基于单视角图像处理的刀具磨损检测研究

针对现有刀具磨损检测存在检测速度慢、人工工作量大、经济效益低等问题,提出一种基于单视角图像处理的刀具磨损检测算法。使用CCD相机对刀具磨损前和磨损后的图像进行单视角提取,通过图像融合变换得到磨损前和磨损后的全景深刀具图像;将刀具磨损前后的全景深图像进行配准,得到磨损区域图像后,将磨损区域图像的像素转化成实际长度值,计算出实际磨损量。实验结果表明:该方案检测速度快,测量误差范围满足检测精度要求,可应用于实际加工。

Research on Tool Wear Detection Based on Single Visual Angle Image Processing

Aiming at the existing problems of tool wear detection,such as slow detection speed,large manual workload and low economic benefit,a tool wear detection algorithm based on single perspective image processing was proposed.The CCD camera was used to extract the images before and after wear from a single perspective.The panoramic deep tool images before and after wear were obtained through image fusion and transformation,and the panoramic deep images before and after wear were registered.After the wear area image was obtained,the pixel of the wear area image was converted into the actual length value,and the actual wear amount was calculated.The experimental results show that the proposed method has high detection speed,and the measurement error range meets the requirement of detection accuracy,so it can be applied to actual production.