Automatic character location and segmentation in color scene images

Abstract

This paper describes a connected component (CC)-based approach to automatic text location and segmentation in natural scene images. A multi-group decomposition scheme is used to deal with the complexity of the color background. Connected component extraction is implemented using the block adjacency graph (BAG) algorithm after noise filtering and runlength smearing (RLS) operation. Some heuristic features and priority adaptive segmentation (PAS) of characters are proposed in block candidate verification and grayscale-based recognition. A prototype system is completed and the experimental results prove the effectiveness of the proposed method.