Exudate-Segmentation

Project on Segmentation of Exudates through Image Processing Techniques

Image Dataset

https://ieee-dataport.org/open-access/indian-diabetic-retinopathy-image-dataset-idrid

Sample Image

Taking the Green Channel of the image and apply CLAHE (Contrast Limited Adaptive Histogram Equalisation)

After Taking CLAHE of the green channel image I am using alternate sequential filtering for Blood Vessel Extraction

The edge candidates

For edge candidates , first we find the edges using Canny Edge Detection and then subtracting the blood vessels from he image using the given function

Now in order to remove the outer edge , we take a mask just as the same shape of the image and then perform a bitwise AND with the image

Now we will know where are the exudates by comparing the images

Credits

Papers taken for reference

1.Guimarães, Juliana & Amorim, Luciana & Ferreira, Flávia & Peixoto, Zélia. (2019). Automatic segmentation of blood vessels in retinal images using 2D Gabor wavelet and sub-image thresholding resulting from image partition. Biomedical Engineering. 39. 10.1007/s42600-019-00028-9.

2. A. Elbalaoui, M. Boutaounte, H. Faouzi, M. Fakir and A. Merbouha, "Segmentation and detection of diabetic retinopathy exudates," 2014 International Conference on Multimedia Computing and Systems (ICMCS), Marrakech, Morocco, 2014, pp. 171-178, doi: 10.1109/ICMCS.2014.6911368.