Approximate Nearest-Neighbour Field has been an area of interest in recent research for a wide variety of topics in graphics and multimedia community. Medical imageprocessing is a relatively unaffected field by these...
详细信息
ISBN:
(纸本)9781479915880
Approximate Nearest-Neighbour Field has been an area of interest in recent research for a wide variety of topics in graphics and multimedia community. Medical imageprocessing is a relatively unaffected field by these developments in ANNF computations, brought about by various extremely efficient algorithms like PatchMatch. In this paper, we use Generalized PatchMatch for Optic Disk detection, in retinal images, and show that by making use of efficient ANNF computations we are able to generate results with 98% accuracy with an average time of 0.5 sec. This is significantly faster than conventional Optic Disk detection methods, which average at 95-97% accuracy with 3-5 sec average computation time.
Automatic eye screening for conditions like diabetic retinopathy critically hinges on detection and localization of Optic disk (OD). In this paper, we present a novel scale-embedded dictionary-based method that poses ...
详细信息
ISBN:
(纸本)9781467325332;9781467325349
Automatic eye screening for conditions like diabetic retinopathy critically hinges on detection and localization of Optic disk (OD). In this paper, we present a novel scale-embedded dictionary-based method that poses the problem of OD localization as that of classification, carried out in sparse representation framework. A dictionary is created with manually marked fixed-sized sub-images that contain OD at the center, for multiple scales. For a given test image, all sub-images are sparsely represented as a linear combination of OD dictionary elements. A confidence measure indicating the likelihood of the presence of OD is obtained from these coefficients. Red channel and gray intensity images are processed independently, and their respective confidence measures are fused to form a confidence map. A blob detector is run on the confidence map, whose peak response is considered to be at the location of the OD. The proposed method is evaluated on publicly available databases such as DIARETDB0, DIARETDB1 and DRIVE. The OD was correctly localized in 253 out of 259 images, with an average computation time of 3.8 seconds/image and accuracy of 97.6%. Comparisons with two existing techniques are also discussed.
暂无评论