Method for detecting and tracking multi targets at real time in monitoring videotape based on characteristic point classification

Abstract

The invention discloses a method for detecting and tracking multi targets at real time in a monitoring videotape based on characteristic point classification. The method comprises a step of off-time pre-processing: according to the distribution of the characteristic points on a target object, dividing the target object into a plurality of sections, taking the character of each section to train a classifier. The method also comprises a step of taking a character around the character points on line: confirming the object part corresponding to the character point by using the trained classifier and calculating a center point of the corresponding target, detecting the target according to the distribution of the center point and finally tracking the target object based on a tracked character point. The method needs no step of estimating the static background, so the method has better robustness on illumination change and the camera vibration. The method utilizes the rapid stable random tree as the classifier and the gradient of the around character points as the classification data, so the method has excellent detecting and tracking effect and meets the real-time demand.

Description

technical field [0001] The invention relates to a target detection and tracking method, in particular to a real-time multi-target detection and tracking method in a traffic monitoring system. Background technique [0002] The detection and tracking of multi-target moving objects is a very important and challenging problem in the field of computer vision, and has a wide range of applications. In the intelligent traffic monitoring system, it is necessary to identify and track the vehicles and pedestrians in real time. Compared with some other sensors, cameras are not only cheap, but also easy to install, so cameras are installed on most roads, and the videos taken by the cameras can be used to count traffic flow, track vehicles and pedestrians, and so on. [0003] In the past ten years, many researchers have proposed many detection and tracking algorithms for vehicles and pedestrians in surveillance video, and some commercial software in this area have also appeared. Most of...

AI Technical Summary

Problems solved by technology

Compared with the background subtraction method, the corner clustering method can better deal with the occlusion of the target object. However, due to the different sizes of the actual targets, it is difficult for the clustering to achieve a stable effect. For example, the sizes of vehicles and pedestrians are much different. If vehicles and pedestrians appear in one place at the same time, the clustering result is likely to deviate from the actual target object

[0005] There are also some methods that use object-based appearance matching (such as particle filtering), combined with some detection methods, such as Michael D. Breitenstein, Fabian Reichlin, Bastian Leibe, EstherKoller-Meier and Luc Van Gool Robust Tracking-by-Detection using a DetectorConfidence Particle Filter.IEEE International Conference on Computer Vision (ICCV′09), although it can achieve a better tracking effect, but the calculation is relatively complicated, and it is difficult to achieve real-time when there are many objects

Images