[opencv]基于dlib和ssd的目标追踪和多线程加速

最新推荐文章于 2025-09-26 01:43:10 发布

原创

最新推荐文章于 2025-09-26 01:43:10 发布 · 1.5k 阅读

5 ·

CC 4.0 BY-SA版权

文章标签：

#opencv #计算机视觉 #python #人工智能

1.opencv的追踪算法

1.1opencv的八个追踪算法

"csrt": cv2.TrackerCSRT_create,
"kcf": cv2.TrackerKCF_create,
"boosting": cv2.TrackerBoosting_create,
"mil": cv2.TrackerMIL_create,
"tld": cv2.TrackerTLD_create,
"medianflow": cv2.TrackerMedianFlow_create,
"mosse": cv2.TrackerMOSSE_create

主要用到 kcf(卡尔曼滤波),效率和准确率都不错

1.2基于kcf的OpenCV追踪检测流程

# 实例化OpenCV's multi-object tracker                         
trackers = cv2.legacy.MultiTracker_create()   #实例化一个追踪器    

# 视频流                         
while True:                   
	# 取当前帧                    
	ret , frame = vs.cv2.VideoCapture(args["video"])   #选择参数列表传入的追踪器  
	# 到头了就结束                  
	if ret is False:          
		print("没有视频")         
		break                             
                              
	# resize每一帧               
	(h, w) = frame.shape[:2]  
	width=600                 
	r = width / float(w)      
	dim = (width, int(h * r)) 
	frame = cv2.resize(frame, 
                              
	# 追踪结果                    
	(success, boxes) = tracker
                              
	# 绘制区域                    
	for box in boxes:         
		(x, y, w, h) = [int(v)
		cv2.rectangle(frame, (
                              
	# 显示                      
	cv2.imshow("Frame", frame)
	key = cv2.waitKey(100) & 0
                              
	if key == ord("s"):       
		# 选择一个区域，按s           
		box = cv2.selectROI("F
			sho