ZHANG Rui-ping1, SHI Jia-yue1, GOU Jun-nian1, DONG Hai-ying1,2, AN Mei1
（1. School of Automation and Electrical Engineering, Lanzhou Jiaotong University, Gansu Lanzhou 730070, China;2. School of New Energy and Power Engineering, Lanzhou Jiaotong University, Gansu Lanzhou 730070, China）
Abstract： In order to solve the problem of internal defect detection in industry, an intelligent detection method for workpiece defect based on industrial computed tomography (CT) images is proposed. The industrial CT slice image is preprocessed first with the combination of adaptive median filtering and adaptive weighted average filtering by analyzing the characteristics of the industrial CT slice images. Then an image segmentation algorithm based on gray change rate is used to segment low contrast information in industrial CT images, and the feature of workpiece defect is extracted by using Hu invariant moment. On this basis, the radial basis function (RBF) neural network model is established and the firefly algorithm is used for optimization, and the intelligent identification of the internal defects of the workpiece is completed. Simulation results show that this method can effectively improve the accuracy of defect identification and provide a theoretical basis for the detection of internal defects in industry.
Key words： industrial computed tomography (CT); defect detection; image segmentation; feature extraction; intelligent identification
CLD number： TN911.73; TH878 Document code： A
Article ID： 1674-8042（2019）03-0299-08 doi： 103969/jissn1674-8042201903014
References
［1］Lu R, Shen Y. An improved image threshold segmentation method for two-dimensional histogram. Systems Engineering and Electronic Technology, 2004, 26(10): 1487-1490.
［2］Wu J, Li J, Liu C Y, et al. Method of selecting gray threshold value based on maximum entropy. Journal of Wuhan University of Science and Technology, 2004, 27(1): 158-60.
［3］Jin R B. Research and implementation of mechanical workpiece defect detection based on acoustic emission technology. Nanjing: Southeast University, 2015: 6-11.
［4］Zhang Y F. Detection of workpiece defects based on photoelectric sensors and related technologies. Sensor Technology, 2004, 23(9): 82-83.
［5］Tang W, Yu G Y, et al. Research on the detection method of surface artifact defects based on Gabor characteristics. Journal of China Institute of Metrology, 2015, 26(1): 46-49.
［6］Jiang C Y, Liu Y, Wei S F. Metal workpiece surface defect detection algorithm based on GPU research. Journal of Computer Applications and Software, 2015, 32(3): 271-274.
［7］Qiao L, Zhao E D, Liu J J, et al. Research on the detection method of workpiece defect based on CNN. Computer science, 2014, 44(11A): 238-242.
［8］Masooleh M G, Moosavi S A. An improved fuzzy algorithm for image segmentation. In: Proceedings of World Academy of Science,Engineering and Technology, 2008, 28(4): 400-404.
［9］Shi W, Zhu X F, Zhu G. Adaptive edge detection algorithm based on morphology of MRI. Journal of Instrumentation, 2013, 34(2): 408-414.
［10］Nie S P, Wang M, Liu F. Low contrast image segmentation algorithm research. China Laser, 2004, 31(1): 89-91.
［11］Lu J F, Lin H, Pan Z G. Application of adaptive regional growth algorithm in medical image segmentation. Journal of Computer Aided Design and Graphics, 2005, 17(10): 2169-2173.
［12］Ng P E, Ma K K. A switching median filter with boundary discriminative noise detection for extremely corrupted images. IEEE Transactions on Image Processing, 2006, 15(6): 1506-1516.
［13］Huang K D, Zhang D H, Jin Y F. Low contrast cone-beam CT image defect segmentation algorithm. Nuclear Technology and Detection Technology, 2009, 29(1): 132-136.
［14］Wang Y, Chen J Z, Tan H, et al. Study on noise characteristics of industrial CT detection system. Nuclear Electronics and Detection Technology, 2010, 30(7): 929-934.
［15］Jing L Y, Song H, L X S. Research and application on job shop planning based on improved glowworm swarm optimization algorithm. In: Proceedings of the Fifth International Conference on Intelligent Human-Machine Systems and Cybernetics. Hangzhou: IEEE, 2013: 139-143.
［16］Liu V, Lariviere N R, Wang G. X-ray micro-CT with a displaced detector array. Medicial Physics, 2003, 29(7): 1634-1636.
基于工业CT图像的工件缺陷智能检测
张蕊萍1， 时佳悦1， 苟军年1， 董海鹰1,2， 安玫1
（1. 兰州交通大学 自动化与电气工程学院, 甘肃 兰州 730070； 2. 兰州交通大学 新能源与动力工程学院, 甘肃 兰州 730070）
摘要:针对工业内部缺陷检测问题， 提出了基于工业（Computed tomography,CT）图像的工件缺陷智能检测方法。 通过分析工业CT切片图像自身特点， 提出了以自适应中值滤波和自适应加权均值滤波相结合的方法对工业CT切片图像进行预处理， 采用基于灰度变化率的图像缺陷分割算法对工业CT图像中的低对比度信息进行分割， 利用Hu不变矩方法对工件缺陷进行特征提取。 在此基础上， 建立了RBF神经网络模型， 采用萤火虫算法进行优化， 进而完成对工件内部缺陷的智能识别。 仿真结果表明， 该方法能够有效地提高缺陷识别准确率， 为工业内部缺陷检测提供理论依据。
关键词:工业CT； 缺陷检测； 图像分割； 特征提取； 智能识别
引用格式：ZHANG Rui-ping, SHI Jia-yue, GOU Jun-nian, et al. Intelligent detection method for workpiece defect based on industrial CT image. Journal of Measurement Science and Instrumentation, 2019, 10（3）： 299-306. ［doi： 103969／jissn1674-8042201903014］
[full text view]
