Public place remnant detection method based on monitoring video
A detection method and video monitoring technology, which is applied in the field of image processing, can solve the problem of being unable to detect the loose connection between tourists and their belongings, and achieve the effects of accurate and effective detection, enhanced stability, and improved stability
Active Publication Date: 2020-12-01
CHONGQING UNIVERSITY OF SCIENCE AND TECHNOLOGY
View PDF7 Cites 8 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0006] In view of the fact that the traditional IOU method cannot detect that there is only a loose connection between tourists and their belongings, the present invention uses a comprehensive distance measurement to determine the correlation between tourists and their belongings, which improves the stability of the algorithm
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment
[0038] Such as figure 1 As shown, a method for detecting remnants based on surveillance video includes the following steps:
[0039] S1: Obtain a video stream through an image acquisition device;
[0040] S2: Match tourists and their belongings, which specifically includes the following steps:
[0041] S21: Use the YOLOv4 network to detect the human body of the passenger in the Nth frame and the common and easily lost items carried by him as the target object;
[0042] S22: Use the comprehensive distance measure to match multiple target objects and their corresponding human bodies, the expression is as follows:
[0043]
[0044] Where P is the human body detection frame detected by the YOLOv4 network, and P c is the center point; O is the carrying object detection frame, O c It is the center point; S is the minimum convex set of the detection frame O and P, S d is its diagonal length; ρ 2 (·) is the Euclidean distance. Considering that some belongings are far away fr...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention provides a public place remnants detection method based on a monitoring video. Firstly, acquiring a video stream through image acquisition equipment; then, according to the relationshipbetween the tourist and the carried object in the video frame, matching tourists and carriers thereof, and carrying out target tracking on the tourist and the carried object thereof by using a Kalmanfiltering and apparent modeling cascade matching method, finally judging whether the carried object of the tourist leaves the tourist or not according to the tracking result and the comprehensive distance measurement of the tourist and the carried object thereof, and if the independent residence time of the carried object exceeds a threshold value, judging that the carried object is a remnant andgiving an alarm. Whether the remnants leave the tourists or not is judged through comprehensive distance measurement, and the situation which cannot be handled by a traditional IOU method can be better handled; and meanwhile, the target object is tracked by using a Kalman filtering and apparent modeling cascade matching method, so that the robustness of a multi-target tracking task is enhanced, and the method has relatively strong pertinence and high accuracy in detection of remnants in public places.
Description
technical field [0001] The present invention relates to the technical field of image processing, in particular to a multi-target detection and tracking method. Background technique [0002] It is very common for tourists to lose their belongings in public places. The lost items range from wallets and mobile phones to travel bags. Being able to discover the leftover objects in time and take relevant measures can effectively protect the property safety of tourists and reduce the time cost of tourists returning to search again. [0003] In terms of judging whether an item belongs to a specific tourist, the traditional leftover detection method uses the IOU to judge, but for the situation where the tourist drags the suitcase behind him, there is only a loose connection between the tourist and the object, that is, the distance between the human body and the object detection frame is very likely. It is relatively close, but there is no overlap at all, and the IOU is 0, so it is i...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/00G06K9/62G06T7/246G06N3/04G06N3/08
CPCG06T7/246G06N3/08G06T2207/10016G06T2207/20081G06T2207/20084G06T2207/30232G06V20/52G06N3/045G06F18/241
Inventor 周伟郑福建郭鑫庞一然汪彦易军黄鸿雷友峰辜小花李太福
Owner CHONGQING UNIVERSITY OF SCIENCE AND TECHNOLOGY