Research On Key Technology Of Cake Pattern Spraying System Based On Machine Vision

With the emergence of "Made in China 2025" strategy,intelligent manufacturing has become the general trend in the development process of China’s manufacturing industry.Collocation of machine vision system production equipment,greatly improve the level of intelligence,can meet higher production and manufacturing needs.Therefore,the automatic production and processing technology equipped with machine vision system is a major research direction in the field of intelligent manufacturing.In this paper,the main research work is the pattern spraying of cake workpiece in the food manufacturing industry.Machine vision system,camera calibration technology,image processing algorithm and improved path planning algorithm are used to complete the identification,location,trajectory extraction and path planning of the target workpiece.Set up the software and hardware system of cake pattern spraying processing,and realize the function of automatically spraying Chinese characters and other patterns on the cake workpiece.The main work of this paper is as follows:Aiming at the camera calibration problem in the vision system,the imaging model and distortion model of the camera are established,and the transformation relationship of the important reference coordinate system in the model is solved.By making calibration plates and collecting calibration images,the single and binocular calibration is realized,and the accurate internal and external parameters of the camera are obtained.The hand-eye calibration experiment was carried out to calculate the hand-eye transformation matrix between the camera coordinate system and the manipulator coordinate system in the actual scene.Aiming at the problem of noise interference in the process of target image acquisition,the effectiveness of mean filtering algorithm to remove noise is determined through filtering experiments.The Sobel operator is used to extract the image edge,establish the shape matching template,and identify and locate the target workpiece,which solves the problem of target identification and location in the scene of uneven illumination,and ensures the accuracy,real-time and accuracy of the identification and location.In the process of 3D positioning,the image is collected by binocular camera and the parallax of the target object is obtained by stereo matching algorithm,and the 3D coordinate of the target is calculated.Experimental results show that the positioning accuracy of the system is within 1 mm,which basically meets the requirements of actual production and processing.In this paper,the trace extraction and path planning are analyzed for the important and difficult Chinese characters pattern spraying processing of target workpiece pattern.In the part of trace extraction of Chinese strokes,three schemes of character base extraction,skeleton extraction and XLD and DXF extraction are designed and implemented.At the same time,it also realizes the processing of other line patterns and filling patterns.In the processing path planning part of Chinese character pattern,TSP model was established by analyzing the characteristics of Chinese character strokes,and the ant colony algorithm was used to solve the optimal spraying path of Chinese character pattern.By improving the pheromone updating strategy and probability selection function of ant colony algorithm,the global optimal solution is obtained and the iteration time of the algorithm is obviously reduced.Finally,this paper completes the software development and hardware design and debugging of the whole system,and realizes the above functions.Through the test of the system,it is shown that the system can complete the identification and location of the cake workpiece and the processing path planning according to the text pattern provided,and control the movement of the manipulator to achieve the function of automatic processing pattern.