Study and Implementation of Morphology for Speckle Noise Removal and Edge Detection

Abstract

The field of computer vision is concerned with extracting features and information from images in order to make analysis of images easier, so that more and more information can be extracted. The primary goal of this thesis is to remove speckle noise present in images used and then to obtain the useful edges in the output image obtained after noise has been removed. The method use for both noise removal and edge detection is mathematical morphology. On the basis of set theory, mathematical morphology is used for image processing, analyzing and comprehending. It is a powerful tool in the geometric morphological analysis and description. Noise removal is very important preprocessing step. The existence of speckle is unwanted since it disgrace image quality and it affects the tasks of individual interpretation and diagnosis. Eliminating such noise is an important preprocessing step. Nonlinear filtering techniques are becoming increasingly important in image processing applications, and are often better than linear filters at removing noise without distorting image features. One structure for designing nonlinear filters is mathematical morphology. By using combination of mathematical morphological operations like opening and closing help in removing speckle noise very effectively. The next step after noise removal is edge detection. The need of edge detection is to find the discontinuities in depth, discontinuities in surface orientation, changes in material properties and variations in scene illumination. Again mathematical morphological operations are used for edge detection and enhancement of edges that are detected. For the evaluation of the performance of proposed algorithm parameters like Signal to Noise Ratio (SNR), Root Mean Square Error (RMSE), Correlation of Coefficient (Coc), and Mean Structural Similarity Index Measure (MSSIM). These parameters are used to calculate the image quality of the output image obtained after speckle noise removed by proposed algorithm and help in comparing the results with that of previously used filters for speckle noise. Another parameter named edge preserving index (EPI) is used for measuring the edges preserved and detected by proposed algorithm with that of previously used edge detectors like Sobel, Prewitt and Canny. Based on the values of these parameters the performance of algorithm in terms of speckle noise removal and edge preservation is discussed.