Milind M. Mushrif

Color image segmentation: Rough-set theoretic approach

A new color image segmentation algorithm using the concept of histon, based on Rough-set theory, is presented in this paper. The histon is an encrustation of histogram such that the elements in the histon are the set of all the pixels that can be classified as possibly belonging to the same segment. In rough-set theoretic sense, the histogram correlates with the lower approximation and the histon correlates with upper approximation. The roughness measure at every intensity level is calculated and then a thresholding method is applied for image segmentation. The proposed approach is compared with the histogram-based approach and the histon based approach. The experimental results demonstrate that the proposed approach yields better segmentation.

Texture classification using a novel, soft-set theory based classification algorithm

In this paper, we have presented a new algorithm for classification of the natural textures. The proposed classification algorithm is based on the notions of soft set theory. The soft-set theory was proposed by D. Molodtsov which deals with the uncertainties. The choice of convenient parameterization strategies such as real numbers, functions, and mappings makes soft-set theory very convenient and practicable for decision making applications. This has motivated us to use soft set theory for classification of the textures. The proposed algorithm has very low computational complexity when compared with Bayes classification technique and also yields very good classification accuracy. For feature extraction, the textures are decomposed using standard dyadic wavelets. The feature vector is obtained by calculating averaged L1-norm energy of each decomposed channel. The database consists of 25 texture classes selected from Bordatz texture Album. Experimental results show the superiority of the proposed approach compared with some existing methods.

Segmentation of brain MR images using intuitionistic fuzzy clustering algorithm

A new algorithm for the segmentation of brain MR images, using intuitionistic fuzzy clustering (IFCM), is proposed in this paper. The algorithm uses intuitionistic fuzzy representation of image to deal with variations in pixel intensities of brain MR images. The proposed intuitionistic fuzzy clustering algorithms segments brain MR image into three regions, gray matter (GM), white Matter (WM) cerebrospinal fluid (CSF). To evaluate the performance of the proposed method, segmentations results are compared on the basis of segmentation accuracy and computational time with bias corrected fuzzy clustering method (BCFCM). The quantitative evaluation demonstrates the superiority of the proposed algorithm.

Multi-level Thresholding Algorithm for Color Image Segmentation

In this paper, we present an effective multilevel threshold selection method of color image segmentation based on minimum class variance thresholding (MCVT) and OTSU method. The input RGB color image is segmented into regions and proceeds to the region merge using JND (Just Noticeable Difference) of human visual property. The result shows that with simple yet very effective JND merge criteria, the proposed algorithm is capable of generating region representations. After analyzing these two results, Otsu method is better than MCVT method.

Left Ventricular Segmentation of 2-D Echocardiographic Image: A Survey

Cardiovascular diseases are a major health concern worldwide. The left ventricle and in particular the endocardium is a structure of particular interest since it performs the task of pumping oxygenated blood to the entire body. Therefore, segmentation of the left ventricle in echocardiography images is a task with important diagnostic power. More concretely, contour extraction is an important criterion for subjective evaluation of the cardiac function and has become an area of focus. Cardiac function is evaluated quantitatively using echocardiography via the analysis of shape attributes, such as the heart wall thickness or the shape change of the heart wall boundaries. This requires that the complete boundaries of the heart wall be detected from a sequence of two-dimensional ultrasonic images of the heart. The image segmentation process is made difficult since these images are plagued by poor intensity contrast and dropouts caused by the intrinsic limitations of the image formation process. Current studies often require the tedious and time-consuming practice of having trained operators manually trace the heart walls. This paper reviews Left ventricular segmentation methods of 2-D echocardiographic image, in a broad sense.

FCM Clustering Algorithms for Segmentation of Brain MR Images

The study of brain disorders requires accurate tissue segmentation of magnetic resonance MR brain images which is very important for detecting tumors, edema, and necrotic tissues. Segmentation of brain images, especially into three main tissue types: Cerebrospinal Fluid CSF, Gray Matter GM, and White Matter WM, has important role in computer aided neurosurgery and diagnosis. Brain images mostly contain noise, intensity inhomogeneity, and weak boundaries. Therefore, accurate segmentation of brain images is still a challenging area of research. This paper presents a review of fuzzy c-means FCM clustering algorithms for the segmentation of brain MR images. The review covers the detailed analysis of FCM based algorithms with intensity inhomogeneity correction and noise robustness. Different methods for the modification of standard fuzzy objective function with updating of membership and cluster centroid are also discussed.

Texture Classification Using Cosine-modulated Wavelets

This paper proposes a technique for image texture classification based on cosine-modulated wavelet transform. Better discriminability and low implementation cost of the cosine-modulated wavelets has been effectively utilized to yield better features and more accurate classification results. Experimental results demonstrate the effectiveness of this approach on different datasets in three experiments. The proposed approach improves classification rates compared to the traditional Gabor wavelet based approach, rotated wavelet filters based approach, DT-CWT approach and the DLBP approach. The computational cost of the proposed method is less as compared to the other two methods.

Extraction of blood vessels and optic disc in retinal images

Abstract Blood vessels and optic disc (OD) detection is important in developing systems for automated diagnosis of various ophthalmic pathologies. This work proposes two efficient approaches for automatic detection of blood vessels and OD in retinal images. For extraction of blood vessels, first the Gabor filters are applied so as to enhance the contrast between the blood vessels and the background, and then, local entropy thresholding is applied for segmentation. The DRIVE database is used for comparative analysis. The average accuracy, sensitivity and specificity of proposed method are 93.85, 70.99 and 97.35%, respectively. The proposed technique produced better segmentation when compared with the existing algorithms. For localisation and detection of OD, a new histon-based algorithm is proposed in which roughness index of the histon and histogram is used for thresholding. The localised OD area is used as a mask, since it reduces execution time for segmentation of the OD. The proposed techniques have been tested over publicly available data-sets. The OD localisation accuracy for DRIVE and DIARETDB1 database is 95 and 91%, respectively.

Image reconstruction by modified exemplar based inpainting

Exemplar based image inpainting technique is one of the inpainting techniques which uses self similarity priors. The main aim is to fill damaged area of image by copying patches from remaining image. Patch priority, governed by confidence term and data term, decides the order in which filling should occur. This paper proposes the calculation of data term by using edge map technique rather than isophotes directions. The proposed paper gives the comparison of different distance metrics such as sum of squared distance, Hamming distance and normalized cross correlation used to find best matching patch.

Intuitionistic fuzzy roughness measure for segmentation of brain MR images

A multilevel thresholding method for the segmentation of Magnetic Resonance (MR) brain images using the concept of intuitionistic fuzzy and rough set is presented here. Intuitionistic fuzzy roughness measure, calculated by considering histogram as lower approximation of rough set and intuitionistic fuzzy histon as upper approximation of rough set, is used to find optimum valley points for segmentation of brain MR images. A new fuzzy complement function is proposed for intuitionistic fuzzy image representation to take into account intensity inhomogeneity and noise in brain MR images. The proposed algorithm segments brain MR image into three regions, gray matter (GM), white matter (WM) and cerebrospinal fluid (CSF). The quantitative evaluation demonstrate the superiority of the proposed algorithm.