Document page segmentation by integrating distributed soft decisions

A new algorithm for image segmentation is suggested. The suggested algorithm is based on making local soft decisions on small blocks and integrating them to reduce their “ambiguities” and increase their “confidence” as more contextual evidence is obtained from the image data. Integration is performed by vote propagation across windows (within and across scales) using majorities of weighted votes. The algorithm has been tested on some document page decomposition tasks; the results of these tests are presented. This method is based on parallel, distributed and independent computations and has low complexity. It is general and can be applied to different segmentation and classification tasks