本文將機器視覺與數字圖像處理技術引入到織物疵點檢測中，提出了一種織物疵點檢測算法——圖像距離差算法。該算法可使用戶根據織物的類型，自行設置相應的疵點檢測控制參數，能夠檢測出三十多種常見的疵點。實驗證明該算法具有識別正確率高，識別速度快等優點。
關鍵詞:疵點檢測，動態閾值，二值化，圖像處理

Fabric Defect Detection Method Based on Image Distance Difference LU Yun1,2，ZHANG Jing-miao3，JIANG Jian-wei1 ?1．College of Information and Technology, Hebei University of Economics and Business，Shijiazhuang，050061； 2．College of Information science and engineering, Hebei University of Science and Technology, Shijiazhuang, 050054； 3．Computer Center, Hebei University of Economics and Business, Shijiazhuang, 050061) Abstract: By using machine vision and digital image processing technology to fabric defect detection, this paper brings forward a new method of detection of fabric defect, namely image distance difference arithmetic. The arithmetic permit user to set appropriate control parameter of fabric defect defection based on the type of the fabric. It can detect more than 30 kinds of common defects, and by experiments proved it has advantages of high identification correctness and fast inspection speed.
Keywords: fabric defect detection, dynamic threshold, two-valuezition, image measure