Soft decision histogram-based image binarization for enhanced ID recognition

Image binarization is one of the main techniques for image segmentation. It segments an image in to foreground and background. The foreground contains interested objects/characters. Usually binarization is carried out with a threshold found from the histogram of an image automatically. In this paper we formulate binarization using Soft Decision Histogram (SDH). The algorithm effectively integrates color clustering and binary texture analysis, and is capable of handling situations with complex and multiple backgrounds. The (SDH) is able to make a soft decision using image histogram while clustering the pixels for binarization. The main advantage by using soft decision is the occurrence of isolated or outlier pixels are overcome, complex background and noise pixels are removed. The proposed algorithm can be use for retrieving both text and objects. Experiments with images collected from Internet has been carried out and compared with existing techniques, both show the effectiveness of the algorithm.