Liveness detection method, apparatus, medium, device and product
By employing two models in the liveness detection method to handle normal and extreme environments respectively, the problem of performance degradation under extreme environments is solved, and efficient liveness detection is achieved under different environments.
CN115810221BActive Publication Date: 2026-07-17ALIPAY (HANGZHOU) INFORMATION TECH CO LTD
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- ALIPAY (HANGZHOU) INFORMATION TECH CO LTD
- Filing Date
- 2022-12-01
- Publication Date
- 2026-07-17
AI Technical Summary
Technical Problem
Existing liveness detection methods exhibit performance degradation in extreme environments, failing to balance the safety levels of both normal and extreme environments, resulting in unbalanced model detection performance.
Method used
Two liveness detection models are used: one for normal environments and the other for extreme environments. Liveness classification is performed using environmental scene recognition and image enhancement techniques.
Benefits of technology
It improves the performance of liveness detection in extreme environments while maintaining the detection effect in normal environments, thereby improving the efficiency of image processing and liveness detection.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN115810221B_ABST
Abstract
Embodiments of the present specification provide a living body detection method and device, computer readable storage medium, electronic equipment and computer program product. The method comprises: obtaining a to-be-recognized face image, recognizing an environment scene corresponding to the to-be-recognized face image, if the environment scene is a first scene, indicating that the environment scene of the collected to-be-recognized face image is a normal light environment scene, then using a first living body detection model to perform living body classification on the to-be-recognized face image collected in the normal light environment scene, if the environment scene is a second scene, indicating that the environment scene of the collected to-be-recognized face image is a strong light environment scene or a dark light environment scene, then using a second living body detection model to perform living body classification on the to-be-recognized face image collected in a strong light environment scene or a dark light environment scene.
Need to check novelty before this filing date? Find Prior Art