一种利用色彩空间实现导出特征的不确定分割方法
By utilizing tonal differences in the color space, the problem of determining the threshold in image segmentation is solved, resulting in more accurate segmentation results. The output is a continuous or binary image, which facilitates subsequent processing.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- NANJING UNIV OF AERONAUTICS & ASTRONAUTICS
- Filing Date
- 2023-12-14
- Publication Date
- 2026-07-17
AI Technical Summary
In existing technologies, it is difficult to effectively determine a suitable segmentation threshold during image segmentation, leading to uncertain segmentation problems and affecting the accuracy of the segmentation results.
By utilizing the independent property of hue and intensity in color space, the uncertain segmentation problem of images is transformed into a problem of hue difference expression in color space. The feature is derived by calculating pixel by pixel, forming an RGB image and calculating the color difference joint index. Finally, the color channels of the derived features are separated for segmentation.
It effectively solves the problem of the segmentation threshold not being able to be adaptively determined, achieves better segmentation results, and outputs continuous values or binary images, which is convenient for subsequent applications.
Smart Images
Figure CN117876389B_ABST