Recognition And Sorting Of Workpiece Based On Pattern Recognition And Machine Vision

Along with the development of China’s manufacturing industry rapidly,the factory making the enterprise can do the workpieces variety, different specifications, the workpiece structure become more and more complicated, the demand of artifacts also have the corresponding enhancement, the general production field of artifacts records and management are mostly by manual record is complete, not only picking efficiency is low, prone to error or omission and the intensity of labor is very big, which also caused the information transmission speed is slow, difficult to realize real-time analysis and information sharing, more difficult to achieve the real time detection to the production status, so the cost of high quality, high efficiency, low to complete the classification of the workpiece picking became the industry must solve the problem of 4.0 in industry, because of simple structure, low price, environmental protection, easy maintenance and a series of advantages, scope of the application of pneumatic manipulator in industrial production is more and more widely, has become a indispensable important technique in industrial automation industry 4.0 time, one of the people’s attention.Artifacts in industrial production line machine vision as one of the important means of information, not only can improve the accuracy of the pneumatic manipulator pick, but also can increase its own ability to recognize. And machine vision problem to be solved in application to pneumatic manipulator are (1) to extract information acquisition, (2) quickly and accurately extract the characteristic parameters of the workpiece in the production line (3) Judge the work piece located in the production line and identify types of work pieces. In this paper, based on the analysis of machine vision in industry at home or abroad on the basis of the status of the application, aiming at the existing problems and limitations of the research work, the recognition of the workpiece on the conveyor belt and grasping the key technology research. This paper research content is as follows:1. Light by commissioning, industrial camera collection for workpiece images on the conveyor belt, and work to get ready for the next step of image processing.2. The industrial camera collected image preprocessing, evaluate the image noise and reduce the image noise, image sharpening, and to enhance image edge gradually smooth, through set manually or automatically select different threshold was carried out on the collected images segmentation, morphology processing for subsequent pattern recognition to reduce the difficulty of workpiece.3. According to the workpiece under the background of the conveyor belt, design the related algorithm of target recognition and grab workpiece, with Halcon successfully simulated experiment, analyzes the features of each pattern recognition, and provide theoretical basis for subsequent workpiece pattern recognition.4. The design of the pneumatic manipulator control system, including the design of the hardware design, software and hardware communication program, the software design. Hardware design includes the following several parts:the upper machine and lower machine and PC control structure design of the hardware selection and circuit design; Specific provisions of software design using c # language to realize the control program, finally realizes the PC real-time control of pneumatic manipulator movement; Communication system structure of hardware and software design includes vs2010 based on c # software realization man-machine exchange interface and serial communication between upper machine under a machine, and confirms the feasibility of the scheme in the platform of building, in addition, also found the workpiece identification and grab some design problems of the pneumatic manipulator, for the future further research and improvement of the hardware of the theory of optimization to provide the basis.