一种陶瓷砖表面缺陷实时检测算法的优化

Optimization of a Real-Time Detection Algorithm for Surface Defects of Ceramic Tiles

摘要: 为了使陶瓷砖表面花纹纹理缺陷检测适应实际生产流水线快速检测的要求，本文采用实时性强的ORB算法在纹理缺陷检测时完成特征匹配，并对其算法进行优化，同时将其与机器学习中数据挖掘KNN算法结合，增强其准确性以及各种环境下的鲁棒性。通过实验测试，该算法具有良好的实时性、准确性以及适应环境变换的鲁棒性，更能适应实际生产流水线的检测。

Abstract: In order to meet the requirement of rapid detection of texture defects on ceramic tile surface in production line, this paper uses ORB algorithm with strong real-time performance to complete feature matching in texture defect detection, and optimizes its algorithm. At the same time, it combines with KNN algorithm of data mining in machine learning to enhance its accuracy and ro-bustness in various environments. The experimental results show that the algorithm has good re-al-time performance, accuracy and robustness to adapt to environmental changes, and is more suitable for the detection of actual production line.