['business'] ),
|
|||
|
|
'name':'yolov5',
|
|||
|
|
'model':yolov5Model,
|
|||
|
|
'par':{ 'half':True,'device':'cuda:0' ,'conf_thres':0.25,'iou_thres':0.45,'allowedList':[0,1,2,3],'segRegionCnt':1, 'trtFlag_det':False,'trtFlag_seg':False, "score_byClass":{"0":0.25,"1":0.3,"2":0.3,"3":0.3 } },
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
|
|||
|
|
],
|
|||
|
|
'detModelpara':[{"id":str(x),"config":{"k1":"v1","k2":"v2"}} for x in [] ],
|
|||
|
|
'postFile': '../AIlib2/conf/forest/para.json',###后处理参数文件
|
|||
|
|
'txtFontSize':50,###文本字符的大小
|
|||
|
|
'digitFont': { 'line_thickness':1,'boxLine_thickness':1, 'fontSize':0.5,'waterLineColor':(0,255,255),'waterLineWidth':1},###显示框、线设置
|
|||
|
|
#'testImgPath':'../AIdemo2/images/tt/',###测试图像的位置
|
|||
|
|
#'testImgPath':'images/smogfire',
|
|||
|
|
'testImgPath':'/mnt/thsw2/DSP2/forest2/videos',
|
|||
|
|
#'testImgPath':'/mnt/thsw2/DSP2/weights/forest2/images_cloud_FP',
|
|||
|
|
#'testImgPath':'/mnt/thsw2/DSP2/weights/forest2/images_visdrone',
|
|||
|
|
'testOutPath':'images/results/',###输出测试图像位置
|
|||
|
|
}
|
|||
|
|
if opt['business'] == 'forestCrowd':
|
|||
|
|
from utilsK.crowdGather import gather_post_process
|
|||
|
|
par={
|
|||
|
|
'device':'0', ###显卡号,如果用TRT模型，只支持0（单显卡）
|
|||
|
|
'labelnames':"../AIlib2/conf/forestCrowd/labelnames.json", ###检测类别对照表
|
|||
|
|
'max_workers':1, ###并行线程数
|
|||
|
|
#'pedestrianId':行人的ID,'crowdThreshold':判断是否是人群时人的数量,'gatherId':人群的ID,'distancePersonScale':人与人之间的距离/人的身高
|
|||
|
|
'postProcess':{'function':gather_post_process,'pars':{'pedestrianId':2,'crowdThreshold':4,'gatherId':5,'distancePersonScale':2.0}},
|
|||
|
|
'models':
|
|||
|
|
[
|
|||
|
|
{
|
|||
|
|
'weight':"../weights/%s/AIlib2/%s/yolov5_%s_fp16.engine"%(opt['gpu'], opt['business'] ,opt['gpu'] ),###检测模型路径
|
|||
|
|
#'weight':'../weights/pth/AIlib2/forestCrowd/yolov5.pt',###检测模型路径
|
|||
|
|
'name':'yolov5',
|
|||
|
|
'model':yolov5Model,
|
|||
|
|
'par':{ 'half':True,'device':'cuda:0' ,'conf_thres':0.25,'iou_thres':0.45,'allowedList':[0,1,2,3],'segRegionCnt':1, 'trtFlag_det':False,'trtFlag_seg':False, "score_byClass":{ "0":0.25,"1":0.25,"2":0.6,"3":0.6,'4':0.6 ,'5':0.6 } },
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
|
|||
|
|
],
|
|||
|
|
'detModelpara':[{"id":str(x),"config":{"k1":"v1","k2":"v2"}} for x in [] ],
|
|||
|
|
'postFile': '../AIlib2/conf/forest/para.json',###后处理参数文件
|
|||
|
|
'txtFontSize':10,###文本字符的大小
|
|||
|
|
'digitFont': { 'line_thickness':1,'boxLine_thickness':1, 'fontSize':0.5,'waterLineColor':(0,255,255),'waterLineWidth':1},###显示框、线设置
|
|||
|
|
'testImgPath':'../AIdemo2/images/tt/',###测试图像的位置
|
|||
|
|
#'testImgPath':'/mnt/thsw2/DSP2/weights/forest2/images_visdrone',
|
|||
|
|
'testOutPath':'images/results/',###输出测试图像位置
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
|
|||
|
|
##智慧工地-- "工人","塔式起重机","悬臂","起重机","压路机","推土机","挖掘机","卡车","装载机","泵车","混凝土搅拌车","打桩","其他车辆"
|
|||
|
|
if opt['business'] == 'smartSite':
|
|||
|
|
par={
|
|||
|
|
'device':'0', ###显卡号,如果用TRT模型，只支持0（单显卡）
|
|||
|
|
'labelnames':"../AIlib2/conf/%s/labelnames.json"%(opt['business']), ###检测类别对照表
|
|||
|
|
'max_workers':1, ###并行线程数
|
|||
|
|
|
|||
|
|
'postProcess':{'function':default_mix,'pars':{}},
|
|||
|
|
'models':
|
|||
|
|
[
|
|||
|
|
{
|
|||
|
|
'weight':"../weights/%s/AIlib2/%s/yolov5_%s_fp16.engine"%(opt['gpu'], opt['business'] ,opt['gpu'] ),###检测模型路径
|
|||
|
|
#'weight':'../weights/pth/AIlib2/%s/yolov5.pt'%(opt['business'] ),
|
|||
|
|
'name':'yolov5',
|
|||
|
|
'model':yolov5Model,
|
|||
|
|
'par':{ 'half':True,'device':'cuda:0' ,'conf_thres':0.25,'iou_thres':0.45,'allowedList':list(range(20)),'segRegionCnt':1, 'trtFlag_det':False,'trtFlag_seg':False, "score_byClass":{"0":0.25,"1":0.3,"2":0.3,"3":0.3 } },
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
|
|||
|
|
],
|
|||
|
|
|
|||
|
|
'txtFontSize':50,###文本字符的大小
|
|||
|
|
'digitFont': { 'line_thickness':1,'boxLine_thickness':1, 'fontSize':0.5,'waterLineColor':(0,255,255),'waterLineWidth':1},###显示框、线设置
|
|||
|
|
#'testImgPath':'../AIdemo2/images/tt/',###测试图像的位置
|
|||
|
|
'testImgPath':'images/%s'%(opt['business'] ),
|
|||
|
|
#'testImgPath':'/mnt/thsw2/DSP2/weights/forest2/images_cloud_FP',
|
|||
|
|
#'testImgPath':'/mnt/thsw2/DSP2/weights/forest2/images_visdrone',
|
|||
|
|
'testOutPath':'images/results/',###输出测试图像位置
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
##烟火检测-- 烟花
|
|||
|
|
if opt['business'] == 'firework':
|
|||
|
|
par={
|
|||
|
|
'device':'0', ###显卡号,如果用TRT模型，只支持0（单显卡）
|
|||
|
|
'labelnames':"../AIlib2/conf/%s/labelnames.json"%(opt['business']), ###检测类别对照表
|
|||
|
|
'max_workers':1, ###并行线程数
|
|||
|
|
|
|||
|
|
'postProcess':{'function':default_mix,'pars':{}},
|
|||
|
|
'models':
|
|||
|
|
[
|
|||
|
|
{
|
|||
|
|
'weight':"../weights/%s/AIlib2/%s/yolov5_%s_fp16.engine"%(opt['gpu'], opt['business'] ,opt['gpu'] ),###检测模型路径
|
|||
|
|
#'weight':'../weights/pth/AIlib2/%s/yolov5.pt'%(opt['business'] ),
|
|||
|
|
'name':'yolov5',
|
|||
|
|
'model':yolov5Model,
|
|||
|
|
'par':{ 'half':True,'device':'cuda:0' ,'conf_thres':0.25,'iou_thres':0.45,'allowedList':list(range(20)),'segRegionCnt':1, 'trtFlag_det':False,'trtFlag_seg':False, "score_byClass":{"0":0.25,"1":0.3,"2":0.3,"3":0.3 } },
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
|
|||
|
|
],
|
|||
|
|
'detModelpara':[{"id":str(x),"config":{"k1":"v1","k2":"v2"}} for x in range(20) ],
|
|||
|
|
'postFile': '../AIlib2/conf/forest/para.json',###后处理参数文件
|
|||
|
|
'txtFontSize':50,###文本字符的大小
|
|||
|
|
'digitFont': { 'line_thickness':1,'boxLine_thickness':1, 'fontSize':0.5,'waterLineColor':(0,255,255),'waterLineWidth':1},###显示框、线设置
|
|||
|
|
#'testImgPath':'../AIdemo2/images/tt/',###测试图像的位置
|
|||
|
|
'testImgPath':'images/%s'%(opt['business'] ),
|
|||
|
|
#'testImgPath':'/mnt/thsw2/DSP2/weights/forest2/images_cloud_FP',
|
|||
|
|
#'testImgPath':'/mnt/thsw2/DSP2/weights/forest2/images_visdrone',
