A Method of Optic Disc Segmentation in Fundus Image Based on Fast Mean Shift
A fundus image and mean shift technology, applied in the field of image processing, can solve the problems of low segmentation efficiency, low robustness and low accuracy, and achieve the effects of improving accuracy, simple calculation and fast speed
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment
[0133] Download the public fundus image dataset HEI_MED on the Internet, and select the picture numbered 37 (original fundus image) for optic disc segmentation, such as figure 2 shown. Input the original fundus image for preprocessing, such as image 3 shown. Using the multi-light source color constancy algorithm, the preprocessed original fundus image is color corrected. Then, based on the color-corrected fundus image, superpixels are divided by fast mean shift, such as Figure 4 shown. The blood vessel map of the color-corrected fundus image is obtained by using Gaussian matched filtering, such as Figure 5 shown. Combining the obtained blood vessel information for optic disc positioning, such as Image 6 shown.
[0134] Select the most complete R component channel image of the optic disc in the optic disc candidate region image, and use the circular structural element of size 7 to perform morphological expansion. The R component channel map of the original fundus i...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


