The Research On Rice Quality Dynamic Detection Algorithm

Most the measuring of appearance quality of rice in our country still relies on manual work sortion,which lacks efficiency and objectivity.So it extends the period of paddy classification and increases complication.In response to this issue,rice quality dynamic detection equipment is designed,which is based on Xilinx's Sparten3E series XC3S500E FPGA chips and high-speed linear array CCD sensors. The function of detection equipment scans the falling rice in high speed by CCD sensor,then extracts features of modal parameters from collected rice images,finally achieves to sort qualified and unqualified rice and calculate rice qualified indexs such as the head rice milling rate.So that it makes feasible to dynamic detecting the quality of rice.The hardware design of detection system includes the core circuit of FPGA; drive circuit of CCD; the interface circuit of digitizer.The research on detection system is divided into two main parts.One part is researches on speedy imaging theories and algorithms during the process of rice dynamic detection,which includes image filtering algorithms (such as median filtering algorithm and wavelet denoising algorthm), the basic image processing algorithms(such as edge detection algorithm); The other part is image parameters extraction algorithm (7 parameters mentioned in the paper) and the head rice milling rate calculation algorithm. Center-edge rotation searching algorithm is mentioned in diameter of rice detection, which reduces complication of classic algorithm.The results of classification can be obtained in 5ms. Algorithms realizes on FPGA chips. The hardware description is realized in Verilog and VHDL language. Using the platform of ISE8.2, Synplify Pro7.2 software has completed the simulation, synthesis, implementation and configuration of every module. It is embedded IP cores by ChipScope Pro8.2 to download the program and accomplish on-line debugging.After conducting experiments and debugging on-line, we have completed the driver of the CCD, rice dynamic image acquisition, image parameters of the extraction and head milling rice rate calculation.The detection accuracy rate is over 90%.