Multi-color LED illumination-based co-localization analysis multi-channel image automatic registration method and multi-color LED illumination-based co-localization analysis multi-channel image automatic registration system
By acquiring the driving current and structural jitter parameters of the multi-color LED lighting acquisition system, analyzing the offset components caused by spectral and mechanical motion, and performing nonlinear fitting and local search, the problem of image registration accuracy and efficiency in the multi-color LED lighting acquisition system is solved, and high-precision and high-robust automatic image registration is achieved.
Patent Information
- Application Number
- CN202610614467.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-05-07
- Publication Date
- 2026-07-24
- Estimated Expiration
- 2046-05-07
AI Technical Summary
Existing multi-color LED lighting acquisition systems struggle to balance image registration accuracy and computational efficiency under complex conditions, resulting in problems such as blurred edges and false-color stripes in synthesized images.
By acquiring the driving current parameters of each illumination channel and the structural jitter parameters of the acquisition device during the facial image acquisition process, the dispersion displacement and structural offset components caused by spectral differences and mechanical motion are analyzed, nonlinear fitting is performed, and local search is conducted in conjunction with the phase correlation map to locate the sub-pixel offset coordinates and achieve high-precision registration.
It achieves high-precision and robust automatic registration in multi-color LED lighting acquisition scenarios with weak textures and significant spectral differences, improving the quality and reliability of co-localization analysis and multispectral imaging.
