一种基于机器视觉的平面加工机床控制系统的设计

针对服装、包装等加工行业中须将人工测量的纸质图纸或模型样件的尺寸信息录入计算机并转换成电子加工图纸而导致的加工周期长、生产效率低的问题，提出了一种基于机器视觉的平面加工机床控制系统，以实现对纸质图纸或模型样件的快速检测。采用“ARM+DSP”方式搭建了主从式运动控制系统，设计了系统各部分功能模块。构建了“工控机+工业CCD （charge coupled device，电荷耦合器件）相机+光源控制”的视觉检测系统，结合FAWS（feature adaptive wavelet shrinkage, 自适应特征的小波收缩）算法和麻雀搜索算法提出一种改进的FAWS算法进行图像降噪，并采用Canny算法进行图像边缘检测，实现图像轮廓的准确提取。设计了图像轮廓提取、轮廓数据转换为加工数据、数据通信等处理程序，实现了基于机器视觉的快速检测以及在系统加工过程中的人机交互。最后，对系统进行了实验测试，对实际加工效果进行了评价。结果表明，采用所研制的平面加工机床控制系统不仅能显著提高生产效率，而且能减小图像轮廓的误差。其性能稳定可靠，具有一定的工程实用价值。

Design of a control system of plane machining machine tool based on machine vision

In order to solve the problems of long processing cycle and low production efficiency caused by inputting the size information of manually measured paper drawings or model samples into the computer and converting it into electronic processing drawings in garment, packaging and other processing industries, a control system of plane machining machine tool based on machine vision was proposed to realize the rapid detection of paper drawings or model samples. The master-slave motion control system was built by using "ARM+DSP" mode, and the functional modules of each part of the system were designed. A visual detection system of "industrial computer+industrial CCD (charge coupled device) camera+light source control" was built. An improved FAWS (feature adaptive wavelet shrinkage) algorithm was proposed by combining the FAWS algorithm and the sparrow search algorithm to reduce image noise, and the Canny algorithm was used to complete achieve image edge detection so as to achieve the accurate extraction of image contour. The processing programs of image contour extraction, contour data conversion to processing data, data communication, etc. were designed to realize rapid detection based on machine vision and human-computer interaction in the system processing process. Finally, the system was tested and the actual machining effect was evaluated. The results showed that the proposed control system of plane machining machine tool could not only significantly improve the production efficiency, but also reduce the error of image contour. Its performance is stable and reliable, and has certain engineering practical value.