|
|||
|
|
'testOutPath':'images/results/',###输出测试图像位置
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
###垃圾检测--"建筑垃圾","白色垃圾","其他垃圾"
|
|||
|
|
if opt['business'] == 'rubbish':
|
|||
|
|
par={
|
|||
|
|
'device':'0', ###显卡号,如果用TRT模型，只支持0（单显卡）
|
|||
|
|
'labelnames':"../AIlib2/conf/%s/labelnames.json"%(opt['business']), ###检测类别对照表
|
|||
|
|
'max_workers':1, ###并行线程数
|
|||
|
|
|
|||
|
|
'postProcess':{'function':default_mix,'pars':{}},
|
|||
|
|
'models':
|
|||
|
|
[
|
|||
|
|
{
|
|||
|
|
'weight':"../weights/%s/AIlib2/%s/yolov5_%s_fp16.engine"%(opt['gpu'], opt['business'] ,opt['gpu'] ),###检测模型路径
|
|||
|
|
#'weight':'../weights/pth/AIlib2/%s/yolov5.pt'%(opt['business'] ),
|
|||
|
|
'name':'yolov5',
|
|||
|
|
'model':yolov5Model,
|
|||
|
|
'par':{ 'half':True,'device':'cuda:0' ,'conf_thres':0.25,'iou_thres':0.45,'allowedList':list(range(20)),'segRegionCnt':1, 'trtFlag_det':False,'trtFlag_seg':False, "score_byClass":{"0":0.25,"1":0.3,"2":0.3,"3":0.3 } },
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
|
|||
|
|
],
|
|||
|
|
'detModelpara':[{"id":str(x),"config":{"k1":"v1","k2":"v2"}} for x in range(20) ],
|
|||
|
|
'postFile': '../AIlib2/conf/forest/para.json',###后处理参数文件
|
|||
|
|
'txtFontSize':50,###文本字符的大小
|
|||
|
|
'digitFont': { 'line_thickness':1,'boxLine_thickness':1, 'fontSize':0.5,'waterLineColor':(0,255,255),'waterLineWidth':1},###显示框、线设置
|
|||
|
|
#'testImgPath':'../AIdemo2/images/tt/',###测试图像的位置
|
|||
|
|
#'testImgPath':'images/%s'%(opt['business'] ),
|
|||
|
|
#'testImgPath':'/mnt/thsw2/DSP2/weights/forest2/images_cloud_FP',
|
|||
|
|
'testImgPath':'/mnt/thsw2/DSP2/rubbish/images_TaiZhouFeedback',
|
|||
|
|
'testOutPath':'images/results/',###输出测试图像位置
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
|
|||
|
|
|
|||
|
|
#第一步加载模型
|
|||
|
|
modelList=[ modelPar['model'](weights=modelPar['weight'],par=modelPar['par']) for modelPar in par['models'] ]
|
|||
|
|
print(' load moder over')
|
|||
|
|
|
|||
|
|
#准备画图字体
|
|||
|
|
labelnames = par['labelnames'] ##对应类别表
|
|||
|
|
names=get_labelnames(labelnames)
|
|||
|
|
label_arraylist = get_label_arrays(names,rainbows,outfontsize=par['txtFontSize'],fontpath="../AIlib2/conf/platech.ttf")
|
|||
|
|
|
|||
|
|
#图像测试
|
|||
|
|
imgpaths,videopaths = get_images_videos( par['testImgPath'])
|
|||
|
|
|
|||
|
|
#开始测试
|
|||
|
|
for imgUrl in imgpaths:
|
|||
|
|
img = cv2.imread(imgUrl);bname = os.path.basename(imgUrl)
|
|||
|
|
|
|||
|
|
ret,timeInfos = AI_process_N([img],modelList,par['postProcess'])
|
|||
|
|
print(ret)
|
|||
|
|
timeInfos=bname+':'+timeInfos
|
|||
|
|
print(timeInfos )
|
|||
|
|
if len(ret)>0:
|
|||
|
|
img0 = drawAllBox(ret,img,label_arraylist,rainbows,par['digitFont'])
|
|||
|
|
else: img0= img
|
|||
|
|
cv2.imwrite(os.path.join('images/results/',bname ) ,img0)
|
|||
|
|
|
|||
|
|
|
|||
|
|
def det_demo(business ):
|
|||
|
|
|
|||
|
|
####森林巡检的参数
|
|||
|
|
if opt['business'] == 'forest':
|
|||
|
|
par={
|
|||
|
|
'device':'0', ###显卡号,如果用TRT模型，只支持0（单显卡）
|
|||
|
|
'labelnames':"../AIlib2/conf/forest/labelnames.json", ###检测类别对照表
|
|||
|
|
'gpuname':'3090',###显卡名称
|
|||
|
|
'max_workers':1, ###并行线程数
|
|||
|
|
'trtFlag_det':True,###检测模型是否采用TRT
|
|||
|
|
'trtFlag_seg':False,###分割模型是否采用TRT
|
|||
|
|
'Detweights':"../weights/%s/AIlib2/%s/yolov5_%s_fp16.engine"%(opt['gpu'], opt['business'] ,opt['gpu'] ),###检测模型路径
|
|||
|
|
'detModelpara':[{"id":str(x),"config":{"k1":"v1","k2":"v2"}} for x in [0,1,2,3,4,5,6] ],###控制哪些检测类别显示、输出
|
|||
|
|
'slopeIndex':[],###岸坡类别（或者其它业务里的类别），不与河道（分割的前景区域）计算交并比，即不论是否在河道内都显示。
|
|||
|
|
'seg_nclass':2,###分割模型类别数目，默认2类
|
|||
|
|
'segRegionCnt':0,###分割模型结果需要保留的等值线数目
|
|||
|
|
'segPar':None,###分割模型预处理参数
|
|||
|
|
'Segweights' : None,###分割模型权重位置
|
|||
|
|
'postFile': '../AIlib2/conf/forest/para.json',###后处理参数文件
|
|||
|
|
'txtFontSize':80,###文本字符的大小
|
|||
|
|
'digitFont': { 'line_thickness':2,'boxLine_thickness':1, 'fontSize':1.0,'waterLineColor':(0,255,255),'waterLineWidth':3},###显示框、线设置
|
|||
|
|
'testImgPath':'../AIdemo2/images/forest/',###测试图像的位置
|
|||
|
|
#'testImgPath':'/mnt/thsw2/DSP2/weights/forest2/',
|
|||
|
|
'testOutPath':'images/results/',###输出测试图像位置
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
###车辆巡检参数
|
|||
|
|
if opt['business'] == 'vehicle':
|
|||
|
|
par={
|
|||
|
|
'device':'0', ###显卡号,如果用TRT模型，只支持0（单显卡）
|
|||
|
|
'labelnames':"../AIlib2/conf/vehicle/labelnames.json", ###检测类别对照表
|
|||
|
|
'gpuname':'2080T',###显卡名称
|
|||
|
|
'max_workers':1, ###并行线程数
|
|||
|
|
'trtFlag_det':True,###检测模型是否采用TRT
|
|||
|
|
'trtFlag_seg':False,###分割模型是否采用TRT
|
|||
|
|
'Detweights':"../weights/%s/AIlib2/%s/yolov5_%s_fp16.engine"%(opt['gpu'], opt['business'] ,opt['gpu'] ),###检测模型路径
|
|||
|
|
'detModelpara':[{"id":str(x),"config":{"k1":"v1","k2":"v2"}} for x in [0,1,2,3,4,5,6] ],###控制哪些检测类别显示、输出
|
|||
|
|
'slopeIndex':[],###岸坡类别（或者其它业务里的类别），不与河道（分割的前景区域）计算交并比，即不论是否在河道内都显示。
|
|||
|
|
'seg_nclass':2,###分割模型类别数目，默认2类
|
|||
|
|
'segRegionCnt':0,###分割模型结果需要保留的等值线数目
|
|||
|
|
'segPar':None,###分割模型预处理参数
|
|||
|
|
'Segweights' : None,###分割模型权重位置
|
|||
|
|
'postFile': '../AIlib2/conf/vehicle/para.json',###后处理参数文件
|
|||
|
|
'txtFontSize':40,###文本字符的大小
|
|||
|
|
'digitFont': { 'line_thickness':2,'boxLine_thickness':1, 'fontSize':1.0,'waterLineColor':(0,255,255),'waterLineWidth':3},###显示框、线设置
|
|||
|
|
'testImgPath':'../AIdemo2/images/vehicle/',###测试图像的位置
|
|||
|
|
'testOutPath':'images/results/',###输出测试图像位置
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
###行人检测模型
|
|||
|
|
if opt['business'] == 'pedestrian':
|
|||
|
|
par={
|
|||
|
|
'device':'0', ###显卡号,如果用TRT模型，只支持0（单显卡）
|
|||
|
|
'labelnames':"../AIlib2/conf/pedestrian/labelnames.json", ###检测类别对照表
|
|||
|
|
'gpuname':'2080T',###显卡名称
|
|||
|
|
'max_workers':1, ###并行线程数
|
|||
|
|
'trtFlag_det':True,###检测模型是否采用TRT
|
|||
|
|
'trtFlag_seg':False,###分割模型是否采用TRT
|
|||
|
|
