Sample collection method and system used for face recognition and based on video
A face recognition and sample collection technology, applied in the field of face recognition, can solve the problems of slow speed, low efficiency, and long time consumption, and achieve the effect of fast speed and high efficiency
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0081] This embodiment provides a video-based face recognition sample collection method, such as figure 1 shown, including the following steps:
[0082] S1: Obtain the image area with motion information in the image frame of the video, such as figure 2 shown, including:
[0083] First use the initial image frame of the video to establish the background model of each pixel or pixel area, and the initial image frame can select the first frame or the first three frames.
[0084] Build a foreground model for each pixel or pixel region in the current image frame.
[0085] Computes the difference between the background model and the foreground model for each pixel or region of pixels.
[0086] Determine a preset threshold of 20*20 pixels. If the connected area of the pixel region that produces the difference is smaller than the preset threshold, it means that the current frame image does not generate motion information, and there is no human face. Delete the impossible to gene...
Embodiment 2
[0111] This embodiment provides a video-based face recognition sample collection method. The topological diagram of face recognition sample collection information network is as follows: Image 6As shown, according to the functions, there are mainly three parts: the acquisition front-end, the back-end server and the client management end. There are two types of acquisition front-ends: ordinary acquisition front-end 1 and intelligent acquisition front-end 2. Front-end processor 3 outputs front-end video files, ordinary acquisition front-end 1 outputs encoded video images, and the output of intelligent acquisition front-end 2 includes part of the data extracted by intelligent front-end. Face information specifically refers to information in the form of condensed video frames, face capture area images, or face feature data. The back-end server comprises a first server 4 and a second server 5. The first server 4 carries out face feature extraction with the video collected by the f...
Embodiment 3
[0125] This embodiment provides a video-based face recognition sample collection system, such as Figure 8 shown, including:
[0126] An image area acquisition module, configured to acquire an image area with motion information in an image frame of a video;
[0127] A face detection module, configured to detect the image region to obtain a face pixel region;
[0128] The face tracking module is used to track the face pixel area in the image frame containing the face, and obtain the historical image frame of the same face;
[0129] The face feature acquisition module is used to extract the face features from the historical image frames of the same face with high quality, and obtain the required face features for face recognition;
[0130] The face feature storage module is used to store the face feature.
[0131] The image area acquisition module includes:
[0132] Background model establishment sub-module, for utilizing the initial image frame in the video to establish bac...
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