Fuzzy segmentation for document image analysis

This paper describes a segmentation method, called the fuzzy segmentation method (FSM), for the analysis of document images. Based on the fuzzy set theory, we define fuzzy blank and fuzzy black lines, and based on these definitions we then define fuzzy blank and fuzzy black blocks. An efficient algorithm is proposed here for their identification. The idea of λ-cut sets is utilized when determining the vertical and horizontal thresholds in the segmentation process. With the definition of fuzzy sets, such a process can be automated in an adaptive process. When comparing to existing approaches, the method gave better results in correct identification of blocks in the document image