'Detweights':"../weights/%s/AIlib2/%s/yolov5_%s_fp16.engine"%(opt['gpu'], opt['business'] ,opt['gpu'] ),###检测模型路径
|
|||
|
|
'detModelpara':[{"id":str(x),"config":{"k1":"v1","k2":"v2"}} for x in [0,1,2,3,4,5,6] ],###控制哪些检测类别显示、输出
|
|||
|
|
'slopeIndex':[],###岸坡类别（或者其它业务里的类别），不与河道（分割的前景区域）计算交并比，即不论是否在河道内都显示。
|
|||
|
|
'seg_nclass':2,###分割模型类别数目，默认2类
|
|||
|
|
'segRegionCnt':0,###分割模型结果需要保留的等值线数目
|
|||
|
|
'segPar':None,###分割模型预处理参数
|
|||
|
|
'Segweights' : None,###分割模型权重位置
|
|||
|
|
'postFile': '../AIlib2/conf/pedestrian/para.json',###后处理参数文件
|
|||
|
|
'txtFontSize':40,###文本字符的大小
|
|||
|
|
'digitFont': { 'line_thickness':2,'boxLine_thickness':1, 'fontSize':1.0,'waterLineColor':(0,255,255),'waterLineWidth':3},###显示框、线设置
|
|||
|
|
'testImgPath':'../AIdemo2/images/pedestrian/',###测试图像的位置
|
|||
|
|
'testOutPath':'images/results/',###输出测试图像位置
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
###烟雾火焰检测模型
|
|||
|
|
if opt['business'] == 'smogfire':
|
|||
|
|
par={
|
|||
|
|
'device':'0', ###显卡号,如果用TRT模型，只支持0（单显卡）
|
|||
|
|
'labelnames':"../AIlib2/conf/smogfire/labelnames.json", ###检测类别对照表
|
|||
|
|
'gpuname':'2080T',###显卡名称
|
|||
|
|
'max_workers':1, ###并行线程数
|
|||
|
|
'trtFlag_det':False,###检测模型是否采用TRT
|
|||
|
|
'trtFlag_seg':False,###分割模型是否采用TRT
|
|||
|
|
#'Detweights':"../weights/%s/AIlib2/%s/yolov5_%s_fp16.engine"%(opt['gpu'], opt['business'] ,opt['gpu'] ),###检测模型路径
|
|||
|
|
'Detweights':"../weights/pth/AIlib2/%s/yolov5.pt"%(opt['business']),
|
|||
|
|
'detModelpara':[{"id":str(x),"config":{"k1":"v1","k2":"v2"}} for x in [0,1,2,3,4,5,6] ],###控制哪些检测类别显示、输出
|
|||
|
|
'slopeIndex':[],###岸坡类别（或者其它业务里的类别），不与河道（分割的前景区域）计算交并比，即不论是否在河道内都显示。
|
|||
|
|
'seg_nclass':2,###没有分割模型，此处不用
|
|||
|
|
'segRegionCnt':0,###没有分割模型，此处不用
|
|||
|
|
'segPar':None,###分割模型预处理参数
|
|||
|
|
'Segweights' : None,###分割模型权重位置
|
|||
|
|
'postFile': '../AIlib2/conf/smogfire/para.json',###后处理参数文件
|
|||
|
|
'txtFontSize':40,###文本字符的大小
|
|||
|
|
'digitFont': { 'line_thickness':2,'boxLine_thickness':1, 'fontSize':1.0,'waterLineColor':(0,255,255),'waterLineWidth':3},###显示框、线设置
|
|||
|
|
'testImgPath':'../AIdemo2/images/smogfire/',###测试图像的位置
|
|||
|
|
'testOutPath':'images/results/',###输出测试图像位置
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
|
|||
|
|
###钓鱼游泳检测
|
|||
|
|
if opt['business'] == 'AnglerSwimmer':
|
|||
|
|
par={
|
|||
|
|
'device':'0', ###显卡号,如果用TRT模型，只支持0（单显卡）
|
|||
|
|
'labelnames':"../AIlib2/conf/AnglerSwimmer/labelnames.json", ###检测类别对照表
|
|||
|
|
'gpuname':'2080T',###显卡名称
|
|||
|
|
'max_workers':1, ###并行线程数
|
|||
|
|
'trtFlag_det':True,###检测模型是否采用TRT
|
|||
|
|
'trtFlag_seg':False,###分割模型是否采用TRT
|
|||
|
|
'Detweights':"../weights/%s/AIlib2/%s/yolov5_%s_fp16.engine"%(opt['gpu'], opt['business'] ,opt['gpu'] ),###检测模型路径
|
|||
|
|
'detModelpara':[{"id":str(x),"config":{"k1":"v1","k2":"v2"}} for x in [0,1,2,3,4,5,6] ],###控制哪些检测类别显示、输出
|
|||
|
|
'slopeIndex':[],###岸坡类别（或者其它业务里的类别），不与河道（分割的前景区域）计算交并比，即不论是否在河道内都显示。
|
|||
|
|
'seg_nclass':2,###没有分割模型，此处不用
|
|||
|
|
'segRegionCnt':0,###没有分割模型，此处不用
|
|||
|
|
'segPar':None,###分割模型预处理参数
|
|||
|
|
'Segweights' : None,###分割模型权重位置
|
|||
|
|
'postFile': '../AIlib2/conf/AnglerSwimmer/para.json',###后处理参数文件
|
|||
|
|
'txtFontSize':40,###文本字符的大小
|
|||
|
|
'digitFont': { 'line_thickness':2,'boxLine_thickness':1, 'fontSize':1.0,'waterLineColor':(0,255,255),'waterLineWidth':3},###显示框、线设置
|
|||
|
|
'testImgPath':'../AIdemo2/images/AnglerSwimmer/',###测试图像的位置
|
|||
|
|
'testOutPath':'images/results/',###输出测试图像位置
|
|||
|
|
}
|
|||
|
|
###航道应急，做落水人员检测, channelEmergency
|
|||
|
|
if opt['business'] == 'channelEmergency':
|
|||
|
|
par={
|
|||
|
|
'device':'0', ###显卡号,如果用TRT模型，只支持0（单显卡）
|
|||
|
|
'labelnames':"../AIlib2/conf/channelEmergency/labelnames.json", ###检测类别对照表
|
|||
|
|
'gpuname':'2080T',###显卡名称
|
|||
|
|
'max_workers':1, ###并行线程数
|
|||
|
|
'trtFlag_det':True,###检测模型是否采用TRT
|
|||
|
|
'trtFlag_seg':False,###分割模型是否采用TRT
|
|||
|
|
'Detweights':"../weights/%s/AIlib2/%s/yolov5_%s_fp16.engine"%(opt['gpu'], opt['business'] ,opt['gpu'] ),###检测模型路径
|
|||
|
|
#'detModelpara':[{"id":str(x),"config":{"k1":"v1","k2":"v2"}} for x in [0,1,2,3,4,5,6] ],###控制哪些检测类别显示、输出
|
|||
|
|
'detModelpara':[{"id":str(x),"config":{"k1":"v1","k2":"v2"}} for x in [] ],###控制哪些检测类别显示、输出
|
|||
|
|
'slopeIndex':[],###岸坡类别（或者其它业务里的类别），不与河道（分割的前景区域）计算交并比，即不论是否在河道内都显示。
|
|||
|
|
'seg_nclass':2,###没有分割模型，此处不用
|
|||
|
|
'segRegionCnt':0,###没有分割模型，此处不用
|
|||
|
|
'segPar':None,###分割模型预处理参数
|
|||
|
|
'Segweights' : None,###分割模型权重位置
|
|||
|
|
'postFile': '../AIlib2/conf/channelEmergency/para.json',###后处理参数文件
|
|||
|
|
'txtFontSize':40,###文本字符的大小
|
|||
|
|
'digitFont': { 'line_thickness':2,'boxLine_thickness':1, 'fontSize':1.0,'waterLineColor':(0,255,255),'waterLineWidth':3},###显示框、线设置
|
|||
|
|
'testImgPath':'../AIdemo2/images/channelEmergency/',###测试图像的位置
|
|||
|
|
'testOutPath':'images/results/',###输出测试图像位置
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
|
|||
|
|
###乡村路违法种植
|
|||
|
|
if opt['business'] == 'countryRoad':
|
|||
|
|
par={
|
|||
|
|
'device':'0', ###显卡号,如果用TRT模型，只支持0（单显卡）
|
|||
|
|
'labelnames':"../AIlib2/conf/countryRoad/labelnames.json", ###检测类别对照表
|
|||
|
|
'gpuname':'2080T',###显卡名称
|
|||
|
|
'max_workers':1, ###并行线程数
|
|||
|
|
'trtFlag_det':True,###检测模型是否采用TRT
|
|||
|
|
'trtFlag_seg':False,###分割模型是否采用TRT
|
|||
|
|
'Detweights':"../weights/%s/AIlib2/%s/yolov5_%s_fp16.engine"%(opt['gpu'], opt['business'] ,opt['gpu'] ),###检测模型路径
|
|||
|
|
'detModelpara':[{"id":str(x),"config":{"k1":"v1","k2":"v2"}} for x in [0,1,2,3,4,5,6] ],###控制哪些检测类别显示、输出
|
|||
|
|
'slopeIndex':[],###岸坡类别（或者其它业务里的类别），不与河道（分割的前景区域）计算交并比，即不论是否在河道内都显示。
|
|||
|
|
'seg_nclass':2,###没有分割模型，此处不用
|
|||
|
|
'segRegionCnt':0,###没有分割模型，此处不用
|
|||
|
|
'segPar':None,###分割模型预处理参数
|
|||
|
|
'Segweights' : None,###分割模型权重位置
|
|||
|
|
'postFile': '../AIlib2/conf/countryRoad/para.json',###后处理参数文件
|
|||
|
|
'txtFontSize':40,###文本字符的大小
|
