International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395 -0056 Volume: 03 Issue: 02 | Feb-2016 www.irjet.net p-ISSN: 2395-0072 © 2016, IRJET ISO 9001:2008 Certified Journal Page 618 Review on Digital Image Segmentation Techniques K.Vidhya 1 , S.Revathi 2 , S. Sahaya Selva Ashwini 3 , S.Vanitha 4 1,2,3,4 Department of Electronics and Communication Engineering, Vel Tech, Chennai, Tamilnadu, India ---------------------------------------------------------------------***--------------------------------------------------------------------- Abstract - Image segmentation is partitioning the images based on their similarities, similarities which include color, texture, intensity value, shape; size .This survey gives the better technique for image segmentation which is applied on particular images. This survey is source from journals, conference and online database based on the technique advantage and disadvantages. Key Words: — Image segmentation, Edge detection, Clustering. 1.INTRODUCTION Image segmentation generally means dividing an image into multiple regions. Segmentation provides a meaningful information about an image. It is used for the identification of an object. Basically an image is segmented based on the features like color, texture and intensity. The property of the selected pixel and the information of the neighboring pixel are the two basic parameters of image segmentation. Image segmentation plays a major role in detection of cancerous cells, detection ofland,water, forest region, military applications, computer visions and biometrics. Particular segmentation cannot be applied to all types of image due to its dissimilar features. So many different techniques are followed in image segmentation. Hence it is difficult to develop a universal technique for image segmentation. 2.TECHNIQUES 2.1. Active Contour Method In this method [1] Qiang Chen,used a new edge based interactive image segmentation. Parametric active contour model is adopted for object segmentation. Object contour is obtained by parametric active contour model which is based on feature images. It is generated according to the small number of uses supplied object contour points. By generating the local intensity values of the object contour feature image prevents the evolving curve running into local optimal solution. This method is better than traditional parametric active contour model. 2.1. Region Growing Method The main goal of segmentation is to partition an image into regions. The first step in region growing [8] is selection of seed points. Seed point selection is based on some user criterion (for example, pixels in a certain gray scale range, pixels evenly, spaced on a grid, etc.). The initial region begins at the exact location of these seeds, the regions are then grown from these seed points to adjacent points depending on the region membership criterion pixel intensity, gray scale texture or color. This segmentation provides images with clear edges. 2.2.Markov Random Field Method In SAR image segmentation [6], the MRF method is popular technology which achieves optimal image segmentation. The SAR image is segmented initially which randomly changes current state to new state. The new state is accepted and judged. If all pixels been inquired global energy is calculated else the state is changed. If the global energy is convergent the image is obtained or else it updates the temperature and changes it state. This approach is better than single SAR image segmentation based on MRF. 2.3.Edge Detection Method Edge detection is the most common approach used for detecting meaningful transition, discontinuities in the grey level of an image. Edge is a set of connected pixels that lie on boundary between two regions. Thus the edge detection finds the points where there is an abrupt change in intensities. Edge detection is suitable for image that are simple and noise free. 2.4.Threshold Method Thresholding is very simple to implement, so that it is used in various application of image segmentation. It is used to separate the objects present in an image from its background. Global thresholding is the simplest thresholding technique. It uses a single threshold, to partition an image. Segmentation is implemented by scanning all the pixels in the image one by one. Then each pixel is labeled either object or background. This depends upon threshold. Global