Research On Intelligent Image Acquisition And Processing System Of Tea Color Sort

The tea color sorter is a kind of high-tech equipment used for separating the leaf and the stalk in various kinds of tea, which contains the newest photovoltaic technology and machine vision; it plays a very important role in improving separating productivity, accuracy and quality. Due to the combined influence of market demand, technological development, the craft of tea color sorter become better and more intelligent. An intelligent image acquisition and processing system against the current is designed in this paper, the system can improve the accuracy and reliability significantly.The image acquisition and processing system can be divided into three parts: Acquisition system, processing system and PC host. In this paper, we analyzed the operating principles and time sequences of CCD and ADC, and designed a3-line color linear CCD image acquisition system; in the processing part, the scheme of FPGA+DDR3+USB is used, which the FPGA is used for controlling and processing, DDR3is used for storing the large image data, USB is used for communicating with the host. In order to achieve satisfying sort result, a new algorithm including image segmentation, color and shape feature extraction and Bayes classifier is present in this paper, which takes both color and shape characters in consideration rather than considering color only. In the last, we analyzed the needs of custom in host software, and develop the software modularization.The image acquisition and processing system designed in this paper had been applied into prototype tea color sorter successfully. The system is efficient and meets the demand.