Identification Algorithm Of Paper Defects Base On Threshold Segmentation And Fractal Characteristics

With the continuous development of the paper industry in China, the increasing web widthand machine speed. In the production process, it is not easy to inspect the paper defects bymanual inspection, which caused by equipments abrasion, quality of raw material andenvironment pollution. In practical production, accurate judgment of a paper disease type anddetermine causes, which is important to monitor production quality.This paper focuses on image processing algorithms of paper defects, gray feature andfractal feature of paper image, paper defects detection and recognition experiment systemdesign, including study the algorithms, the image processing algorithms, algorithm proceduresrealize. Aiming at the problems of typical paper defects identification such as specks, void, flawand goffer, through the analysis of the image gray characteristics and fractal feature, arecognized algorithm of paper defects base on threshold segmentation and fractal characteristicsis proposed. The algorithm using dual threshold segmentation paper disease area and judgingpaper disease image shape by the box dimension of fractal characteristics. Based on thesetheories,designing the paper defect recognition algorithm, experiment results shows that thisapproach is effective.The method applied in the system still has a long distance to the utility because of themanifold paper defects, although this paper has acquired some achievements. Therefore, thispaper concludes deficiencies in research and gives suggestions for focus of next work.