Non-contact fingerprint acquisition method, device, apparatus and storage medium
By training a model to predict optical control parameters and dynamically adjusting camera settings, the problem of capturing clear fingerprints in contactless fingerprint authentication was solved, achieving high-quality fingerprint image acquisition.
CN122347818APending Publication Date: 2026-07-07ANT BLOCKCHAIN TECHNOLOGY (SHANGHAI) CO LTD
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- ANT BLOCKCHAIN TECHNOLOGY (SHANGHAI) CO LTD
- Filing Date
- 2026-02-28
- Publication Date
- 2026-07-07
AI Technical Summary
Technical Problem
Traditional cameras struggle to capture clear fingerprint images for contactless fingerprint authentication, as they are significantly affected by factors such as light, distance, and angle.
Method used
The trained parameter prediction model is used to predict optical control parameters. By cyclically adjusting parameters such as camera focus, white balance and exposure, the shooting conditions are dynamically adjusted to improve fingerprint image quality.
Benefits of technology
This improves the fingerprint acquisition quality of the non-contact fingerprint acquisition solution, ensuring that the fingerprint image meets the preset shooting quality requirements.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN122347818A_ABST
Abstract
Embodiments of the present specification provide a non-contact fingerprint acquisition method, device, equipment and storage medium. In the method, a finger image currently collected by a camera and a position of a target finger fingerprint in the finger image are input to a trained parameter prediction model to predict optical control parameters required for improving the shooting quality of the target finger fingerprint; the camera is adjusted according to the optical control parameters to collect a new finger image; whether the target finger fingerprint in the new finger image meets a preset shooting quality requirement is judged; when the target finger fingerprint in the new finger image does not meet the shooting quality requirement, the step of inputting the finger image currently collected by the camera to the trained parameter prediction model to predict the optical control parameters required for improving the shooting quality of the target finger fingerprint is returned to be executed until the target finger fingerprint in the new finger image meets the shooting quality requirement.
Need to check novelty before this filing date? Find Prior Art