|||
|
|
'digitFont': { 'line_thickness':2,'boxLine_thickness':1, 'fontSize':1.0,'waterLineColor':(0,255,255),'waterLineWidth':3},###显示框、线设置
|
|||
|
|
'testImgPath':'../AIdemo2/images/countryRoad/',###测试图像的位置
|
|||
|
|
'testOutPath':'images/results/',###输出测试图像位置
|
|||
|
|
}
|
|||
|
|
###河道上大型船只
|
|||
|
|
if opt['business'] == 'ship':
|
|||
|
|
par={
|
|||
|
|
'device':'0', ###显卡号,如果用TRT模型，只支持0（单显卡）
|
|||
|
|
'labelnames':"../AIlib2/conf/%s/labelnames.json"%(opt['business']), ###检测类别对照表
|
|||
|
|
'gpuname':'2080T',###显卡名称
|
|||
|
|
'max_workers':1, ###并行线程数
|
|||
|
|
'trtFlag_det':True,###检测模型是否采用TRT
|
|||
|
|
'trtFlag_seg':False,###分割模型是否采用TRT
|
|||
|
|
'Detweights':"../weights/%s/AIlib2/%s/yolov5_%s_fp16.engine"%(opt['gpu'], opt['business'] ,opt['gpu'] ),###检测模型路径
|
|||
|
|
'detModelpara':[{"id":str(x),"config":{"k1":"v1","k2":"v2"}} for x in [0,1,2,3,4,5,6] ],###控制哪些检测类别显示、输出
|
|||
|
|
'slopeIndex':[],###岸坡类别（或者其它业务里的类别），不与河道（分割的前景区域）计算交并比，即不论是否在河道内都显示。
|
|||
|
|
'seg_nclass':2,###没有分割模型，此处不用
|
|||
|
|
'segRegionCnt':0,###没有分割模型，此处不用
|
|||
|
|
'segPar':None,###分割模型预处理参数
|
|||
|
|
'Segweights' : None,###分割模型权重位置
|
|||
|
|
'postFile': '../AIlib2/conf/%s/para.json'%(opt['business']),###后处理参数文件
|
|||
|
|
'txtFontSize':40,###文本字符的大小
|
|||
|
|
'digitFont': { 'line_thickness':2,'boxLine_thickness':1, 'fontSize':1.0,'waterLineColor':(0,255,255),'waterLineWidth':3},###显示框、线设置
|
|||
|
|
'testImgPath':'../../../data/XunHe/shipData/',###测试图像的位置
|
|||
|
|
'testOutPath':'images/results/',###输出测试图像位置
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
###城管项目，检测城市垃圾和车辆
|
|||
|
|
if opt['business'] == 'cityMangement':
|
|||
|
|
par={
|
|||
|
|
'device':'0', ###显卡号,如果用TRT模型，只支持0（单显卡）
|
|||
|
|
'labelnames':"../AIlib2/conf/%s/labelnames.json"%(opt['business']), ###检测类别对照表
|
|||
|
|
'gpuname':'2080Ti',###显卡名称
|
|||
|
|
'max_workers':1, ###并行线程数
|
|||
|
|
'trtFlag_det':True,###检测模型是否采用TRT
|
|||
|
|
'trtFlag_seg':False,###分割模型是否采用TRT
|
|||
|
|
'Detweights':"../weights/%s/AIlib2/%s/yolov5_%s_fp16.engine"%(opt['gpu'], opt['business'] ,opt['gpu'] ),###检测模型路径
|
|||
|
|
'detModelpara':[{"id":str(x),"config":{"k1":"v1","k2":"v2"}} for x in [0,1,2,3,4,5,6] ],###控制哪些检测类别显示、输出
|
|||
|
|
'slopeIndex':[],###岸坡类别（或者其它业务里的类别），不与河道（分割的前景区域）计算交并比，即不论是否在河道内都显示。
|
|||
|
|
'seg_nclass':2,###没有分割模型，此处不用
|
|||
|
|
'segRegionCnt':0,###没有分割模型，此处不用
|
|||
|
|
'segPar':None,###分割模型预处理参数
|
|||
|
|
'Segweights' : None,###分割模型权重位置
|
|||
|
|
'postFile': '../AIlib2/conf/%s/para.json'%(opt['business']),###后处理参数文件
|
|||
|
|
'txtFontSize':40,###文本字符的大小
|
|||
|
|
'digitFont': { 'line_thickness':2,'boxLine_thickness':1, 'fontSize':1.0,'waterLineColor':(0,255,255),'waterLineWidth':3},###显示框、线设置
|
|||
|
|
'testImgPath':'images/tmp',###测试图像的位置
|
|||
|
|
'testOutPath':'images/results/',###输出测试图像位置
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
###城管项目，检测道路情况，输入类别为五个："护栏","交通标志","非交通标志","施工","施工“(第4，第5类别合并，名称相同)
|
|||
|
|
###实际模型检测输出的类别为："护栏","交通标志","非交通标志","锥桶","水马"
|
|||
|
|
if opt['business'] == 'cityRoad':
|
|||
|
|
par={
|
|||
|
|
'device':'0', ###显卡号,如果用TRT模型，只支持0（单显卡）
|
|||
|
|
'labelnames':"../AIlib2/conf/%s/labelnames.json"%(opt['business']), ###检测类别对照表
|
|||
|
|
'gpuname':'2080Ti',###显卡名称
|
|||
|
|
'max_workers':1, ###并行线程数
|
|||
|
|
'trtFlag_det':True,###检测模型是否采用TRT
|
|||
|
|
'trtFlag_seg':False,###分割模型是否采用TRT
|
|||
|
|
'Detweights':"../weights/%s/AIlib2/%s/yolov5_%s_fp16.engine"%(opt['gpu'], opt['business'] ,opt['gpu'] ),###检测模型路径
|
|||
|
|
'detModelpara':[{"id":str(x),"config":{"k1":"v1","k2":"v2"}} for x in [0,1,2,3,4,5,6] ],###控制哪些检测类别显示、输出
|
|||
|
|
'slopeIndex':[],###岸坡类别（或者其它业务里的类别），不与河道（分割的前景区域）计算交并比，即不论是否在河道内都显示。
|
|||
|
|
'seg_nclass':2,###没有分割模型，此处不用
|
|||
|
|
'segRegionCnt':0,###没有分割模型，此处不用
|
|||
|
|
'segPar':None,###分割模型预处理参数
|
|||
|
|
'Segweights' : None,###分割模型权重位置
|
|||
|
|
'postFile': '../AIlib2/conf/%s/para.json'%(opt['business']),###后处理参数文件
|
|||
|
|
'txtFontSize':40,###文本字符的大小
|
|||
|
|
'digitFont': { 'line_thickness':2,'boxLine_thickness':1, 'fontSize':1.0,'waterLineColor':(0,255,255),'waterLineWidth':3},###显示框、线设置
|
|||
|
|
'testImgPath':'images/%s'%(opt['business'] ),###测试图像的位置
|
|||
|
|
'testOutPath':'images/results/',###输出测试图像位置
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
|
|||
|
|
if opt['business'] == 'pothole':
|
|||
|
|
par={
|
|||
|
|
'device':'0', ###显卡号,如果用TRT模型，只支持0（单显卡）
|
|||
|
|
'labelnames':"../AIlib2/conf/%s/labelnames.json"%(opt['business']), ###检测类别对照表
|
|||
|
|
'gpuname':'2080Ti',###显卡名称
|
|||
|
|
'max_workers':1, ###并行线程数
|
|||
|
|
'trtFlag_det':True,###检测模型是否采用TRT
|
|||
|
|
'trtFlag_seg':False,###分割模型是否采用TRT
|
|||
|
|
'Detweights':"../weights/%s/AIlib2/%s/yolov5_%s_fp16.engine"%(opt['gpu'], opt['business'] ,opt['gpu'] ),###检测模型路径
|
|||
|
|
'detModelpara':[{"id":str(x),"config":{"k1":"v1","k2":"v2"}} for x in [0,1,2,3,4,5,6] ],###控制哪些检测类别显示、输出
|
|||
|
|
'slopeIndex':[],###岸坡类别（或者其它业务里的类别），不与河道（分割的前景区域）计算交并比，即不论是否在河道内都显示。
|
|||
|
|
'seg_nclass':2,###没有分割模型，此处不用
|
|||
|
|
'segRegionCnt':0,###没有分割模型，此处不用
|
|||
|
|
'segPar':None,###分割模型预处理参数
|
|||
|
|
'Segweights' : None,###分割模型权重位置
|
|||
|
|
'postFile': '../AIlib2/conf/%s/para.json'%(opt['business']),###后处理参数文件
|
|||
|
|
'txtFontSize':40,###文本字符的大小
|
|||
|
|
'digitFont': { 'line_thickness':2,'boxLine_thickness':1, 'fontSize':1.0,'waterLineColor':(0,255,255),'waterLineWidth':3},###显示框、线设置
|
|||
|
|
'testImgPath':'images/%s'%(opt['business'] ),###测试图像的位置
|
|||
|
|
'testOutPath':'images/results/',###输出测试图像位置
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
|
|||
|
|
#segRegionCnt=par['segRegionCnt']
|
|||
|
|
trtFlag_seg = par['trtFlag_seg'];segPar=par['segPar']
|
|||
|
|
##使用森林，道路模型,business 控制['forest','road']
|
|||
|
|
##预先设置的参数
|
|||
|
|
gpuname=par['gpuname']#如果用trt就需要此参数，只能是"3090" "2080Ti"
|
|||
|
|
device_=par['device'] ##选定模型，可选 cpu,'0','1'
|
|||
|
|
|
|||
|
|
|
|||
|
|
