A Moving Target Detection Method Based on Statistical Matching of Local Features in Spatio-temporal Domain

Abstract

The invention discloses a moving target detection method combining time-space domain statistical matching and weight distribution. The method is as follows: firstly, a 3‑D LWR operator is proposed and studied to distinguish the importance of neighboring pixels for extracting finer spatiotemporal local features of the video. Secondly, the features of the 3‑D LWR template set are removed from the background and scaled in multiple scales to form a composite template set, which is matched with the local features of the video to be tested to obtain a local similarity matrix. Finally, the space-time statistics are carried out to obtain the position probability matrix of the moving target, and the moving target is extracted by analyzing the position probability matrix. This method combines the traditional LARK operator and weight assignment to construct a new spatiotemporal statistical matching detection model for moving targets. Compared with the existing supervised methods, the present invention can achieve the same detection accuracy without a large amount of training; compared with the existing unsupervised methods, the present invention expands the applicable scenes and shooting angles of the video to be tested, reduces the false detection rate, and Both visible and infrared video are available.

Description

technical field [0001] The invention belongs to the moving target detection technology in the field of artificial intelligence, in particular to a moving target detection method combining time-space domain similarity judgment, statistical overall similarity and video local weighted features. Background technique [0002] In order to more efficiently extract target information from the ever-increasing mass of videos and improve search efficiency, research on moving target detection models has always been a key development technology in the field of artificial intelligence. Existing supervised methods require a lot of training and have high algorithm complexity, while unsupervised methods have low detection accuracy and depend on features. In this paper, according to the requirements of high detection accuracy, fast timeliness, few parameters and easy implementation, a moving target detection method based on statistical matching of local features in time and space domain is ex...

AI Technical Summary

Problems solved by technology

[0003] The three-dimensional LARK feature was proposed by Seo et al. in 2010. It has rotation and scale invariance, captures the potential structure of the image without being affected by noise, and has the advantages of good stability, but it cannot distinguish the importance of central pixels and neighboring pixels. ; while HOG features have no scale and rotation invariance, LBP features cannot retain image details, and SIFT features are easily affected by background and noise; CNN features extract features of different levels through the convolution kernel from shallow to deep, the effect is good but the training samples are lengthy complex

[0004] In terms of detection methods, Seo’s method in the unsupervised category uses a full-background overall template, and the target matches the template as a whole, resulting in limited applicable scenarios for the video to be tested; When the angle difference is large, when the background is not close to the template, the detection accuracy is very low; the supervised method needs to train the target and the background separately, and then verify and adjust after training, the method process is complicated and inefficient

Images