Aiming at the problem of large calculation and long time consumption for the validity detection for the production process of video surveillance.The dataset is established based on the collected detection video surveillance images for the hand-held probe,and the detection model for the hand-held probe is trained;The K-Nearest-Neighbors(KNN)algorithm is applied to analyze the continuous frames of the video surveillance and separate the foreground and background for the video frame,the artificial handheld probe model is used to extract the probe from surveillance videos in real time,and then the foreground image of the hand-held probe can be obtained;Based on the foreground image of the hand-held probe,the algorithm based on the pixel search is proposed to calcu-late the detection position of the hand-held probe.Then,compared with the actual position need to be theoretically detected for the probe,the validity for detection of the hand-held probe will be evaluated.The experiment is carried out in the factory surveillance vid-eo,and the results show that the average accuracy for the validity detection of the hand-held probe in the video surveillance is about 93.26％,the recall rate is about 81.11％,and the F1 score is about 86.76％with the detection speed of 9.66 fps/s,which achieves the validity detection of the hand-held probe in the factory surveillance video.
维普
万方数据