International Journal of Computer Applications (0975 – 8887) Volume 16– No.3, February 2011 1 Performance of Ant System over other Convolution Masks in Extracting Edge A. Amali Asha S.P. Victor A. Lourdusamy St.Xavier’s College (Autonomous) Department of Computer Science Department of Mathematics Palayamkottai – 627002, India St.Xavier’s College (Autonomous) St.Xavier’s College (Autonomous) Palayamkottai – 627 002, India. Palayamkottai – 627 002, India. ABSTRACT The front end of most vision systems consists of edge detection as preprocessing. The vision of objects is easy for the human because of the natural intelligence of segmenting, pattern matching and recognizing very complex objects. But for the machine, everything needs to be artificially induced and it is not so easy to recognize and identify objects. Towards Computer vision, the Machine needs pattern recognition; extracting the important features so as to recognize the objects, where the boundary detection or the edge detection is very crucial. Edge detection is finding the points where there are sudden changes in the intensity values and linking them suitably. This paper aims, at presenting a comparative study on the Gradient based edge detectors with a swarm intelligence. Though, these detectors are applied on to the same image, they may not see the same edge pixels. Some detectors seems to be good only for simple transparent images which are less noise prone, and marks pseudo and congested edges in case of denser images. Hence it would be appreciated, to have an edge detector, which is sensitive in detecting edges in majority of the common types of edges. With this in mind, the authors propose a new edge detector based on swarm intelligence, which fairly detects the edges of all types of images with improved quality, and with a low failing probability in detecting edges. Key words: Edge, Segmentation, Feature Extraction, Swarm intelligence, Ant System 1. INTRODUCTION Computer vision aims to duplicate the effect of human vision by electronically perceiving and understanding an image. Giving computers the ability to see is not an easy task. Towards computer vision the role of edge detection is very crucial as it is the preliminary or fundamental stage in pattern recognition. Edges characterize object boundaries and are therefore useful for segmentation and identification of objects in a scene. The idea that the edge detection is the first step in vision processing has fueled a long term search for a good edge detection algorithm. Swarm intelligence methods are computational methods inspired by animals such as social insects acting together to solve complex problems. The main application of these techniques has been to combinatorial optimization problems. This paper discusses work-in-progress on the application of swarm intelligence ideas to image processing problem, such as extracting boundaries or edges of objects. This paper presents an Ant Colony Optimization based mechanism to extract the edges in an image. Experimental results indicate that the proposed method is more efficient than the Gradient based edge detection techniques 2. EXTRACTING EDGES FROM IMAGES An edge [1], [2], [3] is a jump in Intensity or otherwise it can be considered as a typical boundary between two dissimilar regions. An edge is not a physical entity, just like a shadow. It is where the picture ends and the wall starts. It is where the vertical and the horizontal surfaces of an object meet. It‘s what happens between a bright window and a dark. Edges in images are areas with strong intensity contrasts. 2.1 The Edge Structure If we look at the concept of a digital edge a little closer, an edge is a set of connected pixels that lie on the boundary between two regions. An ideal edge is a set of connected pixels, in the vertical direction, each of which is located at an orthogonal step transition in gray level. In practice the imperfections in image acquisition yield edges that are blurred, with the degree of blurring being determined by factors such as the quality of the image acquisition system, the sampling rate, and illumination conditions under which the image is acquired. Effects such as refraction or poor focus can result in objects with boundaries defined by a gradual change in intensity. As a result, if we closely observe the cross section of the edge it is nothing but the shape of the ramp. An ideal edge is a ramp with an infinite slope. The slope of the ramp is inversely proportional to the degree of blurring in the edge. In this model, we no longer have a thin ( one pixel thick) path. Instead, an edge point now is any point contained in the ramp, and an edge would then be ‗a set of such points that are connected. The ―thickness‖ of the edge is determined by the length of the ramp, as it transitions from an initial to a final gray level. This length is determined by the slope, which, in turn, is determined by the degree of blurring. Blurred edges tend to be thick and sharp edges tend to be thin. 2.2 Edge Detection Categories Though, a variety of edge detection Techniques are available, the most of them may be grouped into two categories, Gradient and Laplacian [2]. The gradient method detects edges by looking for a maximum and minimum in the first derivatives of the images [2] ie, it assumes a local maximum at an edge. The laplacian method searches for zero crossing in the second derivatives of the image to find the edges [2]. In gradient method for a continuous image say f(x, y) we consider the two edge directions; horizontal and vertical represented by x(f(x, y)) and y(f(x, y)) . The gradient vector points in the direction of maximum rate of change of ‗f‘ at co-ordinates f(x,y). The