基于多阈值形状度量的图像暗斑消除方法
By using a multi-threshold shape metric method, dark spots in fundus images can be eliminated quickly and efficiently, solving the problem that existing technologies cannot effectively handle dark spots and achieving high-quality dark spot elimination on fundus cameras.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- NORTHWESTERN POLYTECHNICAL UNIV
- Filing Date
- 2024-09-23
- Publication Date
- 2026-07-17
AI Technical Summary
Existing image processing methods cannot effectively eliminate dark spots in fundus images. In particular, due to the limitations of fundus camera hardware resources, deep learning methods are not well-suited, and traditional methods lack specificity.
A method based on multi-threshold shape measurement is adopted to quickly and with high quality eliminate dark spots in fundus images through ROI region extraction, grayscale processing, multi-threshold segmentation, candidate dark spot detection, voting screening, and baseline region elimination.
It enables efficient and convenient removal of dark spots on common fundus cameras, improving the accuracy of dark spot removal and the utilization of computing resources, while maintaining the color consistency of the area after dark spot removal.
Smart Images
Figure CN119180767B_ABST