| In recent years,many industries such as aerospace,automotive production,mold processing,and machine manufacturing have greatly increased the demand for integrated visual inspection of CNC(Computer Numerical Control)machining.Only by integrating the online detection technology and the online detection system into the CNC working cycle to form a CNC integrated process can we further standardize and deepen the application and popularization of CNC online detection technology,and lay the foundation for the integrated perception of high-end CNC equipment and intelligent CNC equipment.The CCD on-line inspection of CNC-processed workpieces,research and development of CNC work cycle integrated vision probe detection technology,and proposed a CNC machining technology system that integrates independently developed CNC program control and can automatically exchange vision probes,taking into account the accuracy and real-time of online inspection requirements,based on the CNC embedded device workpiece online detection process method,to achieve automatic online acquisition of workpiece image,size and error processing,data transmission and storage functions in the CNC work cycle.The specific research contents of this article are as follows:1.Aiming at the detection problems of typical CNC machining process systems,an integrated vision inspection process for CNC machining centers was proposed.Through the analysis of the complexity of the process,the specific requirements of the process system were determined,and CNC processing technology system was constructed based on the independently developed CCD vision probe.The research and development process of CCD vision probe,the composition of the process system,the detection mechanism and the implementation scheme are introduced.2.For the process of online visual inspection of different types of part sizes,the visual inspection method and process integration implementation technology were studied.Aiming at the problem that the single corner detection algorithm is inaccurate in the linear dimension visual feature extraction,FAST + SUSAN corner detection method is proposed.The FAST algorithm is used for rough corner extraction,and the SUSAN algorithm is used for fine extraction.Combining algorithms to improve detection accuracy;Aiming at the problem of low detection efficiency of traditional Hough transform algorithm applied to the extraction of visual features of arc size,an Hough transform optimization algorithm was proposed,and the algorithm execution speed was improved by optimizing algorithm parameters to make it better Meet real-time needs.The circuit board of the embedded system in the CCD vision probe adopts DSP + FPGA architecture.The CNC system program and the embedded processing system control together to realize the process integration of the visual inspection method of the workpiece size.3.Two typical workpieces were selected for image experiments,and the measurement accuracy and real-time performance of the two algorithms proposed in this paper were tested respectively.The experimental results show that the FAST + SUSAN algorithm proposed in this paper is more accurate than the single method in removing a large number of false corners;the feature extraction time of the Hough transform optimization algorithm using the optimal parameters is reduced from the original 7.046 4 s to 3.723 2 s.And the detection accuracy can reach 0.02 mm.Using the technology system proposed in this paper,NC machining of complex workpieces and in-sequence online detection experiments of key dimensions are performed.The experimental results show that the process has high detection efficiency,which ensures high-precision machining of CNC machine tools,and fully meets the needs of automated online detection. |
PDF Full Text Request