Research On The Basic Parameters Of Cnc Mills Vision Measurement Method

The tool during machining of the workpiece, the wear arising directly affect the machini ng accuracy, so the real-time or periodically measured parameter of the tool is particularly im portant. Today tool detection equipment can not meet the growing requirements of industrial pr oduction, primarily for conventional measurement methods can not keep up with the demand for the development of a long measurement time, the workload is not suitable for measuring t he disadvantages in production, the development of industrial adaptation demand is particular ly tool testing equipment needs. Tool geometry parameters can overcome the subjective vision measurement error due to the large, measuring less parameters, time and other shortcomings, adapted to the high-speed, high-precision CNC technology modern, efficient requirements.Visual measurement method of this research to CNC mills, for example, also be applied to other CNC tool in the tool to overcome the error in the detection of geometric parameters, relying on manually adjust the lens produced. In this paper, Matlab image processing software for CCD imaging technology resulting cutter photos for image processing, the naked eye observation into a computer vision testing. Then treated tool for image edge detection, boundary tracking, feature point detection tool to get a clear image features. Angle tooling to introduce a test is required cutter geometry, based on the obtained image feature extraction and calculation of its various parameters.CNC tool in order to achieve the correct image segmentation, to lay the foundation for the subsequent parameter measurement tool, the paper has been divided on the basis of the method proposed thresholding method based on wavelet theory. The proposed approach to the target image wavelet decomposition, first estimate segmentation threshold to narrow the search in the large size, and then downscaling, small correction in the calculation results obtained optimal threshold wavelet domain. Opposition cutter images experimental results show that the new algorithm segmentation result is correct, adaptability and robustness has improved significantly. Proposed algorithm can be used for adaptive segmentation CNC tool provides an ideal segmentation results for subsequent processing.In addition, the tool geometry detection, this paper presents a global reference angle establish local coordinate system to the local coordinate system with the combination of image coordinate measuring reference, the detection method in the past due to human error rotary tool positions and observation points can be generated by different avoided, effectively improve the measurement accuracy.