Pedestrian rerecognition method and system based on segmented silhouette

Abstract

The invention belongs to the field of computer vision, in particular to a pedestrian re-recognition method and system based on segmented silhouette, aiming at solving the problem of how to eliminate background noise of an image and improve the accuracy of pedestrian re-recognition. For the purpose, the pedestrian re-recognition method of the present invention comprises the following steps: obtaining pedestrian characteristics of the pedestrian to be tested based on a pre-constructed pedestrian re-recognition model and according to a color image of the pedestrian to be tested and a corresponding segmented silhouette; obtaining the pedestrian characteristics of the pedestrian to be tested according to the color image of the pedestrian to be tested; calculating the similarity between the pedestrian features and each preset pedestrian identity feature; the pedestrian identity features corresponding to the maximum similarity degree are obtained, and the identity information of the pedestrian to be tested is obtained according to the obtained pedestrian identity features. The pedestrian re-identification method based on the invention can better solve the pedestrian re-identification problem under the condition of chaotic background and improve the identification accuracy. At the same time, the pedestrian re-identification system of the present invention can perform and implement theabove method.

Description

technical field [0001] The invention belongs to the field of computer vision, and in particular relates to a pedestrian re-identification method and system based on segmented silhouettes. Background technique [0002] Pedestrian re-identification is a technology that uses computer vision technology to determine whether a specific pedestrian exists in an image or video sequence. This technology has a wide range of applications in security monitoring, intelligent video analysis, and personnel search and rescue retrieval. [0003] Pedestrian re-identification technology has been studied in academia for many years by processing image data for target detection. It was not until recent years with the development of deep learning that a huge breakthrough was made. The feature description of pedestrians is extracted based on deep convolutional neural network. The ideal state is that the extracted features have a strong distinguishing effect and are relatively stable. However, due to...

AI Technical Summary

Problems solved by technology

[0004] In order to solve the above-mentioned problems in the prior art, that is, to solve the problem of how to eliminate image background noise and improve the accuracy of pedestrian re-identification, one aspect of the present invention provides a pedestrian re-identification method based on segmented silhouettes, including:

Images