期刊文献+

基于机器视觉的雨刮器表面缺陷检测

Surface Defects Detection of Wipers Based on Machine Vision
下载PDF
导出
摘要 针对雨刮器表面典型缺陷,比如擦伤、划痕、斑点(凹坑与砂眼)等,提出了一种基于机器视觉的雨刮器表面缺陷检测方法。首先根据CCD采集得到的雨刮器表面图像,对图像采用图像形态学处理与Gabor滤波相结合的方法,构建出图像预处理特定的结构元素,利用其对雨刮器表面纹理噪声进行抑制,并使用伽马灰度图像增强算法增强雨刮器缺陷与背景的对比度,然后采用图像最大熵阈值缺陷分割出雨刮器表面缺陷,最后实现金属雨刮器外壳表面典型缺陷特征提取与分类,并对相关缺陷进行尺寸计算。实验结果表明,文章提出的方法对工业生产中雨刮器表面典型缺陷检测速度达到平均0.724 s/根,具有良好的检测精度,能较好地适应于工业环境中。 Aiming at the common typical surface defects of wiper,such as scratches,spots(pits and sand holes),a method of wiper surface defect detection based on machine vision is proposed.Firstly,according to the surface image of the wiper collected by CCD,the image morphology processing and Gabor filtering method are used to suppress the texture noise of the metal frosting by designing specific structural elements of image preprocessing,and the gamma gray image enhancement algorithm is used to enhance the contrast between the wiper defect and the background.Then,the image maximum entropy threshold defect is used to segment the surface defects of the wiper.Finally,the feature extraction and classification of typical defects on the surface of the metal wiper shell are realized,and the size of the relevant defects is calculated.The experimental results show that the detection speed of the proposed method for typical surface defects of wiper in industrial production reaches an average of 0.724 s/piece,which has good detection accuracy and can be well adapted to the industrial environment.
作者 谢丹 贺福强 何昊 纪家平 XIE Dan;HE Fuqiang;HE Hao;JI Jiaping(College of Mechanical Engineering,Guizhou University,Guiyang 550025)
出处 《计算机与数字工程》 2024年第4期1221-1227,共7页 Computer & Digital Engineering
关键词 雨刮器 形态学 GABOR滤波器 机器视觉 缺陷检测 wiper morphology Gabor filter machine vision defects detection
  • 相关文献

参考文献10

二级参考文献59

  • 1董银文,郭雷,姚俊.一种从噪声图像中提取边缘的启发式搜索算法[J].模式识别与人工智能,2006,19(1):14-19. 被引量:8
  • 2[2]Rautaruukki New Technology. Defect Classification in Surface Inspection of Strip Steel. Steel Times, 1992(5): 214~216
  • 3[3]Badger J C, Enright Sean T. Automated surface inspection system. Iron and Steel Engineer, 1996 (3): 48~51
  • 4[4]Parsytech Computer GmbH. Software controlled on-line surface inspection. Steel Times International, 1998(3): 30~35
  • 5[5]Karayiannis N B. Accelerating the training of feed forward Neural Networks using generalized hebbian rules for inintializing the internal representation. IEEE Transactions on Neural Networks, 1996, (7)2: 419~426
  • 6[6]Sking J, Jorg R. Self-learning fuzzy controllers based on temporal back propagation. IEEE Trans. on Neural Networks, 1992, 3(5): 714~723
  • 7[7]Amari S, Murata N, Muller K R, et al. Asymptotic statistical theory of overtraining and cross-validation. In: Anon. ed. METR 95-06. Tokyo: Dept. of Mathematical Engineering and Information, Physics, Univ. of Tokyo, 1995.
  • 8田娟,郑郁正.模板匹配技术在图像识别中的应用[J].传感器与微系统,2008,27(1):112-114. 被引量:60
  • 9高程程,惠晓威.基于灰度共生矩阵的纹理特征提取[J].计算机系统应用,2010,19(6):195-198. 被引量:175
  • 10刘中华,殷俊,金忠.一种自适应的Gabor图像特征抽取和权重选择的人脸识别方法[J].光子学报,2011,40(4):636-641. 被引量:12

共引文献291

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部