Video Object Detection for Tractability with Deep Learning Method

In this paper, a video based objection detection method is proposed for traceability system with deep learning method. The surveillance video is collected first, from which an annotated image database of target object such as people or vehicle was constructed to train convolutional neural network model off-line. With the trained model, a real-time target detection and recognition system is designed and implemented. The proposed method mainly includes three aspects: video processing, target detection and object recognition. It provides a variety of video interfaces to support the downloaded video and real-time video stream. The experimental results indicate that the proposed deep learning based detection method is efficient for the traceability application.