Jujube-based Online Classified Computer Vision Systems

Machine to replace human labor is the human in the long-term social production practice the pursuit of the goal. Among them, the machine vision is applied computer vision has been widely development. It is to use machines to replace human eye for target identification, judgment, measurement, the main research using computer to simulate the visual function. Computer vision technology involves target image acquisition technology, the image information processing technology and the target object measurement and identification technology.Computer vision system is mainly composed of visual perception unit, image information processing and identification unit, results show that unit and visual system control unit. In the industrial production process, compared with the traditional detection method of measurement, the computer vision technology the biggest advantage is rapid, accurate, reliable and intelligent, to improve the consistency of the product inspection, production safety, reduce labor intensity and realize the effective safety production and automation management plays an irreplaceable role.The subject, according to the current status quo of domestic and foreign fruit quality detection research, learn from foreign advanced technology and experience, the use of computer vision technology, image information processing Xinjiang jujube of size, color, defects, and then through man-machine interface to control on-site, intelligent automatic grading purposes.This paper studies content:(1) Analysis of the current situation at home and abroad of fruit automated detection, Xinjiang jujube and discuss the feasibility and necessity of computer vision detection;(2) Research on intelligent classification the hardware platform and working principle, determine the overall scheme, the main components (light source, camera, computer, DSP) selection, and set up the system hardware platform;(3) To study computer vision system for automatic classification of software foundation, including image acquisition, image processing (including threshold segmentation, median filtering, etc.), and applies it to the big jujube jujube grade, size, shape, color, defects analysis, using VC to write the corresponding processing program;(4) The development of the grading system for the PC program, according to the need for on-site production, the use of the modular design concept, the system function module division, mainly user login configuration parameter processing operation, fault alarm diagnosis designed to meet the requirements of the man-machine interface of the host computer.(5) Analysis algorithm by the experimental data.(6) Analysis of the papers could be improved and some of the issues for further research and innovation of this paper.