ENHANCED EDGE DETECTION FOR DISCONTINUTITY ISSUE USING MORPHOLOGICAL OPERATIONS

Classification of banana fruit based on its quality and maturity is an interesting application of image processing in agricultural sector for enhancing banana export. Complete automation of this system leads various challenges for researcher in computer vision. Isolating banana fruit from the background image using efficient method is an initial phase of this system. Edge detection using image segmentation is a conventional method used for identifying edges of banana image from entire scene. Major problem in most of the edge detection method is discontinuity of edge which records its higher influence on banana segregation phase. Hence development of better segmentation method helps to resolve discontinuity issue which has higher impact on automation system for banana fruit industry. This paper proposes an efficient and simple edge detection method using convolution and morphological operations. Performance of the proposed method has been compared with the existing first order derivative methods using Mean Square Error (MSE) and Peak Signal to Noise Ratio (PSNR). The result shows significant enhancement in the results by proposed methods compare with existing methods.