Dynamic low-resolution pedestrian re-identification method based on scale distance gradient function interface learning

Abstract

The invention discloses a dynamic low-resolution pedestrian re-identification method based on scale distance gradient function interface learning, and is to obtain a good re-identification effect when images are low in resolution and change in scale. The method comprises the following steps: to begin with, obtaining an image pair, that is, a positive sample pair, from the same person, and image pairs, that is, negative sample pairs, from different persons; then, introducing distance scale gradient functions, wherein the positive sample pair generates a feasible distance scale gradient function, and the negative sample pair generates an infeasible distance scale gradient function, and the two kinds of distance scale gradient functions form a distance scale gradient function space, a positive parameter vector expressing the feasible distance scale gradient function, and a negative parameter vector expressing the infeasible distance scale gradient function; and finally, carrying out classification through a trained random forest classifier to judge whether the images are from the same object according to the information that whether some distance scale gradient function is in a positive domain or a negative domain in the function space.

Description

technical field [0001] The invention belongs to the technical field of surveillance video retrieval, and relates to a pedestrian re-identification method, in particular to a dynamic low-resolution pedestrian re-identification method based on scale-distance gradient function interface learning. Background technique [0002] In recent years, surveillance networks have been more and more widely used and popularized in public security, mobile detection, passenger flow statistics and other fields. Video surveillance is playing an increasingly important role in combating crime and maintaining social security. Video surveillance has become a public security An effective means for agencies to investigate and solve crimes. However, in the actual video investigation, investigators need to quickly check, track and lock the suspected target according to the moving pictures and trajectories of the designated pedestrian object under the multi-camera, which requires a lot of manpower, mate...

AI Technical Summary

Problems solved by technology

[0005] Chinese Patent Document No. CN104298992A, public (announcement) date 2015.02.01, discloses a data-driven scale-adaptive pedestrian re-identification method, the invention obtains the cross-domain relationship between the query pedestrian and the pedestrian to be measured by using a sparse method Support consistency to calculate the distance between pedestrian pairs, and use the consistency of training data under different viewing angles to adjust the scale adaptively. This method takes into account that the scale of pedestrian images changes, but does not take into account that the resolution of pedestrian images is different. , so the result obtained by the algorithm is not optimal

[0006] Chinese Patent Document No. CN103793702A, Publication (Announcement) Date 2014.05.14, discloses a pedestrian re-identification method based on collaborative scale learning. This invention studies pedestrian re-identification technology based on scale learning under the semi-supervised framework. Auxiliary labeling samples are used for scale learning. This method does not take into account that the scale and resolution changes of pedestrian image pairs will affect the re-identification results. Therefore, the re-identification performance of this method in the case of fewer labeled samples needs to be improved.

In addition, Chinese Patent Document No. CN104200206A, published (announcement) date 2014.12.10, discloses a pedestrian re-identification method based on dual-angle sorting optimization. This method obtains the initial sorting result based on global features and local features, and then further rearrangement, and then get a better re-identification effect; and Chinese patent document number CN103810476A, published (announcement) day 2014.05.21, discloses a pedestrian re-identification method in a video surveillance network based on small group information association, the method According to the relevance of small group information, the small group characteristics of pedestrians are used as an important feature of pedestrian re-identification, which mainly solves the problem of low accuracy and low precision of pedestrian re-identification in video surveillance networks; these methods are also pedestrian re-identification field, but it is also different from our research angle of dynamic low-resolution person re-identification method based on scale-distance gradient function interface learning

Images