基于改进K-means算法的工件表面缺陷分割算法研究

Research on Workpiece Surface Defect Segmentation Algorithm Based on Improved K-means Algorithm

工件表面缺陷的存在影响工件产品的质量以及工件的安全使用,传统的工件表面缺陷检测由人工完成,工作量大且易受到检测人员主观因素的影响,很难保证检测的效率与精度.本文提出了一种基于改进的K-means算法的工件表面缺陷分割算法,将自适应人类学习优化算法应用到Kmeans聚类算法中,使自适应人类学习优化算法初始化K-means聚类算法的聚类中心,最后将改进的K-means聚类算法结合形态学进行工件表面缺陷的检测.实验表明,该算法能够较理想的分割出工件表面的缺陷,具有分割精度高、实用价值较好的特性.

The existence of workpiece surface defects affects the quality of workpiece products and the safe use of workpiece.The traditional surface defect detection of workpiece is completed by manual work,and it is easy to be affected by subjective factors of inspectors,so it is difficult to ensure the efficiency and accuracy of detection.In this paper,an improved k-means algorithm for workpiece surface defect segmentation is proposed.The adaptive human learning optimization algorithm is applied to the K-means clustering algorithm.The adaptive human learning optimization algorithm initializes the clustering center of the K-means clustering algorithm.Finally,the improved k-means clustering algorithm is combined with morphology to detect the surface defects of workpiece.The experimental results show that the algorithm can segment the surface defects of the workpiece with high segmentation accuracy and can achieve the purpose of workpiece surface defect detection,which proves that the algorithm has good practical value.