device = select_device(device_)
|
|||
|
|
half = device.type != 'cpu' # half precision only supported on CUDA
|
|||
|
|
trtFlag_det=par['trtFlag_det'] ###是否采用TRT模型加速
|
|||
|
|
|
|||
|
|
##以下参数目前不可改
|
|||
|
|
imageW=1536 ####道路模型
|
|||
|
|
digitFont= par['digitFont']
|
|||
|
|
|
|||
|
|
|
|||
|
|
if trtFlag_det:
|
|||
|
|
Detweights=par['Detweights']
|
|||
|
|
logger = trt.Logger(trt.Logger.ERROR)
|
|||
|
|
with open(Detweights, "rb") as f, trt.Runtime(logger) as runtime:
|
|||
|
|
model=runtime.deserialize_cuda_engine(f.read())# 输入trt本地文件，返回ICudaEngine对象
|
|||
|
|
print('####load TRT model :%s'%(Detweights))
|
|||
|
|
else:
|
|||
|
|
Detweights=par['Detweights']
|
|||
|
|
model = attempt_load(Detweights, map_location=device) # load FP32 model
|
|||
|
|
if half: model.half()
|
|||
|
|
|
|||
|
|
labelnames = par['labelnames']
|
|||
|
|
postFile= par['postFile']
|
|||
|
|
print( Detweights,labelnames )
|
|||
|
|
#conf_thres,iou_thres,classes,rainbows=get_postProcess_para(postFile)
|
|||
|
|
detPostPar = get_postProcess_para_dic(postFile)
|
|||
|
|
conf_thres,iou_thres,classes,rainbows = detPostPar["conf_thres"],detPostPar["iou_thres"],detPostPar["classes"],detPostPar["rainbows"]
|
|||
|
|
if 'ovlap_thres_crossCategory' in detPostPar.keys(): ovlap_thres_crossCategory=detPostPar['ovlap_thres_crossCategory']
|
|||
|
|
else:ovlap_thres_crossCategory = None
|
|||
|
|
|
|||
|
|
####模型选择参数用如下：
|
|||
|
|
mode_paras=par['detModelpara']
|
|||
|
|
|
|||
|
|
allowedList,allowedList_string=get_needed_objectsIndex(mode_paras)
|
|||
|
|
slopeIndex = par['slopeIndex']
|
|||
|
|
##只加载检测模型，准备好显示字符
|
|||
|
|
|
|||
|
|
names=get_labelnames(labelnames)
|
|||
|
|
#imageW=4915;###默认是1920，在森林巡检的高清图像中是4920
|
|||
|
|
outfontsize=int(imageW/1920*40);###
|
|||
|
|
label_arraylist = get_label_arrays(names,rainbows,outfontsize=par['txtFontSize'],fontpath="../AIlib2/conf/platech.ttf")
|
|||
|
|
|
|||
|
|
segmodel = None
|
|||
|
|
##图像测试
|
|||
|
|
#url='images/examples/20220624_响水河_12300_1621.jpg'
|
|||
|
|
impth = par['testImgPath']
|
|||
|
|
outpth = par['testOutPath']
|
|||
|
|
|
|||
|
|
|
|||
|
|
imgpaths=[]###获取文件里所有的图像
|
|||
|
|
for postfix in ['.jpg','.JPG','.PNG','.png']:
|
|||
|
|
imgpaths.extend(glob.glob('%s/*%s'%(impth,postfix )) )
|
|||
|
|
videopaths=[]###获取文件里所有的视频
|
|||
|
|
for postfix in ['.MP4','.mp4','.avi']:
|
|||
|
|
videopaths.extend(glob.glob('%s/*%s'%(impth,postfix )) )
|
|||
|
|
|
|||
|
|
|
|||
|
|
imgpaths.sort()
|
|||
|
|
|
|||
|
|
for i in range(len(imgpaths)):
|
|||
|
|
#for i in range(2):
|
|||
|
|
#imgpath = os.path.join(impth, folders[i])
|
|||
|
|
imgpath = imgpaths[i]
|
|||
|
|
bname = os.path.basename(imgpath )
|
|||
|
|
im0s=[cv2.imread(imgpath)]
|
|||
|
|
time00 = time.time()
|
|||
|
|
#使用不同的函数。每一个领域采用一个函数
|
|||
|
|
p_result,timeOut = AI_process_forest(im0s,model,segmodel,names,label_arraylist,rainbows,half,device,conf_thres, iou_thres,allowedList,font=digitFont,trtFlag_det=trtFlag_det,SecNms=ovlap_thres_crossCategory)
|
|||
|
|
|
|||
|
|
time11 = time.time()
|
|||
|
|
image_array = p_result[1]
|
|||
|
|
cv2.imwrite( os.path.join( outpth,bname ) ,image_array )
|
|||
|
|
|
|||
|
|
print('----image:%s, process:%.1f ,save:%.1f, %s'%(bname,(time11-time00) * 1000, (time.time() - time11) * 1000,timeOut ) )
|
|||
|
|
|
|||
|
|
##process video
|
|||
|
|
|
|||
|
|
print('##begin to process videos, total %d videos'%( len(videopaths)))
|
|||
|
|
for i,video in enumerate(videopaths):
|
|||
|
|
print('process video%d :%s '%(i,video))
|
|||
|
|
par0={'model':model,'segmodel':segmodel, 'names':names,'label_arraylist':label_arraylist,'rainbows':rainbows,'outpth':par['testOutPath'],
|
|||
|
|
'half':half,'device':device,'conf_thres':conf_thres, 'iou_thres':iou_thres,'allowedList':allowedList,'digitFont':digitFont,'trtFlag_det': trtFlag_det
|
|||
|
|
}
|
|||
|
|
process_video(video,par0,mode='det')
|
|||
|
|
|
|||
|
|
|
|||
|
|
|
|||
|
|
def OCR_demo2(opt):
|
|||
|
|
|
|||
|
|
from ocr import ocrModel
|
|||
|
|
if opt['business'] == 'ocr_ch':
|
|||
|
|
par={
|
|||
|
|
#weights = '/home/thsw2/WJ/src/OCR/benchmarking-chinese-text-recognition/weights/scene_base.pth'
|
|||
|
|
'weights' : '/mnt/thsw2/DSP2/weights/ocr2/crnn_ch_2080Ti_fp16_192X32.engine',
|
|||
|
|
'modelpar':{
|
|||
|
|
'char_file':'/home/thsw2/WJ/src/OCR/benchmarking-chinese-text-recognition/src/models/CRNN/data/benchmark.txt',
|
|||
|
|
'mode':'ch',
|
|||
|
|
'nc':3,
|
|||
|
|
'imgH':32,
|
|||
|
|
'imgW':192,
|
|||
|
|
'hidden':256,
|
|||
|
|
'mean':[0.5,0.5,0.5],
|
|||
|
|
'std':[0.5,0.5,0.5],
|
|||
|
|
'dynamic':False
|
|||
|
|
},
|
|||
|
|
'inputDir' : '/home/thsw2/WJ/src/OCR/shipNames'
|
|||
|
|
}
|
|||
|
|
if opt['business'] == 'ocr_en':
|
|||
|
|
par={
|
|||
|
|
'weights' : '/home/thsw2/WJ/src/DSP2/weights/pth/AIlib2/ocr2/crnn_448X32.pth',
|
|||
|
|
#'weights' : '/mnt/thsw2/DSP2/weights/ocr2/crnn_en_2080Ti_fp16_448X32.engine',
|
|||
|
|
'modelpar':{
|
|||
|
|
#'cfg':'../AIlib2/conf/OCR_Ch/360CC_config.yaml',
|
|||
|
|
'char_file':'/home/thsw2/WJ/src/DSP2/AIlib2/conf/ocr2/chars2.txt',
|
|||
|
|
'mode':'en',
|
|||
|
|
'nc':1,
|
|||
|
|
'imgH':32,
|
|||
|
|
'imgW':448,
|
|||
|
|
'hidden':256,
|
|||
|
|
'mean':[0.588,0.588,0.588],
|
|||
|
|
'std':[0.193,0.193,0.193 ],
|
|||
|
|
'dynamic':True
|
|||
|
|
},
|
|||
|
|
'inputDir':'/home/thsw2/WJ/src/DSP2/AIdemo2/images/ocr_en'
|
|||
|
|
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
|
|||
|
|
model = ocrModel(weights=par['weights'],par=par['modelpar'] )
|
|||
|
|
imgUrls = glob.glob('%s/*.jpg'%(par['inputDir']))
|
|||
|
|
|
|||
|
|
for imgUrl in imgUrls[0:]:
|
|||
|
|
img = cv2.imread(imgUrl)
|
|||
|
|
res_real,timeInfos = model.eval(img)
|
|||
|
|
#res_real="".join( list(filter(lambda x:(ord(x) >19968 and ord(x)<63865 ) or (ord(x) >47 and ord(x)<58 ),res_real)))
|
|||
|
|
print(res_real,os.path.basename(imgUrl),timeInfos )
|
|||
|
|
|
|||
|
|
|
|||
|
|
|
|||
|
