A comparison of visual attention models for the salient text content detection in natural scene

Visual saliency approaches have been acquainted to recognize the most noticeable object in the scene that are going to pull in human attention. These methodologies have been adequately utilized for the prominent region detection. The text or character detection as a salient region in image remains a challenging research problem. Intrinsically about the image, a text in the scene conveys vital information. This paper evaluates six state-of-art models of visual attention in context to scene content (i.e. text or characters). Saliency map generated by these models of visual saliency have been intended for facilitating salient text detection in a natural scene by highlighting text and concealing non-text regions.