Detecting the Optic Disc Boundary in Digital Fundus Images Using Morphological, Edge Detection, and Feature Extraction Techniques

Abstract

Abstract-Optic disc (OD) detection is an important step in developing systems for automated diagnosis of various serious ophthalmic pathologies. This paper presents a new template-based methodology for segmenting the OD from digital retinal images. This methodology uses morphological and edge detection techniques followed by the Circular Hough Transform to obtain a circular OD boundary approximation. It requires a pixel located within the OD as initial information. For this purpose, a location methodology based on a voting-type algorithm is also proposed. The algorithms were evaluated on the 1200 images of the publicly available MESSIDOR database. The location procedure succeeded in 99% of cases, taking an average computational time of 1.67 s. with a standard deviation of 0.14 s. On the other hand, the segmentation algorithm rendered an average common area overlapping between automated segmentations and true OD regions of 86%. The average computational time was 5.69 s with a standard deviation of 0.54 s. Moreover, a discussion on advantages and disadvantages of the models more generally used for OD segmentation is also presented in this paper.

Optic disc segmentation using the sliding band filter

“…where SBF (x i , y i ) is the filter response at point (x i , y i ) using (4), and x m and y m represent the centroid of the K points. After excluding the outliers, the new centroid of the remaining points in set Q is the new candidate for ODC (O l1 ), that will be the center of the new ROI to apply the high-resolution SBF in the next phase.…”

“…Similarly to [8] and [4], we excluded the images with no discernible OD or with severe enough cataracts to prevent meaningful segmentation, leaving a final set of 90 images for assessing the SBF method. The size of the images in this dataset is equal to the reference size.…”

“…For this reason, after getting the result from the low-resolution SBF, the boundary was extracted as mentioned in 2.4.2. After finalizing the OD shape using the smoothing algorithm, the results were compared with those of the Circular Hough Transform (CHT) [4], Morphology-based method (MBM) [16], Temporal Lock, Simple and DVHough methods [8].…”

“…There are several works on the automatic segmentation of OD in retinal images which can mainly be grouped into four categories, namely template-based methods [4,5,6,7], deformable model methods [8,9,10,11,12,13], morphological-based approaches [14,15,16], and pixel classification methods [17,18]. Within the first category, Aquino et al [4] follow a voting-type algorithm to locate a pixel within the OD as initial information to define a starting sub-image.…”

“…Within the first category, Aquino et al [4] follow a voting-type algorithm to locate a pixel within the OD as initial information to define a starting sub-image. Then morphological and edge detection algorithms are applied on the sub-image to segment the OD in the red and green channels separately.…”