|
|
|||
|
|
def OBB_demo(opt):
|
|||
|
|
###倾斜框（OBB）的ship目标检测
|
|||
|
|
par={
|
|||
|
|
'model_size':(608,608), #width,height
|
|||
|
|
'K':100, #Maximum of objects'
|
|||
|
|
'conf_thresh':0.18,##Confidence threshold, 0.1 for general evaluation
|
|||
|
|
'device':"cuda:0",
|
|||
|
|
|
|||
|
|
'down_ratio':4,'num_classes':15,
|
|||
|
|
#'weights':'../AIlib2/weights/conf/ship2/obb_608X608.engine',
|
|||
|
|
'weights':'../weights/%s/AIlib2/%s/obb_608X608_%s_fp16.engine'%(opt['gpu'],opt['business'],opt['gpu']),
|
|||
|
|
'dataset':'dota',
|
|||
|
|
'test_dir': 'images/ship/',
|
|||
|
|
'result_dir': 'images/results',
|
|||
|
|
'half': False,
|
|||
|
|
'mean':(0.5, 0.5, 0.5),
|
|||
|
|
'std':(1, 1, 1),
|
|||
|
|
'model_size':(608,608),##width,height
|
|||
|
|
'heads': {'hm': None,'wh': 10,'reg': 2,'cls_theta': 1},
|
|||
|
|
'decoder':None,
|
|||
|
|
'test_flag':True,
|
|||
|
|
'postFile': '../AIlib2/conf/%s/para.json'%(opt['business'] ),###后处理参数文件
|
|||
|
|
'drawBox':False,#####是否画框
|
|||
|
|
'digitWordFont': { 'line_thickness':2,'boxLine_thickness':1,'wordSize':40, 'fontSize':1.0,'label_location':'leftTop'},
|
|||
|
|
'labelnames':"../AIlib2/conf/%s/labelnames.json"%(opt['business'] ), ###检测类别对照表
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
####加载模型
|
|||
|
|
model,decoder2=load_model_decoder_OBB(par)
|
|||
|
|
par['decoder']=decoder2
|
|||
|
|
|
|||
|
|
names=get_labelnames(par['labelnames']);par['labelnames']=names
|
|||
|
|
conf_thres,iou_thres,classes,rainbows=get_postProcess_para(par['postFile']);par['rainbows']=rainbows
|
|||
|
|
label_arraylist = get_label_arrays(names,rainbows,outfontsize=par['digitWordFont']['wordSize'],fontpath="../AIlib2/conf/platech.ttf")
|
|||
|
|
par['label_array']=label_arraylist
|
|||
|
|
|
|||
|
|
img_urls=glob.glob('%s/*'%( par['test_dir'] ))
|
|||
|
|
for img_url in img_urls:
|
|||
|
|
#print(img_url)
|
|||
|
|
ori_image=cv2.imread(img_url)
|
|||
|
|
ori_image_list,infos = OBB_infer(model,ori_image,par)
|
|||
|
|
|
|||
|
|
ori_image_list[1] = draw_obb(ori_image_list[2] ,ori_image_list[1],par)
|
|||
|
|
|
|||
|
|
|
|||
|
|
imgName = os.path.basename(img_url)
|
|||
|
|
saveFile = os.path.join(par['result_dir'], imgName)
|
|||
|
|
ret=cv2.imwrite(saveFile, ori_image_list[1] )
|
|||
|
|
|
|||
|
|
if not ret:
|
|||
|
|
print(saveFile, ' not created ')
|
|||
|
|
print( os.path.basename(img_url),':',infos)
|
|||
|
|
|
|||
|
|
|
|||
|
|
def jkm_demo():
|
|||
|
|
from utilsK.jkmUtils import pre_process,post_process,get_return_data
|
|||
|
|
img_type = 'plate' ## code,plate
|
|||
|
|
par={'code':{'weights':'../AIlib2/weights/jkm/health_yolov5s_v3.jit','img_type':'code','nc':10 },
|
|||
|
|
'plate':{'weights':'../AIlib2/weights/jkm/plate_yolov5s_v3.jit','img_type':'plate','nc':1 },
|
|||
|
|
'conf_thres': 0.4,
|
|||
|
|
'iou_thres':0.45,
|
|||
|
|
'device':'cuda:0',
|
|||
|
|
'plate_dilate':(0.5,0.1)
|
|||
|
|
}
|
|||
|
|
###加载模型
|
|||
|
|
device = torch.device(par['device'])
|
|||
|
|
jit_weights = par['code']['weights']
|
|||
|
|
model = torch.jit.load(jit_weights)
|
|||
|
|
|
|||
|
|
jit_weights = par['plate']['weights']
|
|||
|
|
model_plate = torch.jit.load(jit_weights)
|
|||
|
|
|
|||
|
|
imgd='images/plate'
|
|||
|
|
imgpaths = os.listdir(imgd)
|
|||
|
|
for imgp in imgpaths[0:]:
|
|||
|
|
#imgp = 'plate_IMG_20221030_100612.jpg'
|
|||
|
|
imgpath = os.path.join(imgd,imgp)
|
|||
|
|
im0 = cv2.imread(imgpath) #读取数据
|
|||
|
|
img ,padInfos = pre_process(im0,device) ##预处理
|
|||
|
|
if img_type=='code': pred = model(img) ##模型推理
|
|||
|
|
else: pred = model_plate(img)
|
|||
|
|
boxes = post_process(pred,padInfos,device,conf_thres= par['conf_thres'], iou_thres= par['iou_thres'],nc=par[img_type]['nc']) #后处理
|
|||
|
|
dataBack=get_return_data(im0,boxes,modelType=img_type,plate_dilate=par['plate_dilate'])
|
|||
|
|
print(imgp,boxes,dataBack['type'])
|
|||
|
|
for key in dataBack.keys():
|
|||
|
|
if isinstance(dataBack[key],list):
|
|||
|
|
cv2.imwrite( 'images/results/%s_%s.jpg'%( imgp.replace('.jpg','').replace('.png',''),key),dataBack[key][0] ) ###返回值： dataBack
|
|||
|
|
|
|||
|
|
def crowd_demo(opt):
|
|||
|
|
if opt['business']=='crowdCounting':
|
|||
|
|
|
|||
|
|
from crowd import crowdModel as Model
|
|||
|
|
par={
|
|||
|
|
'mean':[0.485, 0.456, 0.406], 'std':[0.229, 0.224, 0.225],'threshold':0.5,
|
|||
|
|
'input_profile_shapes':[(1,3,256,256),(1,3,1024,1024),(1,3,2048,2048)],
|
|||
|
|
'modelPar':{'backbone':'vgg16_bn', 'gpu_id':0,'anchorFlag':False, 'width':None,'height':None ,'line':2, 'row':2},
|
|||
|
|
|
|||
|
|
'weights':"../weights/%s/AIlib2/%s/crowdCounting_%s_dynamic.engine"%(opt['gpu'], opt['business'] ,opt['gpu'] ),###检测模型路径
|
|||
|
|
'testImgPath':'images/%s'%(opt['business'] ),###测试图像的位置
|
|||
|
|
'testOutPath':'images/results/',###输出测试图像位置
|
|||
|
|
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
|
|||
|
|
#weights='weights/best_mae.pth'
|
|||
|
|
cmodel = Model(par['weights'],par)
|
|||
|
|
|
|||
|
|
img_path = par['testImgPath']
|
|||
|
|
File = os.listdir(img_path)
|
|||
|
|
targetList = []
|
|||
|
|
for file in File[0:]:
|
|||
|
|
COORlist = []
|
|||
|
|
imgPath = img_path + os.sep + file
|
|||
|
|
|
|||
|
|
|
|||
|
|
img_raw = cv2.cvtColor(cv2.imread(imgPath),cv2.COLOR_BGR2RGB)
|
|||
|
|
# cmodel.eval---
|
|||
|
|
# 输入读取的RGB数组
|
|||
|
|
# 输出:list,0--原图，1-人头坐标list,2-对接OBB的格式数据，其中4个坐标均相同，2-格式如下：
|
|||
|
|
# [ [ [ (x0,y0),(x1,y1),(x2,y2),(x3,y3) ],score， cls ], [ [ (x0,y0),(x1,y1),(x2,y2),(x3,y3) ],score ，cls ],........ ]
|
|||
|
|
|
|||
|
|
prets, infos = cmodel.eval(img_raw)
|
|||
|
|
print(file,infos,' 人数：',len(prets[1]))
|
|||
|
|
img_to_draw = cv2.cvtColor(np.array(img_raw), cv2.COLOR_RGB2BGR)
|
|||
|
|
# 打印预测图像中人头的个数
|
|||
|
|
for p in prets[1]:
|
|||
|
|
img_to_draw = cv2.circle(img_to_draw, (int(p[0]), int(p[1])), 2, (0, 255, 0), -1)
|
|||
|
|
COORlist.append((int(p[0]), int(p[1])))
