Edge detection is important in image analysis to form the shape of an object. Edge is the boundary between different textures, which helps with object segmentation and recognition. Currently, several edge detection techniques are able to identify objects but are unable to localize the shape of an object. To address this problem, this paper proposes a fusion of selected edge detection algorithms with mathematical morphology to enhance the ability to detect the object shape boundary. Edge detection algorithm is used to simplify image data by minimizing the amount of pixel to be processed, whereas the mathematical morphology is used for smoothing effects and localizing the object shape using mathematical theory sets. The discussion section focuses on the improved edge map and boundary morphology (EmaBm) algorithm as a new technique for shape boundary recognition. A comparative analysis of various edge detection algorithms is presented. It reveals that the LoG's edge detection embedded in EmaBM algorithm performs better than the other edge detection algorithms for fruit shape boundary recognition. Implementation of the proposed method shows that it is robust and applicable for various kind of fruit images and is more accurate than the existing edge detection algorithms.