Moving Object Detection and Recognition Algorithm Based on Deep Learning

Aiming at the problem that the deep learning detection framework based on video moving object detection cannot make full use of the motion information of the upper and lower frames, and the computational complexity is high, an improved SSD detection algorithm is proposed. The algorithm uses the lightweight network MobileNet as the basic network in the SSD model to reduce the time and computation spent in extracting image features, and establish the correlation between frames. Based on the Tensorflow deep learning framework, training is carried out on COCO (Common Objects in Context) dataset, and comparative experiments on moving object recognition accuracy and recognition speed are carried out. Experimental results show that the algorithm has good performance in terms of detection accuracy and speed.