blood vessel segmentation using morphological Top-Hat transform

This program implements 2 tophat-based algorithms for vessel segmentation.

The algorithms are available through 'Process' Menu.
They needs at least 2 inputs. one color fundus image and one mask image. If you don't have the mask image, you can create it simply by using 'Create FOV mask' command from the 'Process' Menu.
if you want to compare the result of the algorithm to the manual segmented image (gold standard), you need to input its gold standard image, too.
please, use 'File' menu commands to enter inputs to the program correctly.
The program can also work in Batch Mode. In this mode, you can open lots of images, masks, and their gold standard images.
Example: segmenting DRIVE dataset images
1- Download the DRIVE dataset. (test set)
2- Run the program by 'vessel_segmenter' command
3- File>Open Images (select all the color fundus images)
4- File>Open Mask Images (select all the mask Images)
5- File>Open Reference Images (select all the images in the '1st_manual' folder of the test images of DRIVE dataset)
6- Process> Batch Process (select this option)
7- Process> select one of the algorithms
WTH1
Heneghan et al. 2002
8- You can see each image by clicking on its name in the list boxes moreover 'Display' menu provide some especial options for example overlaping the result on the original image, displaying the statistics related to evaluation measures, namely true positive rate, false positive rate, accuracy & precision.