【摘要】 刀具磨损监测是自动化生产的核心技术之一，为保证监测结果可靠性，提出一种基于光栅投影图像的硬质合金刀具磨损监测方法。使用光栅投影检测系统收集被测物体栅线图样，计算光栅在投影墙面的长度，运用格雷码编码算法划分光栅条带信息，通过三维信息重构获得刀具形貌数据；扫描所有图像数据，将稀疏数据作为孤立点，采用基于层次结构的平衡迭代算法聚类相同特征图像，使用强度均值、属性和清晰度三个指标优化图像质量；在视觉显著特征基础上，利用灰度密度函数梯度找出光栅投影图像突变点，采用小波分解划分刀具磨损图像分辨率图层，运用扩散操作处理图像噪声、光照不均等问题，通过高斯函数锁定刀具磨损方位。仿真结果表明，所提方法的刀具磨损监测精度较高，可信性强，具有极大的应用空间。更多还原

【Abstract】 Tool wear monitoring is one of the core technologies of automatic production. In order to ensure the reliability of monitoring results, this article presented a method of monitoring cemented carbide tool wear based on raster projection images. Firstly, we used the grating projection detection system to collect the pattern of the gate line of the measured object and calculated the length of the grating on the projection wall. Secondly, we used the gray coding algorithm to divide the grating strip information, and thus to obtain the tool profile data through three-dimensional information reconstruction. After scanning all image data, we regarded the sparse data as isolated points, and clustered the same feature images by the balanced iterative algorithm based on hierarchical structure. Secondly, we used three indicators, namely the intensity mean, attribute and clarity, to optimize the image quality. On the basis of the visual salient feature, we used the gradient of the gray density function to find out the abrupt change point of the grating projection image. Moreover, we divided the resolution layer of the tool wear image by wavelet decomposition, and solved the problem of image noise and uneven illumination by the diffusion operation. Finally, the tool wear orientation was locked by the Gaussian function. Simulation results show that the proposed method has high tool wear monitoring accuracy and reliability, as well as wide application space.更多还原