Living Body Detection Using Optical Flow Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video-based living body detection methods fail to effectively preserve local feature information, leading to reduced recognition accuracy due to the focus on high-level information and interference from background motion.
Innovation Solution
The method involves dividing video frames into foreground and background images using optical flow information, with the classifier (such as a neural network model) performing category judgment on these images to determine the likelihood of a living body presence, thereby reducing background interference and enhancing feature extraction accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If video-based living body detection is performed using static image judgment technology, then the detection process is simple, but local feature information is lost and recognition accuracy is reduced
Solution Approach 1:
The video frames are segmented into foreground and background regions using optical flow information. The foreground contains the target object while the background is separated out. This segmentation allows the classifier to focus on local feature information from the foreground without being distracted by background elements, thereby improving recognition accuracy while maintaining a relatively simple detection process.
Solution Approach 2:
Background interference is extracted and separated from the video frames using optical flow-based foreground-background segmentation. By taking out the background component, the system preserves local feature information in the foreground that would otherwise be obscured, leading to improved recognition accuracy without significantly increasing process complexity.
2Area of stationary object
If background interference is included in living body detection, then the detection covers the entire scene, but local feature information is obscured and accuracy decreases
Solution Approach 1:
The detection area is segmented into foreground and background regions using optical flow information. This segmentation maintains comprehensive scene coverage by processing both regions, while simultaneously improving feature extraction accuracy by separating the foreground containing local features from the interfering background.
Solution Approach 2:
Different quality processing is applied to different regions: the foreground receives focused attention for local feature extraction, while the background is separated to remove interference. This local quality approach ensures that areas with important local features (foreground) are processed with higher precision requirements, improving overall recognition accuracy while maintaining full scene coverage.
Data Source
AI summary
A living body detection method and system, and a computer-readable storage medium are disclosed. The living body detection method includes: acquiring a video including an object to be detected; extracting at least two images to be detected from the video, and determining optical flow information according to the at least two images to be detected; dividing each image to be detected into a foreground image and a background image according to the optical flow information; using a classifier to perform category judgment on the foreground image and the background image to obtain a category distribution of the foreground image and a category distribution of the background image; and obtaining a probability that the object to be detected is a living body according to the category distribution of the foreground image and the category distribution of the background image.