Improved Fusion based Breast Ultrasound Image Segmentation technique

Abstract

Image segmentation plays vital role in the biomedical imaging .Image segmentation is the process of dividing image into segmentation to extract the information. Breast Cancer is most common form of cancer found in women. The main cause of the disease is still known, only way to cure the disease is the detection of this disease in early stages. Image segmentation is considered as the one of the efficient method of detection of the disease. Traditionally various image segmentation techniques have been proposed for the cancer detection. The existing system has some problems faced that as in the final stage the optimization based approach is used to get segmented image there will be some chances to get degraded results because of some disadvantages that are noticed in case of optimization algorithm . In this approach a new method is proposed for the detection of cancer. In this frequency as well as spatial domain segmentation is applied on shows the proposed method is more accurate and efficient than the traditional method. the image. As by applying the image segmentation is used for dividing the image, to obtain the image again the image fusion is applied. For fusing the image PCA technique of image fusion is used. So by applying this technique the quality of the image is not degraded. From the results obtained it is concluded that this proposed algorithm is better than the traditional used algorithm. A comparison is performed that shows the proposed method is more accurate and efficient than the traditional method. Keywords--image segmentation; biomedical imaging; image fusion; PCA. Introduction Cancer is the most horrendous disease that is the leading cause of illness among the humans worldwide. According to the survey conducted it was found that is one of major cause of the death among the humans and is growing at fast pace. Since the roots of the cancer are still unknown the numbers of deaths due to the cancer are increasing every year. As it is not possible to completely prevent or take caution of the disease. [2] The early detection of the disease can increase the rate of survival. Breast cancer is the common type of cancer that is found in the women. This is result of the abnormal and unconstrained growth of the breast cells that are divided and give rise to tumor. [3] Benign and the Malignant are two types in which the tumor is defined. Benign is not harmful as it grows but does not spread to other part so this is not cancerous. Malignant tumor grows and spread to other body parts so it is cancerous. They can be treated if they are early detected. Due to the advancement in the biomedical imaging now it’s possible to detect the disease so that it can be diagnosed. Various image processing techniques like preprocessing, segmentation, feature extraction etc are used for the detection of the disease. [4] In case of medical image the processing done before the diagnosis needs special attention as the detection of the disease is important step for diagnosis. Segmentation is one of the basis processes of image