Research And Implementation Of FPGA-Based Center Positioning And Matching Algorithm Of Reflective Markers On A 3D Scanning Probe

The 3D optical scanning measurement system has been used in the aerospace manufacturing,shipbuilding,and automobile manufacturing in recent years due to the advantages of high measurement accuracy,fast measurement speed,and no contact with the workpiece to be tested.With the development of science and technology,there are a large number of parts with large size and high precision,which puts forward a higher requirement for the measurement accuracy and efficiency of 3D scanning instruments.Aiming at a 3D robotic scanning system,which combined 3D laser scanning with multi-vision tracking system,a multi-view 3D tracking system is investigated in this paper.Taking in consideration the large amount of data and high real-time data processing requirements of the multi-view 3D tracking system,the parallel operation FPGA is selected as the core processor,and the existing image processing algorithms are improved to make them more efficiently on the embedded platform of FPGA.The main research results of this paper were as follows:(1)Based on the existing GSI coding scheme and cross-ratio invariance,a new coding scheme is designed to simplify the grouping and matching processes of the coded markers,and a new improved scanning window is designed,which can reduce the probability of label conflicts and improve the efficiency of the algorithm,based on the traditional two-step connected domain labeling algorithm.(2)The arbitration module is employed to control and manage the read and write requests in the system to realize resource sharing of DDR3 by multiple data processing modules.(3)Based on the parallel operation characteristics of the FPGA processor,image data synchronous acquisition,image filtering,marker recognition and center point positioning and coded marker matching modules of the four-channel camera are implemented in a pipeline way to maximize the advantages of the FPGA processor.The experimental results have shown that the maximum operating frequency of the image processing system based on FPGA is 150 MHz,and the processing speed of image data reaches 35.7 fps.Compared with PC platform,the processing speed has been increased by nearly 10 times with the same image processing algorithm,which can satisfy the real-time data processing requirement of the multi-vision positioning and tracking system.