Design Of Sanitary Ware Automatic Packaging System Based On Machine Vision

With the continuous progress and development of science and technology,industrial robots have been successfully applied in aviation manufacturing,automobile manufacturing,food packaging and electronic packaging industries,covering a variety of working methods including spraying,palletizing,cutting,welding,sorting,packaging,etc.Industrial robots have become an important tool to improve work efficiency,reduce production costs,and effectively control product quality.With the improvement of production informationization level,machine vision technology has been closely integrated with industrial robots,and has been applied in product packaging.In recent years,the demand for robots has become more and more vigorous in the intelligent upgrading of enterprises.This trend is particularly prominent in the packaging industry.Due to the complexity of sanitary ware packaging in the packaging industry,a sanitary ware automatic packaging system based on machine vision can be constructed with the help of robot technology and visual technology.The main work of this system is as follows:(1)By understanding the current situation of machine vision and automated packaging at home and abroad,taking the automated packaging of right-angle valves as the research object,the existing manual packaging process was analyzed,the structure of the automatic packaging system for sanitary ware was established,and the light source,camera and robot were selected according to the actual needs.(2)By studying the imaging model and distortion model mechanism of the camera and calibrating the camera,the main parameters of the camera are obtained.Then the camera-conveyor belt-robot system is calibrated,and the mathematical model from the camera coordinate system to the robot coordinate system is obtained.The forward and inverse kinematics of IRB 910 SC is analyzed by D-H parameter method.(3)In the image processing part,bilateral filtering is selected as the image smoothing scheme by comparison.The filling flooding is used to separate the sanitary ware from the background.Then the image is processed by open operation,binarization,ROI determination,affine transformation and image normalization,and finally image recognition scheme based on Haar wavelet fast template matching is adopted.This paper innovatively uses the standard template of right angle valve as the characteristic figure,establishes the position and posture label of the standard template,and derives the corresponding relationship between the standard template and the sanitary ware in the image.With this relationship,the position and posture label of the standard template is geometrically transformed to obtain the position and posture of the sanitary ware in the picture.(4)Robotics Toolbox of MATLAB is used to build kinematics model of robot IRB 910 SC.The coordinates of robot's terminal coordinates are obtained by known coordinates of robot's coordinates.The kinematics parameters and trajectory planning of robot are obtained by knowing the position of robot's terminal actuator.Robot Studio's Smart component is used to set up the logic of the system components and build the simulation model of the robot system.Then the program is compiled and the simulation results are obtained.The correctness of the system operation logic and device selection is demonstrated.