Generic Feature Selection and Document Processing

This paper presents a generic features selection method and its applications on some document analysis problems. The method is based on a genetic algorithm (GA), whose tness function is de ned by combining Adaboot classi ers associated with each feature. Our method is not linked to a classi er achieving the nal recognition task; we have used a combination of weak classi ers to evaluate a subset of features. So we select features that can further be used in the most appropriate classi ers. This method has been tested on three applications: Drop caps classi cation, handwritten digits recognition and text detection. The results show the ef ciency and robustness of the proposed approach.

• Type: Conference
• Year: 2009
• Where: ICDAR
• Authors: Hassan Chouaib, Nicole Vincent, Florence Cloppet, Salvatore Tabbone
• Keywords: Adaboot Classi Ers, Classi Ers, Document Analysis, ICDAR 2009, Weak Classi Ers