Date of Publication :11th January 2018

Abstract: Nowadays caption in digital media images, capture images or natural images as important for text retrieval in images. In this paper, we proposed a fast text localization method in which Sobel edge detection and clustering method are used to localize all the possible edges of the image. As text is made up of edge strokes and opposite edge pair lies in the text. Sobel edge detection algorithm finds the magnitude of edges in X and Y direction and angle of edge pixels. The clustering method is used to group the text having same properties while Optical character recognition method to recognize is the localized edges are text or not. As these recognized texts can be useful for many purposes like license plate reading, sign detection and visually impair to get the text for accessing text to speech algorithm. Experimental results show the improvements in the proposed algorithm.

Reference :