Abstract：Ultrasound image defects are classified at a low accuracy due to problems such as small sample size, large sample categories, and difficulty in distinguishing. Aiming at these problems, an ultrasonic image defect classification method based on support vector machine optimized by genetic algorithm was proposed. First, the feature data of ultrasonic image defect is extracted by image processing. Then, the support vector machine was trained as ultrasonic image defect classifiers. Finally the parameters of the classifiers were optimized by genetic algorithm to obtain the optimal classifiers. The experimental results show that the proposed ultrasonic image defect classifier is superior to the other methods in the recognition rate, and the comprehensive recognition rate reaches 90%, which can effectively assist the staff to classify and identify the ultrasonic image defects.
［1］倪豪,郑慧峰,王月兵,等. 基于自动种子区域生长的超声B图像缺陷分割方法［J］. 计量学报, 2018, 39(6): 878-883.
Ni H, Zheng H F, Wang Y B, et al. Ultrasonic B Image Defect Segmentation Method Based on Automatic Seeded Region Growing［J］. Acta Metrologica Sinica, 2018, 39(6): 878-883.
［2］张惠璇. 金属薄板缺陷超声检测技术与信号处理研究［D］. 太原: 中北大学, 2017.
［3］胡文刚. 基于多值域特征及数据融合的焊缝缺陷超声检测与识别［D］. 哈尔滨: 哈尔滨工业大学, 2012.
［4］张明星. X射线钢管焊缝缺陷的图像处理与识别技术研究［D］. 成都: 电子科技大学, 2015.
［5］Cenate C F T, Rani B S, Ramadevi R, et al. Optimization of the Cascade Feed Forward Back Propagation network for defect classification in ultrasonic images［J］. Russian Journal of Nondestructive Testing, 2016, 52(10):557-568.
［6］丁世飞, 齐丙娟, 谭红艳. 支持向量机理论与算法研究综述［J］. 电子科技大学学报, 2011, 40(1):2-10.
Ding S f, Qi B J, Tan H Y. A Survey of Support Vector Machine Theory and Algorithms［J］. Journal of University of Electronic Science and Technology of China, 2011, 40(1): 2-10.
［7］刘雪鸥. 医学图像模式识别技术的研究及应用［D］. 太原: 太原理工大学, 2016.
［8］Du P, Tan K, Xing X. A novel binary tree support vector machine for hyperspectral remote sensing image classification［J］. Optics Communications, 2012, 285(13-14):3054-3060. ［J］.
［9］王梅. 一种改进的核函数参数选择方法［D］. 西安: 西安科技大学, 2011.
［10］郭利斌. 基于支持向量机算法的癌症组织拉曼光谱数据分析［D］. 福州: 福建师范大学, 2015.
［11］杜必强, 孙立江. 基于PSO-SVM模型的焊接转子环焊缝超声缺陷识别［J］. 动力工程学报, 2017, 37(5):379-385.
Du B Q, Sun L J. Ultrasonic Defect Identification of Welded Rotor Ring Welds Based on PSO-SVM Model［J］. Journal of Power Engineering, 2017, 37(5): 379-385.
［12］Pang-Ning Tan, Michael Steinbach, Vipin Kumar. 数据挖掘导论(完整版)［M］. 2版. 范明,范宏建,译. 北京: 人民邮电出版社, 2011.
［13］雷亚国, 陈吴, 李乃鹏,等. 自适应多核组合相关向量机预测方法及其在机械设备剩余寿命预测中的应用［J］. 机械工程学报, 2016, 52(1):87-93.
Lei Y G, Chen W, Li N P, et al. Adaptive multi-core combined correlation vector machine prediction method and its application in mechanical equipment residual life prediction［J］. Journal of Mechanical Engineering, 2016, 52(1): 87-93.
［14］茅嫣蕾, 魏贇, 贾佳. 一种基于KKT条件和壳向量的SVM增量学习算法［J］. 电子科技, 2016, 29(2):38-40.
Mao Y L, Wei Y, Jia J. An SVM Incremental Learning Algorithm Based on KKT Condition and Shell Vector［J］. Electronic Science and Technology, 2016, 29(2): 38-40.
［15］丁勇, 秦晓明, 何寒晖. 支持向量机的参数优化及其文本分类中的应用［J］. 计算机仿真, 2010, 27(11):187-190.
Ding Y, Qin X M, He H H. Parameter Optimization of Support Vector Machine and Its Application in Text Classification［J］. Computer Simulation, 2010, 27(11): 187-190.
［16］李琳, 张晓龙. 基于RBF核的SVM学习算法的优化计算［J］. 计算机工程与应用, 2006, 42(29):190-192.
Li L, Zhang X L. Optimization calculation of SVM learning algorithm based on RBF kernel［J］. Computer Engineering and Applications, 2006, 42(29): 190-192.
［17］赵长春, 姜晓爱, 金英汉. 非线性回归支持向量机的SMO算法改进［J］. 北京航空航天大学学报, 2014, 40(1):125-130.
Zhao C C, Jiang X A, Jin Y H. Improvement of SMO Algorithm for Nonlinear Regression Support Vector Machine［J］. Journal of Beijing University of Aeronautics and Astronautics, 2014, 40(1): 125-130.
2019, Vol. 40 
