A pedestrian detection and recognition method based on a deep learning cascade neural network

Abstract

The invention relates to a pedestrian detection and recognition method based on a deep learning cascade neural network, and the method comprises the steps: (1) sending a preprocessed video image sequence to a first-level neural network, and obtaining the original information of a pedestrian in an image; (2) segmenting a local image of the pedestrian in the image and carrying out normalization processing to construct a pedestrian recognition data set; and (3) sending the pedestrian recognition data set to a second-level neural network, and extracting feature information of the pedestrians to realize identity recognition of the pedestrians. According to the method, the problems of inaccurate target positioning, low pedestrian resolution, low pedestrian identity recognition accuracy and the like in the image are solved, relatively good image information of the target pedestrian can be obtained, and the pedestrian detection and recognition accuracy is improved. The method is good in practice effect and high in operation speed, detection and identity recognition of the target pedestrian can be rapidly and accurately achieved in real time, and the method is suitable for various fields ofvideo monitoring, intelligent communities, specific place supervision and the like.

Description

technical field [0001] The invention relates to a pedestrian detection and recognition method based on a deep learning cascaded neural network, belonging to the fields of target detection, identity recognition and the like. Background technique [0002] With the development of target detection technology, the application of target detection to video surveillance systems to simplify human operations and provide convenience for humans has become a current research hotspot. Although object detection has made some progress in object category detection, it is not sensitive to differences between objects of the same category. For example, when applied to pedestrian identification, if the original video image sequence is directly fed into the neural network The network is used to identify the identity of the target pedestrian. The preprocessing of the image by the neural network (such as scale scaling) will reduce the resolution of the pedestrian and lose part of the feature inform...

AI Technical Summary

Problems solved by technology

First, the image of the fabric sample is directly obtained by using an electron microscope, and the desired data set of the fabric sample can be obtained according to the experimental requirements, while the pedestrian image needs to collect raw data from cameras with different resolutions, and by locating pedestrians in the image The position of the pedestrian, cropping the image of the area where the pedestrian is located to construct a pedestrian data set, the two are essentially different; second, the AlexNet used in this patent belongs to the network used in the early stage of deep learning, the structure is relatively simple, and the effect is not good when extracting complex features in the image. not good

Images