Anirban Mitra1, Sudipta Roy2, Somais Roy3, Sanjit Kumar Setua3. 1. Department of Computer Science & Engineering, Academy of Technology, Adisaptagram, West Bengal 712121, India; Department of Computer Science & Engineering, Calcutta University Technology Campus, JD-2, Sector-III, Salt Lake, Kolkata 700098, India. 2. Department of Computer Science & Engineering, Ganpat University, Kherva, Mehsana, Gujarat 384012, India; Department of Computer Science & Engineering, Calcutta University Technology Campus, JD-2, Sector-III, Salt Lake, Kolkata 700098, India. Electronic address: sudiptaroy01@yahoo.com. 3. Department of Computer Science & Engineering, Calcutta University Technology Campus, JD-2, Sector-III, Salt Lake, Kolkata 700098, India.
Abstract
BACKGROUND AND OBJECTIVES: Retinal fundus images are extensively used in manually or without human intervention to identify and analyze various diseases. Due to the comprehensive imaging arrangement, there is a large radiance, reflectance and contrast inconsistency within and across images. METHOD: A novel method is proposed based on the cataract physical model to reduce the generated blurriness of the fundus image at the time of image acquisition through the thin layer of cataract by the fundus camera. After the blurriness reduction the method is proposed the enhancement procedure of the images with an objective on contrast perfection with no preamble of artifacts. Due to the uneven distribution of thickness of the cataract, the cataract surroundings are first predicted in the domain of frequency. Second, the resultant image of first step enhanced by the intensity histogram equalization in the adapted Hue Saturation Intensity (HSI) color image space such as the gamut problem can be avoided. The concluding image with suitable color and disparity is acquired by using the proposed max-min color correction approach. RESULTS: The result indicates that not only the proposed method can more effectively enhanced the non-uniform image of retina obtain through thin layer of cataract, but also the resulting image show appropriate brightness and saturation and maintain complete color space information. The projected enhancement method has been tested on the openly available datasets and the result evaluated with the standard used image enhancement algorithms and the cataract removal method. Results show noticeable development over existing methods. CONCLUSIONS: Cataract often prevents the clinician from objectively evaluating fundus feature. Cataract also affect subjective test. Enhancement and restoration of non-uniform illuminated Fundus Image of Retina obtained through thin layer of Cataract has shown here to be potentially beneficial.
BACKGROUND AND OBJECTIVES: Retinal fundus images are extensively used in manually or without human intervention to identify and analyze various diseases. Due to the comprehensive imaging arrangement, there is a large radiance, reflectance and contrast inconsistency within and across images. METHOD: A novel method is proposed based on the cataract physical model to reduce the generated blurriness of the fundus image at the time of image acquisition through the thin layer of cataract by the fundus camera. After the blurriness reduction the method is proposed the enhancement procedure of the images with an objective on contrast perfection with no preamble of artifacts. Due to the uneven distribution of thickness of the cataract, the cataract surroundings are first predicted in the domain of frequency. Second, the resultant image of first step enhanced by the intensity histogram equalization in the adapted Hue Saturation Intensity (HSI) color image space such as the gamut problem can be avoided. The concluding image with suitable color and disparity is acquired by using the proposed max-min color correction approach. RESULTS: The result indicates that not only the proposed method can more effectively enhanced the non-uniform image of retina obtain through thin layer of cataract, but also the resulting image show appropriate brightness and saturation and maintain complete color space information. The projected enhancement method has been tested on the openly available datasets and the result evaluated with the standard used image enhancement algorithms and the cataract removal method. Results show noticeable development over existing methods. CONCLUSIONS:Cataract often prevents the clinician from objectively evaluating fundus feature. Cataract also affect subjective test. Enhancement and restoration of non-uniform illuminated Fundus Image of Retina obtained through thin layer of Cataract has shown here to be potentially beneficial.