Design Of Jujube Detection And Classification System Based On FPGA

In recent years,with the continuous development of the semiconductor industry,FPGA technology has been rapid development,at the same time,the machine vision industry in China is the rapid development in the field of industrial production,medical research have been widely used.Machine vision is include optical technology,mechanical engineering,semiconductor technology,image processing,computer hardware and software technology and so on,the characteristics of the machine vision system with high efficiency and high automation is decided by the machine vision and development in modern society is more and more important role.In this paper,we use the machine vision system to develop a jujube detection and classification system based on FPGA.The whole system includes optical,mechanical,image acquisition,data processing and transmission,FPGA classification algorithm,etc..The main contents are as follows:(1)the design of a high speed color linear CCD image acquisition system,using Kodak's KLI2113 color line array CCD image sensor,image resolution of up to 2098,the highest frequency can reach 8500fps.The image acquisition system has been studied and completed the linear array CCD sensor,professional ADC driver,gain control,as well as the acquisition of the CCD analog signal digital acquisition.(2)the collected image data is filtered to get low noise image data,and the image quality is improved.At the same time,the Bayer data compression of the image data is processed and the data is transmitted through the USB2.0 communication interface,the color image is restored in the upper computer,and the moving image can be displayed in real time.(3)a classification algorithm for the size and edge detection of jujube is proposed.The collected image data were processed by two value processing,dilation and gray value extraction.The detection and classification of red dates were realized.