Method and device for detecting object with anomaly in motion direction

Abstract

The invention discloses a method and a device for detecting an object with anomaly in motion direction, which can obtain spatial derivatives and time derivatives, obtain characteristic patterns in the motion direction by fusing the two, further carry out normalization by taking the area of a motion region as comparison basis, merge the characteristic patterns in the motion direction in all the directions, and obtain a final saliency map in the motion direction. The device has simple structure and can significantly reduce the calculation quantity; the normalization algorithm relying on area comparison has better detection effect than the current normalization algorithm relying on extreme values, and is more in line with the definition of the anomaly in the motion direction; and the method and the device can lead the detection success rate for natural scenes and non-natural scenes to be higher than that of the prior art, and have the capability of sustainedly tracking the object with the anomaly in the motion direction. Obviously, the method and the device for detecting the object with the anomaly in the motion direction can reduce the calculation quantity, improve the reliability of the algorithm and improve the detection effect.

Description

technical field [0001] The invention relates to detection technology, in particular to a method and device for detecting an abnormal object in a moving direction. Background technique [0002] Motion direction anomaly detection is an important research content in the field of video surveillance for automatic monitoring of abnormal objects in motion. The application in the field of traffic monitoring is gradually gaining attention, to realize the automation of vehicles that violate traffic rules such as retrograde and abnormal steering. Monitoring will greatly improve the enforcement efficiency of traffic supervision departments and effectively prevent the recurrence of similar violations. [0003] The existing abnormal motion direction detection methods can extract the motion direction features of each moving object in the scene, and compare the extracted motion direction features with each other to determine the most significant direction in each motion direction, and final...

AI Technical Summary

Problems solved by technology

This operator is effective for the features of static images, but for the motion direction feature map, since the feature maps corresponding to different directions only respond to the object area moving in this direction, and the response of different areas is not very different; therefore, if the extreme value To measure the significance, it is easy to cause the weight of the feature map that should be given a high weight to be reduced instead, resulting in false detection

At the same time, the pseudo extreme points generated in the feature extraction process will also affect the detection effect.

Images