Algorithm of contrast enhancement for visual document images with underexposure

摘要: The visual document image is the electronic about newspapers, books or magazines taken by digital camera, vidicon etc. Whose getting more convenient than got from scanner. Along with development of OCR technology, images could be recognized OCR. Affected some factors, will degraded during its acquisition, processing, transmission. One main problems affecting quality, leading to unpleasant pictures, comes improper exposure light. So preprocessing becoming much significant before recognition in order get an appropriate satisfied requirements. For low contrast underexposure, according image's characteristic, a new algorithm, based on background separation, for object enhance proposed, proposed method calculate threshold separation firstly, And different processing foreground and background: Various gray values merged into unitary value, whereas enhanced. algorithm implemented Visual C++ 6.0, compared result results Otsu's histogram equalization. experimental show clearly that this details adequately, increase rate, avoid block effect at same time.