|
|||
|
|
# 将各测试图像中的人头坐标存储在targetList中, 格式：[[(x1, y1),(x2, y2),...], [(X1, Y1),(X2, Y2),..], ...]
|
|||
|
|
targetList.append(COORlist)
|
|||
|
|
#time.sleep(2)
|
|||
|
|
# 保存预测图片
|
|||
|
|
cv2.imwrite(os.path.join(par['testOutPath'], file), img_to_draw)
|
|||
|
|
def customization_demo(opt):
|
|||
|
|
from AI import AI_process_C
|
|||
|
|
|
|||
|
|
if opt['business'] == 'channel2':
|
|||
|
|
|
|||
|
|
from ocr import ocrModel
|
|||
|
|
from utilsK.channel2postUtils import channel2_post_process
|
|||
|
|
|
|||
|
|
par={
|
|||
|
|
'device':'0', ###显卡号,如果用TRT模型，只支持0（单显卡）
|
|||
|
|
'labelnames':"../AIlib2/conf/%s/labelnames.json"%(opt['business']), ###检测类别对照表
|
|||
|
|
'max_workers':1, ###并行线程数
|
|||
|
|
'postProcess':{'function':channel2_post_process,'name':'channel2','pars':{
|
|||
|
|
'objs':[2],'wRation':1/6.0,'hRation':1/6.0,'smallId':0, 'bigId':3, #船只
|
|||
|
|
'newId':4, #未挂国旗船只
|
|||
|
|
'uncoverId':5, #未封仓标签
|
|||
|
|
'recScale':1.2,
|
|||
|
|
'target_cls':3.0, #目标种类
|
|||
|
|
'filter_cls':4.0 #被过滤的种类
|
|||
|
|
}},
|
|||
|
|
'models':[
|
|||
|
|
{
|
|||
|
|
#'weight':"../weights/%s/AIlib2/%s/yolov5_%s_fp16.engine"%(opt['gpu'], opt['business'] ,opt['gpu'] ),###检测模型路径
|
|||
|
|
#'weight':'../weights/pth/AIlib2/%s/yolov5.pt'%(opt['business'] ),
|
|||
|
|
#'weight':'/mnt/thsw2/DSP2/channel2/weights/yolov5.pt',
|
|||
|
|
'weight':"/mnt/thsw2/DSP2/channel2/weights/yolov5_20250515.pt",
|
|||
|
|
#'weight':'/mnt/thsw2/DSP2/weights/channel2/yolov5_2080Ti_fp16.engine',
|
|||
|
|
'name':'yolov5',
|
|||
|
|
'model':yolov5Model,
|
|||
|
|
'par':{ 'half':True,'device':'cuda:0' ,'conf_thres':0.25,'iou_thres':0.45,'allowedList':list(range(20)),'segRegionCnt':1, 'trtFlag_det':False,'trtFlag_seg':False, "score_byClass":{ "0":0.2,"1":0.2,"2":0.2,"3":0.2 } }
|
|||
|
|
|
|||
|
|
},
|
|||
|
|
|
|||
|
|
{
|
|||
|
|
#'weight' : '../weights/%s/AIlib2/ocr2/crnn_ch_%s_fp16_192X32.engine'%(opt['gpu'], opt['gpu']),
|
|||
|
|
'weight':'../weights/pth/AIlib2/ocr2/crnn_ch.pth',
|
|||
|
|
'name':'ocr',
|
|||
|
|
'model':ocrModel,
|
|||
|
|
'par':{
|
|||
|
|
'char_file':'../AIlib2/conf/ocr2/benchmark.txt',
|
|||
|
|
'weight' : '../weights/%s/AIlib2/ocr2/crnn_ch_%s_fp16_192X32.engine'%(opt['gpu'], opt['gpu']),
|
|||
|
|
'mode':'ch',
|
|||
|
|
'nc':3,
|
|||
|
|
'imgH':32,
|
|||
|
|
'imgW':192,
|
|||
|
|
'hidden':256,
|
|||
|
|
'mean':[0.5,0.5,0.5],
|
|||
|
|
'std':[0.5,0.5,0.5],
|
|||
|
|
'dynamic':False,
|
|||
|
|
},
|
|||
|
|
}
|
|||
|
|
],
|
|||
|
|
|
|||
|
|
'detModelpara':[{"id":str(x),"config":{"k1":"v1","k2":"v2"}} for x in [0,1,2,3,5,6,7,8,9] ],###控制哪些检测类别显示、输出
|
|||
|
|
'postFile': '../AIlib2/conf/%s/para.json'%(opt['business'] ),###后处理参数文件
|
|||
|
|
'txtFontSize':20,###文本字符的大小
|
|||
|
|
'digitFont': { 'line_thickness':2,'boxLine_thickness':1, 'fontSize':1.0,'waterLineColor':(0,255,255),'segLineShow':True,'waterLineWidth':2},###显示框、线设置
|
|||
|
|
#'testImgPath':'/mnt/thsw2/DSP2/weights/channel_tmp/videos/',
|
|||
|
|
#'testImgPath':'/home/thsw2/WJ/src/OCR/shipNames',
|
|||
|
|
'testImgPath':'images/tt',
|
|||
|
|
'testOutPath':'images/results/',###输出测试图像位置
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
if opt['business'] == 'crackMeasurement':
|
|||
|
|
from utilsK.crackUtils import Crack_measure
|
|||
|
|
print( '%s 只能测试图像，不能测试视频%s'%('#'*20,'#'*20))
|
|||
|
|
par={
|
|||
|
|
'device':'0', ###显卡号,如果用TRT模型，只支持0（单显卡）
|
|||
|
|
'labelnames':"../AIlib2/conf/%s/labelnames.json"%(opt['business']), ###检测类别对照表
|
|||
|
|
'max_workers':1, ###并行线程数
|
|||
|
|
'postProcess':{
|
|||
|
|
'name':'crackMeasurement',
|
|||
|
|
'function':Crack_measure,
|
|||
|
|
'pars':{'dsx':(123-30)*1000/35*0.004387636 ,'objs':[0,1,2]}
|
|||
|
|
},
|
|||
|
|
'models':[
|
|||
|
|
{
|
|||
|
|
'weight':"../weights/%s/AIlib2/%s/yolov5_%s_fp16.engine"%(opt['gpu'], opt['business'] ,opt['gpu'] ),###检测模型路径
|
|||
|
|
#'weight':'../weights/pth/AIlib2/%s/yolov5.pt'%(opt['business'] ),
|
|||
|
|
'name':'yolov5',
|
|||
|
|
'model':yolov5Model,
|
|||
|
|
'par':{ 'half':True,'device':'cuda:0' ,'conf_thres':0.1,'iou_thres':0.45,'allowedList':[0,1,2,3],'segRegionCnt':1, 'trtFlag_det':False,'trtFlag_seg':False, "score_byClass":{"0":0.1,"1":0.1,"2":0.1 } }
|
|||
|
|
|
|||
|
|
},
|
|||
|
|
|
|||
|
|
{
|
|||
|
|
#'weight':"../weights/%s/AIlib2/%s/stdc_360X640_%s_fp16.engine"%(opt['gpu'], opt['business'] ,opt['gpu'] ),###检测模型路径
|
|||
|
|
'weight':'../weights/pth/AIlib2/%s/stdc_360X640.pth'%(opt['business'] ),
|
|||
|
|
'par':{
|
|||
|
|
#'modelSize':(640,360),
|
|||
|
|
'modelSize':(1920,1080),
|
|||
|
|
'dynamic':True,
|
|||
|
|
'mean':(0.485, 0.456, 0.406),'std' :(0.229, 0.224, 0.225),'predResize':True,'numpy':False, 'RGB_convert_first':True,'seg_nclass':2},###分割模型预处理参数
|
|||
|
|
'model':stdcModel,
|
|||
|
|
'name':'stdc'
|
|||
|
|
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
|
|||
|
|
|
|||
|
|
],
|
|||
|
|
|
|||
|
|
'detModelpara':[{"id":str(x),"config":{"k1":"v1","k2":"v2"}} for x in [0,1,2,3,5,6,7,8,9] ],###控制哪些检测类别显示、输出
|
|||
|
|
'postFile': '../AIlib2/conf/%s/para.json'%(opt['business'] ),###后处理参数文件
|
|||
|
|
'txtFontSize':20,###文本字符的大小
|
|||
|
|
'digitFont': { 'line_thickness':2,'boxLine_thickness':1, 'fontSize':1.0,'waterLineColor':(0,255,255),'segLineShow':True,'waterLineWidth':2},###显示框、线设置
|
|||
|
|
'testImgPath':'images/%s/'%(opt['business']),
|
|||
|
|
#'testImgPath':'/mnt/thsw2/DSP2/weights/cityMangement2_1102/images/debug',
|
|||
|
|
'testOutPath':'images/results/',###输出测试图像位置
|
|||
|
|
}
|
|||
|
|
|
|||
|
|
#第一步加载模型
|
|||
|
|
modelList=[ modelPar['model'](weights=modelPar['weight'],par=modelPar['par']) for modelPar in par['models'] ]
|
|||
|
|
print(' load moder over')
|
|||
|
|
|
|||
|
|
#准备画图字体
|
|||
|
|
labelnames = par['labelnames'] ##对应类别表
|
|||
|
|
names=get_labelnames(labelnames)
|
|||
|
|
label_arraylist = get_label_arrays(names,rainbows,outfontsize=par['txtFontSize'],fontpath="../AIlib2/conf/platech.ttf")
|
|||
|
|
|
|||
|
|
#图像测试
|
|||
|
|
imgpaths,videopaths = get_images_videos( par['testImgPath'])
|
|||
|
|
print('imgs:',imgpaths,'\n videos:',videopaths)
|
|||
|
|
|
|||
|
|
#开始测试
|
|||
|
|
for imgUrl in imgpaths[0:]:
|
|||
|
|
img = cv2.imread(imgUrl);bname = os.path.basename(imgUrl)
|
|||
|
|
if opt['business'] == 'crackMeasurement':
|
|||
|
|
ret,timeInfos = AI_process_C([img],modelList,par['postProcess'])
|
|||
|
|
#返回类型ret-list,[[ x0,y0,x1,y1,score,class,裂缝长度,平均宽度,最大宽度,最小宽度],[...],[...]]
