Intensity and Texture Gradient Based Boundary Detection Algorithm for Medical Image

In today’s technological world images plays a very important role. Images are very helpful in several fields like non-destructive testing. Medical diagnostics, surveillance and military etc. in image analysis and computer vision edge detection and boundary detection plays a very fundamental role. For image interpretation tasks and in image analysis processes the boundary of image will provide valuable information. The main objective behind the boundary detection is to detect the shape or outline of object. Generally Image segmentation is aimed to find the objects and object boundaries in images and it assigns a label in every pixel in an image such that pixels with the same level share have certain virtual characteristics. However finding the exact boundary in images is a very difficult task. In this paper we propose an efficient and robust technique for the detection of boundaries in noisy images. Our proposed method uses the information or data from texture and intensity gradients. The texture information will get using edge map model and intensity gradients will get using the edge vector gradient model. The proposed concept for boundary detection and edge following technique is efficient and as well as robust and it is applicable to several kinds of medical images such as magnetic resonance (MR) images, ultrasound images, and computerized tomography images etc. Finally the results for this proposed method shows that the proposed method performs very well and yields better performance than the classical contour models.