Informatics Publishing Limited

Marykutty Cyriac ¹, C. Chellamuthu ²

Affiliations
1 Department of Information Technology, Jerusalem College of Engineering, Pallikaranai, Chennai, IN
2 Department of Electrical and Electronics Engineering, RMK Engineering College, Kavaraipettai, Chennai, IN

Abstract

Recent research in transform based compression techniques has concentrated on wavelet transform due to its superior performance. Compression in transform domain is possible either in the quantization stage or bit allocation stage. In this paper, an approach to compress the image by integrating the human visual system model into the quantization stage of the wavelet transform is presented. Evaluations are performed on medical images and the performance is analysed quantitatively using mean square error and peak signal to noise ratio.

Keywords

Contrast Sensitivity Function, Discrete Wavelet Transform, Human Visual System, Quantization.

Full Text

     

Marykutty Cyriac ¹, B. Karthik ²

Affiliations
1 Department of Electronics and Communication Engineering, Jerusalem College of Engineering, Chennai - 600100, Tamil Nadu, IN
2 Department of Electronics and Communication Engineering, Bharath University, Chennai - 600073, Tamil Nadu, IN

Abstract

One of the visual indications for the presence of diabetic retinopathy is the exudates in the retina area of Fundus images. Detection of exudates is useful information for the early diagnosis of the eye disease. In this paper, a new approach for the detection of exudates is presented. A position based approach is combined with morphological operations for extracting the optical disc and detect the exudates. The proposed method is applied to the standard retinal image dataset and the results are presented.

Keywords

Diabetic Retinopathy, Exudates, Fundus Images Median Filter, Morphological Operations

Full Text