Wednesday, September 23, 2020 Home   | Search
Viewing Abstract

Page: 100-111 ------------------------------> Last_modified :7/11/2017 12:10:00 PM

Title: Evaluating the Hausdorff Distance for Contour Segmentation of Brain Images
Authors: Malathi; R. ; Nadirabanu Kamal; A.R
Aff: Department of Computer Science, Box.3030. Ramanathapuram, Tamil Nadu, India. SAAS College, Ramanathapuram, Tamil Nadu, India.
Author Email:
Keywords: Boundary Detection; Canny Operator; Contour Segmentation; Edge Detection; Genetic Algorithm (GA); Genetic Optimizer; Image Segmentation
Abstract:Background: Evaluating the hausdorff distance for contour segmentation of brain images. Objective: Brain tumors are composed of cells that exhibit unstrained growth in brain. Brain tumor detection is one of the challenging task in medical image processing. Since, brain tumors are intricate and tumors can be analyzed only by medical experts. Detecting the accurate boundary in brain images is a crucial task, so this analysis recommends a new edge following technique for correct boundary detection in brain images. The edge detection process serves to facilitate the scrutiny of images by intensely diminishing the amount of data to be processed, while at the same time conserving useful structural information about object boundaries. The canny edge detection algorithm can be used an optimal edge detector based on a set of principle, which comprise finding the most edges by diminishing the error rate. Canny operator is an edge detection technique containing three processes, namely, edge detection, thresholding and edge thinning. Genetic Algorithm (GA) is type of evolutionary systems that simulates the process of natural selection over generations. In this paper, a genetic optimizer is used to predict a suitable threshold value to detect the edges of medical image. The main intent of this paper is to segment the tumor from brain image using the combination of canny operator and active contours. The performance of the proposed method have been tested in medical images, including brain MRIs, brain CT and brain ultrasound images. In this paper, the value of Hausdorff distance is minimized in the range of 0 to 2 and the level of accuracy is increased by 98%. Experimental results shows that the proposed contour segmentation performs very well and give better results, when compared with the existing methods. Results: Edge detection and boundary detection plays an important role in image analysis. Boundaries are mainly used to detect the outline or shape of the object. Image segmentation is used to locate objects and boundaries in images. The proposed edge detection technique for detecting the boundaries of the object using the information from intensity gradient using the vector model and texture gradient using the edge map. The results show that the technique achieves very well and yields better performance than the classical contour models. Conclusion: In this paper, a new edge following technique is designed for boundary detection and applied it to object segmentation problem in medical images. An edge is a property attached to an individual pixel and is calculated from the image function behavior in a neighborhood of the pixel. The purpose of edge detection in general is to significantly diminish the amount of data in an image, while preserving the structural properties to be used for further image processing. This edge following technique incorporates a vector image model and the edge map information. The proposed technique was applied to detect the object boundaries in several types of noisy images where the ill-defined edges were encountered. The proposed integrated image processing algorithm is based on a modified canny edge detection algorithm and implemented using MATLAB.