Publication

Abstract : Document images are becoming more popular in today’s world and being made available over the internet. Information retrieval from the document images becomes a difficult task; it is a challenging problem as it compared with digital texts. Edge detection is an important task in the document image retrieval, it indicates to the process of finding and locating sharp discontinuation of characters in the document images. In this work we have compared six different types of edge detection techniques, Roberts, Sobel, Prewitt, Canny, Laplacian (Zero Cross) and Laplacian of Gussian (LOG) are used to extract edge points from different types of document images. Performance factors are analyzed in terms of processing time and accuracy on the basis of Structural Similarity (SSIM). From the experimental results, it is observed that Laplacian and Roberts edge detection technique found as best among other edge detection techniques.