|
|||
|
|
for re in ret:
|
|||
|
|
print('Summarized Cracklength = %.1f mm Mean crack width = %.1f mm Max crack width = %.1f mm Min crack width = %.1f mm '%( re[6], re[7],re[8],re[9] ) )
|
|||
|
|
elif opt['business'] == 'channel2':
|
|||
|
|
ret,timeInfos = AI_process_C([img],modelList,par['postProcess'])
|
|||
|
|
|
|||
|
|
timeInfos=bname+':'+timeInfos
|
|||
|
|
print(timeInfos,ret )
|
|||
|
|
img0 = img.copy()
|
|||
|
|
if len(ret)>0:
|
|||
|
|
img0 = drawAllBox(ret,img0,label_arraylist,rainbows,par['digitFont'])
|
|||
|
|
|
|||
|
|
|
|||
|
|
'''
|
|||
|
|
ret_shipName=list(filter(lambda x: int(x[5])==2, ret))
|
|||
|
|
ret_ship=list(filter(lambda x: int(x[5]) in [3,4], ret))
|
|||
|
|
ret_others = list(filter(lambda x: int(x[5]) not in [2,3,4], ret))
|
|||
|
|
img0 = img.copy()
|
|||
|
|
if len(ret_others)>0:
|
|||
|
|
img0 = drawAllBox(ret_others,img0,label_arraylist,rainbows,par['digitFont'])
|
|||
|
|
if len(ret_shipName) >0:
|
|||
|
|
for rett in ret_shipName:
|
|||
|
|
x0,y0,x1,y1=rett[0:4]
|
|||
|
|
print(' shipName width:%d , height:%d'%( (x1-x0),(y1-y0) ))
|
|||
|
|
img0= plot_one_box_PIL( rett[0:4], img0, color=rainbows[2], label=rett[6], line_thickness=par['digitFont']['line_thickness'])
|
|||
|
|
|
|||
|
|
if len(ret_ship) >0:
|
|||
|
|
for rett in ret_ship:
|
|||
|
|
label = '船只 %.2f: 已封仓'%( rett[4] )
|
|||
|
|
img0= plot_one_box_PIL( rett[0:4], img0, color=rainbows[4], label=label, line_thickness=par['digitFont']['line_thickness'])
|
|||
|
|
|
|||
|
|
'''
|
|||
|
|
cv2.imwrite(os.path.join('images/results/',bname ) ,img0)
|
|||
|
|
|
|||
|
|
#测试视频
|
|||
|
|
|
|||
|
|
###如果是视频文件
|
|||
|
|
|
|||
|
|
for video in videopaths:
|
|||
|
|
cap=cv2.VideoCapture(video)
|
|||
|
|
fps,width,height,framecnt = get_video_para(cap)
|
|||
|
|
save_path_AI = os.path.join(par['testOutPath'],os.path.basename(video))
|
|||
|
|
vid_writer_AI = cv2.VideoWriter(save_path_AI, cv2.VideoWriter_fourcc(*'mp4v'), fps, (width,height))
|
|||
|
|
num=0
|
|||
|
|
iframe=0;post_results=[];fpsample=30*10
|
|||
|
|
|
|||
|
|
while cap.isOpened():
|
|||
|
|
ret, img = cap.read() #读取摄像头画面
|
|||
|
|
iframe +=1
|
|||
|
|
if not ret:break
|
|||
|
|
ret,timeInfos = AI_process_C([img],modelList,par['postProcess'])
|
|||
|
|
|
|||
|
|
ret_shipName=list(filter(lambda x: int(x[5])==2, ret))
|
|||
|
|
ret_ship=list(filter(lambda x: int(x[5]) in [3,4], ret))
|
|||
|
|
ret_others = list(filter(lambda x: int(x[5]) not in [2,3,4], ret))
|
|||
|
|
img0 = img.copy()
|
|||
|
|
if len(ret_others)>0:
|
|||
|
|
img0 = drawAllBox(ret_others,img0,label_arraylist,rainbows,par['digitFont'])
|
|||
|
|
if len(ret_shipName) >0:
|
|||
|
|
for rett in ret_shipName:
|
|||
|
|
x0,y0,x1,y1=rett[0:4]
|
|||
|
|
#print(' shipName width:%d , height:%d'%( (x1-x0),(y1-y0) ))
|
|||
|
|
|
|||
|
|
if x1-x0>60 and y1-y0>20:
|
|||
|
|
label = rett[6]
|
|||
|
|
#else: label ='船名'
|
|||
|
|
img0= plot_one_box_PIL( rett[0:4], img0, color=rainbows[2], label=label, line_thickness=par['digitFont']['line_thickness'])
|
|||
|
|
|
|||
|
|
if len(ret_ship) >0:
|
|||
|
|
for rett in ret_ship:
|
|||
|
|
label = '船只 %.2f: 已封仓'%( rett[4] )
|
|||
|
|
img0= plot_one_box_PIL( rett[0:4], img0, color=rainbows[4], label=label, line_thickness=par['digitFont']['line_thickness'])
|
|||
|
|
|
|||
|
|
ret = vid_writer_AI.write(img0)
|
|||
|
|
view_bar(iframe, framecnt,time.time(),prefix=os.path.basename(video))
|
|||
|
|
vid_writer_AI.release();
|
|||
|
|
|
|||
|
|
def AI_process_C_multi( ps ):
|
|||
|
|
return AI_process_C( ps[0] ,ps[1],ps[2] )
|
|||
|
|
|
|||
|
|
###用视频文件做一个多线程测试
|
|||
|
|
'''
|
|||
|
|
max_workers=4
|
|||
|
|
bs=4
|
|||
|
|
for video in videopaths:
|
|||
|
|
cap=cv2.VideoCapture(video)
|
|||
|
|
fps,width,height,framecnt = get_video_para(cap)
|
|||
|
|
print('-'*10,' line1307 fps:{}, width:{},height:{},framecnt:{} '.format( fps,width,height,framecnt) )
|
|||
|
|
iframe =0
|
|||
|
|
parsIns=[]
|
|||
|
|
while cap.isOpened():
|
|||
|
|
ret, img = cap.read() #读取摄像头画面
|
|||
|
|
if not ret:break
|
|||
|
|
parsIns.append( [ [img],modelList,par['postProcess'] ] )
|
|||
|
|
iframe+=1
|
|||
|
|
if iframe%bs==0:
|
|||
|
|
with ThreadPoolExecutor(max_workers=max_workers) as t:
|
|||
|
|
results = t.map(AI_process_C_multi, parsIns)
|
|||
|
|
results = list(results)
|
|||
|
|
print(iframe,len( parsIns ))
|
|||
|
|
parsIns=[]
|
|||
|
|
|
|||
|
|
view_bar(iframe, framecnt,time.time(),prefix=os.path.basename(video))
|
|||
|
|
'''
|
|||
|
|
if __name__=="__main__":
|
|||
|
|
|
|||
|
|
#jkm_demo()
|
|||
|
|
businessAll=['river2','AnglerSwimmer', 'countryRoad','forest2', 'forestCrowd','pedestrian' , 'smogfire' , 'vehicle','ship2',"highWay2","channelEmergency","cityMangement","drowning","noParking","illParking",'cityMangement2',"cityRoad","crowdCounting",'cityMangement3','ocr_en','ocr_ch','pothole','crackMeasurement','channel2','riverT','rubbish','firework','smartSite','highWaySpill']
|
|||
|
|
#businessAll=['crackMeasurement']
|
|||
|
|
businessAll = ['highWaySpill']
|
|||
|
|
|
|||
|
|
|
|||
|
|
|
|||
|
|
# forest 、 ocr2 、ocr_en 、 river 、 road 、 ship ,目前都没有在用
|
|||
|
|
for busi in businessAll:
|
|||
|
|
print('-'*40,'beg to test ',busi,'-'*40)
|
|||
|
|
opt={'gpu':'2080Ti','business':busi}
|
|||
|
|
if opt['business'] in ['highWay2','river2','drowning','noParking','river',"illParking","cityMangement2","riverT"]:
|
|||
|
|
detSeg_demo(opt)
|
|||
|
|
elif opt['business'] in ['cityMangement3','forest2','forestCrowd','rubbish','firework','smartSite','highWaySpill'] :
|
|||
|
|
detSeg_demo2(opt)
|
|||
|
|
elif opt['business'] in ['crowdCounting'] :
|
|||
|
|
crowd_demo(opt)
|
|||
|
|
elif opt['business'] in ['ship2']:
|
|||
|
|
OBB_demo(opt)
|
|||
|
|
elif opt['business'] in ['ocr']:
|
|||
|
|
OCR_demo(opt)
|
|||
|
|
elif opt['business'] in ['crackMeasurement','channel2'] :
|
|||
|
|
customization_demo(opt)
|
|||
|
|
elif opt['business'] in ['ocr_en','ocr_ch']:
|
|||
|
|
OCR_demo2(opt)
|
|||
|
|
|
|||
|
|
|
|||
|
|
else:
|
|||
|
|
det_demo( opt )
|
|||
|
|
|
|||
|
|
|
|||
|
|
|