Application Of Digital Image Processing In Office Automation System

On the one hand, we all know the advantages of office automation system system can besummarized in three points. First, a variety of office automation systems can greatly reducestaff repetitive mechanical labor, reduce the burden on staff, improve administrativeefficiency. Second, the office automation system can reduce the probability of error of manualoperation, to reduce the loss caused by the error. Third, office automation system to limit therights of participants, it is advantageous to prevent emergence favoritism behavior. Therefore,various types of office automation system has long been widely used in a variety of occasions.On the other hand, digital image processing technology has made great progress, especially inthe Adaboost face detection algorithm for the emergence of a series of excellentrepresentatives, makes digital image processing has a broad application space in their dailywork and study. And the existing computing power of personal computers have long beenable to smoothly run these algorithms. And enhance the processing capabilities of personalcomputers, and digital image processing related to office automation system is technicallyripe. In this context, and digital image processing-related office automation system istechnically ripe.To keep up the pace to keep up with the information, the paper chose two officeautomation systems and digital image processing issues related to research. These two issuesare based on objective questions answer sheet automatic marking of digital image processingtechnology and paper files HD technology. The traditional answer card reader is based on theOMR (optical mark reader device) developed compared to digital image processing, digitalimage processing technology intelligence and greater flexibility with better, can be easilyovercome OMR reader specifically dedicated card disadvantages. Automatic definition ofpaper files can be reduced burden on staff in the value of paper files to completehigh-definition processing.In answer card issues, to high accuracy positioning Tiantu area should prevail. Tiantuarea usually answer on the answer card was equally spaced array-like distribution, for thisfeature, horizontal and vertical positioning projection can be very, very good answer cardTiantu area. Determine the answer card after Tiantu region, according to each area is blackedarea Tiantu Tiantu area of automatic classification, principles of classification for the principle of maximum variance between classes. In the definition of the file in question, thispaper solves the tilt correction image files, automatic extraction passport, stamp colordeepened, writing deepened and background replacement. Tilt correction of the image usingthe Hough transform to find the image file straight, straight as an image of the tilt angle ofdeflection angle, then the image affine transformation. In the automatic extraction of passportproblems, Adaboost face detection to find images of human face region, locate the appropriatepost by scaling is the photo area. Seal the color deepened, use Gaussian mixture model formodeling stamp color, find the area of the stamp, to increase the color saturation of theposition. Writing deepened and background replacement method with the stamp of handlingsimilar, this article does not go into details.