Savings Slip Handwritten Extraction And Purification

Today, more and more works has been done automatically by developing of science technology. As an essential operation in many business and government organizations on telecommunication, health care ,finance. insurance, and public utilities, form processing remains a labor-intensive task, and the automation of this procedure has attracted intensive research interests. Chinese handwritings are important information in bank checks, we must separate them from the image and save them in binary image in order to recognize it. It can be used for either OCR system or structuring electric image document. The system is proposed to automatically extract and clean handwritten items from business forms. Handwritten data usually touch or cross preprinted form frames and texts. The paper describes components and arithmetic in the system. The automatic bank check processing system includes six important parts: smoothing, baseline location ,baseline delete, information restore, image segment, clean. Important arithmetic: morphological erosion, morphological dilate, open and close, Hough transform, region increase ,projection, etc. When the handwritings are found touching or crossing preprinted texts, morphological operations based on statistical features are used to clean them. If filled-images are slant, which can be rectified by Hough transform. The subjective evaluations show promising results of the proposed system.