Research on the Algorithm of Defect Feature Extraction in Trademark Printing

Abstract:

A defect feature extraction algorithm which depends on the separation and subtraction of color image and mathematical morphology is proposed in this paper, based on the defects in trademark printing processes, such as smudges, pin holes, misprints, misting etc. The difference images of each monochrome are operated with morphology methods and then the defect information of high precision printing processes in colorful trademark, such as shape, boundary contour, quantity, size and location etc, are extracted and recognized with suitable structure elements. The defect extraction algorithm is programmed and run by MATLAB and the results validated the high speed and precision of the algorithm about extracting some kinds of defect information in color trademark printing processes. The algorithm has met the real-time requirements of defect feature extraction in trademark printing processes. This research laid a theoretical foundation for researching and developing the color trademark quality detection system.

• Periodical: Applied Mechanics and Materials (Volume 731)
• Pages: 205-209
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.731.205
• Online since: January 2015
• Authors: Yu Chen*, Bang Gui He, Wen Juan Gu
• Keywords: Defect Feature Extraction Algorithm, Mathematical Morphology, Trademark Printing

References

[1] Zhong Hui. Algorithm Study on the Automatic Detection System for Image Defect of Colors Printer. Jilin: Jilin University, (2007).

[2] Gao Shengli. Research on Workpiece Detection Technology based on Machine Vision. Zhengzhou: The PLA Information Engineering University, (2006).

[3] S. H. Hajimowlana. Detection in Web inspection Using Fuzzy Fusion of Texture Features. Circuits and Systems, 2000, 15(3): 18~21.

[4] Luo. Z. Zhang．Automatic Color Printing Inspection by Image Processing. Journal of Materials Technology. 2003 (139): 373-378.

[5] Shi Jianfang, Zhang Fujun, Hao Baofeng. Image Segmentation Algorithm base on Wavelet Transform and Mathematical Morphology. Journal of Taiyuan University of Technology, (2009).

[6] Dai Xiaohong, Wang Guangli. Inspection and Recognition on Defects of Printed Matter Based on Machine Vision. Semiconductor Optoelectronics. 2011 (10).

[7] Liu Bingqing. Research of Print Defects Recognition System based on Machine Vision. Xi'An University of science and technology, (2010).

[8] Zhang Xue-wu, Ding Yan-qiong, Lv Yan-yun, Shi Ai-ye, Liang Rui-yu. A vision inspection system for the surface defects of strongly reflected metal based on multi-class SVM. Expert Systems with Applications, 2011, 38(5): 5930-5939.

DOI: 10.1016/j.eswa.2010.11.030

[9] Guo xuan. Research on Visual detection algorithms of color Printed matter. Huazhong University of Science and Technology, (2010).

[10] Hou Baosheng. Method of Image Edge Detection Based on Extended Mathematical Morphology. Modern Electronics Technique. 2010 (08).

[11] Zhang Zheng, Wang Yanpin, Xue Guixiang. Digital Image Processing and Machine Vision-Visual C++ and Matlab. Beijing: Addison Wesley Publishing, 2013: 325-331.