Chinese document image retrieval system based on proportion of black pixel area in a character image

In order to preserve the original state of a document, a document is usually saved in computer in image format as backup data after a scanner scans it. Presently, many retrieval systems used to deal with this sort of duplicate document images have been proposed, but most of them are only suitable for English duplicate document images. This paper proposes a system for Chinese duplicate document images, which uses the proportion of black pixel area in each character image as the feature of this character image. According to experimental results, the proposed system can efficiently find out the desired duplicate document image.