Multi-target detection and tracking method for video surveillance
A video surveillance, multi-target technology, used in instruments, character and pattern recognition, computer parts and other directions, can solve the problems of light conditions and background motion sensitivity, target missed detection rate, high false alarm rate, poor accuracy and other problems
Active Publication Date: 2016-05-25
北京知宏科技有限公司 +1
View PDF3 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0005] The inventor found in the research that there are at least the following technical problems in the prior art: the background modeling method used for target detection in the prior art is sensitive to the lighting conditions of the environment and the movement of the background, and the shadow of the moving object will also affect the comparison results. influences
Therefore, the background modeling method is not suitable for scenes with changing lighting conditions and severe weather, and the accuracy is relatively poor.
In addition, the existing multi-target tracking algorithm has a relatively high target miss rate and false alarm rate.
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 Construction
[0015] In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain the present invention, not to limit the present invention.
[0016] see figure 1 , in one embodiment, provides a kind of video surveillance multi-target detection and tracking method, comprising:
[0017] In step S102, the input target samples are trained by the Adaboost learning algorithm to generate an Adaboost cascade classifier.
[0018] In the field of image target recognition technology, it is often necessary to search for any given image according to a certain strategy to determine whether it contains a specific target, such as a car, boat, human face, etc. As an iterative algorithm, Adaboost's basic idea is: for a specific recognition problem, a gr...
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 discloses a multi-target detection and tracing method for video monitoring. The method comprises the following steps that input target samples are trained according to an Adaboost learning algorithm to generate an Adaboost cascade classifier; target detection is conducted on input image data through the Adaboost cascade classifier to position targets and output the confidence coefficient of the targets; the targets positioned through the Adaboost cascade classifier are used for observation, and target tracking is conducted according to a multi-hypothesis tracing algorithm. According to the technical scheme, the multi-target detection and tracing method for the video monitoring can increase the accuracy rate of the target detection and lower the omission rate and the false report rate at the time of multi-target tracing.
Description
technical field [0001] The invention relates to the technical field of video surveillance, in particular to a video surveillance multi-target detection and tracking method. Background technique [0002] Multi-target detection and tracking technology is the key technology of video surveillance. To detect multiple targets of interest in a surveillance scene, each target needs to be tracked to determine the target's trajectory for subsequent analysis. [0003] In the prior art, the background modeling algorithm is generally used for target detection. The background modeling algorithm is a method for detecting moving targets in video images. The basic idea is to model the background of the image. Once the background model is established , compare the current image with the background model, and determine the foreground target according to the comparison result. Background modeling algorithms include color background models and texture background models. [0004] After the exis...
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 Patents(China)
IPC IPC(8): G06K9/66
Inventor 雷明万克林
Owner 北京知宏科技有限公司