Particle Image Velocimetry Algorithms Based On Fpga

Digital particle image velocimetry (DPIV) technique is widely used in the measurement ofthe complex and unsteady flow field. However, the traditional image processing system, which isbased on the computer processing can not meet the requirement of real-time measurement. Theapplication of DPIV technology is restricted.The image processing algorithm of DPIV is researched in this thesis, and the directcross-correlation algorithm is proposed, which is implemented in the Field Programmable GateArray (FPGA).The real-time processing of flow field is realized after the parallel computing.Firstly, the research status and history of the DPIV development is introduced. Then thesystem constructer, operating principle and characteristics of DPIV are represented. The selectionof diagnosis windows and hardware algorithms is discussed.Secondly, based on FPGA, the algorithms of image matching, direct cross-correlation, peaksearching, are designed. Then the data is sent to the host computer by the USB bus aftersubsequent processing.At last, in order to verify the feasibility of the system, Quartus II software is used. Thesimulated result shows that the value and address of cross-correlation peak can be accuratelyacquired. Then the program is downloaded to Cyclone II device for hardware test, and theprocessing data can be read in the embedded SignalTaP II logic analyzer. The experimentalresults show that the hardware algorithm is correct, the system design is feasible, and theprocessing speed is improved.The DPIV algorithm that based on FPGA is realized in this thesis. The feasible of real-timemeasurement of DPIV is verified, and it provides a new idea for DPIV researching.