A human body re-identification method and human body re-identification system
A technology of re-identification and human body, which is applied in the direction of character and pattern recognition, instruments, computer parts, etc., can solve the problems of reducing the accuracy rate of human body re-identification and image blurring, and achieve the effect of improving clarity and accuracy
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0084] figure 1 It is a flow chart of the human body re-identification method provided by Embodiment 1 of the present invention, see figure 1 shown, including:
[0085] Step 101: Determine the motion area in the video to be detected to form the range of human body detection;
[0086] Step 102: Divide the image forming the human body detection range into multiple color channels, perform grayscale stretching, and form a grayscale stretched image;
[0087] Step 103: Perform human body detection on the image after gray-scale stretching to obtain a human body image;
[0088] Step 104: initially matching the human body in the human body image with the pre-stored human body sample;
[0089] Step 105: judge whether the matching is successful according to the preliminary matching result; if not, then jump to step 106; if yes, then jump to step 107;
[0090] Step 106: saving the detected human body to a human body sample;
[0091] Step 107: Exactly match the human body in the hum...
Embodiment 2
[0099] see figure 2 , in the human body re-identification method provided in Embodiment 2 of the present invention, on the basis of the above-mentioned Embodiment 1, the motion area in the video to be detected is determined, and the range of human body detection formed specifically includes:
[0100] Step 201: Use the frame difference method to detect the moving target of the video, and obtain the foreground picture of each frame of the picture in the video;
[0101] Step 202: Determine the area where the foreground picture is located as the motion area, forming a human body detection range.
[0102] In the second embodiment, in a video to be detected, when a moving person or other object passes by, the first frame in the video is used as the background image, and the other images are compared with the background image, and there is no change The changed part is used as the background, and the changed part is used as the foreground image. The changed part is the moving area ...
Embodiment 3
[0105] see image 3 , in the human body recognition method provided by Embodiment 3 of the present invention, on the basis of Embodiment 2 above,
[0106] Determining motion regions in the video to be detected also includes:
[0107] After obtaining the foreground picture of each frame picture in the video, the obtained foreground picture is filtered using erosion and dilation algorithms.
[0108] In the third embodiment, the motion area in the video to be detected is determined to form the range of human body detection, which specifically includes:
[0109] Step 301: Use the frame difference method to detect the moving target of the video, and obtain the foreground picture of each frame of the picture in the video;
[0110] Step 302: After obtaining the foreground picture of each frame of picture in the video, filter the obtained foreground picture using erosion and expansion algorithm;
[0111] Step 303: Determine that the area where the filtered foreground picture is loc...
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com