Adaptive Algorithm for the Quality Control of Braided Sleeving

We describe the development and application of a robot vision based adaptive algorithm for the quality control of the braided sleeving of high pressure hydraulic pipes. With our approach, we can successfully overcome the limitations, such as low reliability and repeatability of braided quality, which result from the visual observation of the braided pipe surface. The braids to be analyzed come in different dimensions, colors, and braiding densities with different types of errors to be detected, as presented in this paper. Therefore, our machine vision system, consisting of a mathematical algorithm for the automatic adaptation to different types of braids and dimensions of pipes, enables the accurate quality control of braided pipe sleevings and offers the potential to be used in the production of braiding lines of pipes. The principles of the measuring method and the required equipment are given in the paper, also containing the mathematical adaptive algorithm formulation. The paper describes the experiments conducted to verify the accuracy of the algorithm. The developed machine vision adaptive control system was successfully tested and is ready for the implementation in industrial applications, thus eliminating human subjectivity.

[1]  Joachim Denzler,et al.  On the Suitability of Different Features for Anomaly Detection in Wire Ropes , 2009, VISIGRAPP.

[2]  Daoliang Li,et al.  Original paper: Classification of foreign fibers in cotton lint using machine vision and multi-class support vector machine , 2010 .

[3]  Tetsuo Asano,et al.  Variants for the Hough Transform for Line Detection , 1996, Comput. Geom..

[4]  Joachim Denzler,et al.  Enhanced anomaly detection in wire ropes by combining structure and appearance , 2013, Pattern Recognit. Lett..

[5]  Ari Visa,et al.  An adaptive texture and shape based defect classification , 1998, Proceedings. Fourteenth International Conference on Pattern Recognition (Cat. No.98EX170).

[6]  Joachim Denzler,et al.  One-Class Classification for Anomaly Detection in Wire Ropes with Gaussian Processes in a Few Lines of Code , 2011, MVA.

[7]  Rafael C. González,et al.  Digital image processing using MATLAB , 2006 .

[8]  Joachim Denzler,et al.  Analysis of Structural Dependencies for the Automatic Visual Inspection of Wire Ropes , 2010, VMV.

[9]  Joachim Denzler,et al.  HMM-Based Defect Localization in Wire Ropes - A New Approach to Unusual Subsequence Recognition , 2009, DAGM-Symposium.

[10]  Xianghua Xie,et al.  A Review of Recent Advances in Surface Defect Detection using Texture analysis Techniques , 2008 .

[11]  Dmitry Chetverikov.  Structural defects: general approach and application to textile inspection , 2000, Proceedings 15th International Conference on Pattern Recognition. ICPR-2000.

[12]  W. Marsden.  I and J , 2012 .

[13]  G. van Straten,et al.  A vision based row detection system for sugar beet , 2005 .