MORPHOLOGICAL EDGE DETECTION AND CORNER DETECTION ALGORITHM USING CHAIN-ENCODING Neeta Nain, Vijay Laxmi, Ankur Kumar Jain & Rakesh Agarwal Department of Computer Engineering Malaviya National Institute of Technology, Jaipur-302017Rajasthan, India neetanain@yahoo.com , vlgaur@yahoo.co.in , ankur10sep@yahoo.com , raj_mnit222@yahoo.co.in Telephone No: +91-141-2529078(O), +91-141-2529140(R) Fax: +91-141-2529029 Submitted to: IPCV'06 Abstract - Edges and corners are regions of interest where there is a sudden change in intensity. These features play an important role in object identification methods used in machine vision and image processing systems. This paper presents a novel method for edge and corner detection in images. The approach used here is extracting Edges of the input image using morphological operator and then sending it for Chain Encoding. We are proposing a new morphological edge detector which returns a one pixel thick m-connected binary boundary image. This is followed by our chain encoding method to detect corners on the extracted edges. The algorithm works on all types of images (i.e. binary, gray level and color images). Since the proposed methods are based on morphological operations, these are very simple, efficient and fast. Experimental results on a variety of images identified all the prominent edges and significant corners efficiently. Index Terms—Morphological operations, edge detection, corner detection, thresholding, neighborhoods etc. [1] 1. Introduction EDGE detection [1-4] and CORNER detection [7-9, 11] are essential tasks in various computer vision and image-understanding systems. Applications include motion tracking, object recognition, and stereo matching. The requirements of edge detector are that it should identify strong as well as weak edges. All the prominent intensity variations must be taken care of. Similarly corner detection should satisfy a number of important criteria: 1. All "true corners" should be detected. 2. No "false corners" should be detected. 3. Corner points should be well localized. 4. Detector should have a high repeatability rate (good stability). 5. Detector should be robust with respect to noise. 6. Detector should be computationally efficient. This paper proposes a new Morphological Edge Detection Method and Corner Detection Algorithm using Chain-Encoding. The task of finding corners is formulated as a two step process. In the very first step one pixel thick noiseless boundary is extracted by various morphological edge extraction methods. For this, first of all image boundary is extracted by using erosion followed by thresholding, hitmiss transformation and thinning operations. And finally pruning is done to remove extra pixels which may be encountered as side effects. These extracted edges are m-connected. In the second step chain encoding procedures are applied on these extracted edges. Encoding is done using 8-way connectivity. Then abrupt changes are identified in encoding to detect potential corners. Finally optimization of corners to remove false positives by suppressing regular intensity changes is done. The proposed method has been qualitatively evaluated over a number of images with different intensity gradation with excellent results. It identifies significant corners with minimal false positive rate. 2. Morphological Edge Detector in detail In order to detect corner we need very fine image boundary. For that purpose we propose our own Morphological edge detector whose output is edge extracted one pixel thick binary image where each image segment is m-connected. The input of this