以物件表面紋理特徵做為工業視覺檢測技術已逐漸成為主要趨勢之一。紋理特徵係藉由像素間的空間域關係來描述平滑度、粗糙度和型態規律性等區域特徵訊息。但空間域法易受到光源和雜訊影響，需融合BLOB分析才能達到精確的區塊特徵檢測目的。本研究因而設計了一個視覺檢測平台，結合影像前處理、BLOB分析、紋理分析功能模組，以及一個機率神經網路分類器，提供應用系統開發者做為紋理特徵選擇策略，可針對不同檢測應用快速選擇最佳的紋理特徵組合，形成智慧化的視覺檢測系統。本文最後採用地瓜品質案例，來驗證我們設計的視覺檢測平台，實驗結果顯示，整體紋理特徵平均辨識率可達76.28%。本系統結合PNN神經網路的智慧型選擇策略，擁有彈性化生成各類應用視覺檢測系統的優點。

Object surface texture features have been increasingly applied in the technology of industrial visual inspection. Texture features refer to region features such as smoothness, roughness, and texture regularity described using inter-pixel spatial domain relationships. However, the spatial domain method is susceptible to light sources and noise, and consequently blob analysis must be incorporated to achieve accurate inspection of block features. To provide application developers with strategies for selecting texture features, this study designed a visual inspection platform that integrates visual preprocessing, blob analysis, and texture analysis modules, as well as a probabilistic neural network classifier. This platform is a smart visual inspection system that enables rapid selection of optimal texture-feature combinations in various inspections. Finally, the quality of sweet potatoes was used to verify the visual inspection platform developed in this study. Overall, the experimental results indicated an average recognition rate of 76.27% for the texture features of the sweet potatoes. By incorporating probabilistic neural network-based smart selection strategies, this platform can flexibly generate various applications of inspection systems.

[1] J. A. Jaleel, S. Salim, and S. Archana, “Textural Features Based Computer Aided Diagnostic System for Mammogram Mass Classification,” IEEE Control, Instrumentation, Communication and Computational Technologies(ICICCT), pp.806-811,Jul. 2014.
[2] X. Wang, T. X. Han, and S. Yan, “An hog-lbp human detector with partial occlusion handlingin,” IEEE 12th International Conference on Computer Vision, pp. 32–39, 2009.
[3] X. Tan, and B. Triggs, “Enhanced local texture feature sets for face recognition under difficult lighting conditions,” IEEE Transactions on Image Processing, vol. 19, No. 6, pp. 1635–1650, Jun. 2010.
[4] Y. Xu, D. Xu, S. Lin, T. X. Han, X. Cao, and X. Li , “Detection of sudden pedestrian crossings for driving assistance systems,” IEEE Transactions on Systems, Man, and Cybernetics, Part B, vol. 42, No. 3, pp. 729–739, Jun. 2012.
[5] F. Lehmann, “Turbo Segmentation of Textured Images, ” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol.33, No.1, pp.16-29, Jan. 2011.
[6] T. Randen, and J. H. Husoy, “Filtering for Texture Classification: A Comparative Study,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 21, No. 4, pp. 291-310, Apr. 1999.
[7] Q. Zhong, Z. Chen, X. Zhang, and G. Hu, "Feature-based object location of IC pins by using fast run length encoding BLOB analysis," IEEE Transactions on Components, Packaging and Manufacturing Technology, Vol.4, No.11, pp. 1887-1898, Nov. 2014.
[8] J. Zavadil, J.Tuma, and V. M. F. Santos, "Traffic signs detection using blob analysis and pattern recognition," IEEE Carpathian Control Conference(ICCC), pp.776-779, May. 2012.
[9] D. D. Nguyen, T. C. Pham, X. D. Pham, S. H. Jin, and J. W. Jeon, "Finger extraction from scene with grayscale morphology and BLOB analysis," IEEE International Conference on Robotics and Biomimetics, pp.324-329,Feb. 2009.
[10] D. Simonnet, S. A. Velastin, E. Turkbeyler, and J. Orwell, "Backgroundless detection of pedestrians in cluttered conditions based on monocular images: a review," IET Computer Vision, Vol.6, No.6, pp.540-550, Nov. 2013.
[11] X. Bao, X. Peng, Y. Wang and Z. Cao, "Textile Image Segmentation Based on Semi-supervised Clustering and Bayes Decision," IEEE International Conference on Artificial Intelligence and Computational Intelligence, Vol.3, pp.559-562, Nov. 2009.
[12] H. Fujiwara, Zhong Zhang, H. Toda, and H. Kawabata, "Textile surface inspection by using translation invariant wavelet transform," IEEE International Symposium on Computational Intelligence in Robotics and Automation, Vol.3, pp.1427-1432, Jul. 2003.
[13] Y. Zhou, G. Song, and M. Fang, "The Automating Detecting of Stitch Distortion in Knitted Fabric by Image Processing Technology," IEEE International Conference on Control, Automation and Systems Engineering(CASE), pp.1-3, Jul. 2011.
[14] I. Foster, C. Kesselman, and S. Tuecke, "The Anatomy of the Grid: Enabling Scalable Virtual Organizations," International Journal of High Performance Computing Applications,vol.15,No.3,pp.200-222,2001.
[15] Q. Zhu, "Exploration and improvement of Ostu threshold segmentation algorithm," IEEE Intelligent Control and Automation, pp.6183-6188, Jul. 2010.
[16] I. S. Msiza1, " Fingerprint segmentation：An investigation of various techniques and a parameter study of a variance-based method," International Journal of Innovative Computing,Vol.7,pp.5313-5326,Sep. 2011.
[17] R. M. Haralick, K. Shanmugam, and H. Dinstein, "Textural Features for Image Classification," IEEE Transactions on Systems, Man and Cybernetics, Vol. SMC-3, No.6, pp.610-621, Nov. 1973.
[18] D. J. Marceau, P. J. Howarth, J. M. Dubois, and D. J. Gratton, "Evaluation Of The Grey-level Co-occurrence Matrix Method For Land-cover Classification Using Spot Imagery," IEEE Transactions on Geoscience and Remote Sensing, Vol. 28, No. 4, pp. 513-519, Jul. 1990.
[19] D. G. Barber, and E. F. LeDrew, "SAR Sea Ice Discrimination Using Texture Statistics :A Multivariate Approach," Photogrammetric Engineering & Remote sensing, Vol.57, No.4, pp.385-395, Apr. 1991.
[20] A. Zaknich, "Introduction to the Modified Probabilistic Neural Network for General Signal Processing Applications," IEEE Transactions On Signal Processing, Vol. 46, No. 7, pp. 1980-1990 , Aug. 1998.
[21] J. Peng ,“On Parzen windows classifiers,” IEEE Applied Imagery Pattern Recognition Workshop, pp1-4, Oct. 2014.
[22] Z. Yong-peng, and Z. Huai-zhou, "Modeling and simulation for Equipment support system based on IDEF Method," IEEE International Conference on Computer Application and System Modeling,Vol.6,pp.19-22,Oct. 2010.
[23] R. David, "Grafcet: A powerful tool for specification of logic controllers," IEEE Transactions on Control Systems Technology, Vol.3, No.3, pp.253-268, Sep. 1995.
[24] F. Schumacher, and A. Fay, "Transforming Time Constraints of a GRAFCET graph into a suitable Petri net formalism," IEEE International Conference on Industrial Technology, pp.210-218, Feb. 2013.