Research On Visual Inspection System For Assembly Of Jacquard Machine Components Based On OpenCV

Jacquard machine components are the main assembly parts of the jacquard machine,which consists of several parts,and the assembly process is complex,so the quality inspection of the assembled jacquard machine component parts is an important part.At present,the assembly inspection of jacquard machine components is mainly carried out manually,which is difficult to ensure the accuracy and efficiency of the final inspection due to its own limitations.Therefore,based on the existing assembly line of the jacquard machine components in the factory,this paper designs and develops an assembly vision inspection system based on the existing assembly line of the jacquard machine components,applies machine vision technology,and uses OpenCV image processing open source library to design and develop.The main research contents of this paper are as follows:1.The assembly process of jacquard components was studied,and then the various problems that can occur during the assembly of jacquard machine components were summarized,and the inspection requirements were clarified.Based on the characteristics of the jacquard component parts to be inspected,and then combined with the existing assembly line,the entire inspection platform is designed.2.Based on the camera imaging principle and coordinate system conversion,this chapter investigates the monocular vision calibration and establishes the imaging model of the monocular vision system to determine its internal and external parameters through the calibration of the monocular camera.Several image processing algorithms are also investigated,including filtering and denoising,thresholding segmentation,etc.,in which two common noises are added to the image and different filtering methods are used for comparison,and then the Otsu method segmentation is found to be slightly better than other methods.3.Based on the traditional edge feature extraction algorithm,this paper proposes an optimal enhancement method based on the Canny edge operator to realize the image acquisition of jacquard machine components,and then for the special object to be measured,this paper designs a template matching method based on the target profile,by dividing the ROI region(region of interest)of the qualified component,the improved edge extraction algorithm is used to complete the extraction of the profile information of the DUT in the area and make it into a template,and then traverse the entire image to be measured through the template.The pixel coordinates of the feature points of the test piece are determined,so as to realize the detection of the assembly quality of the part to be tested.4.Based on the C++ language,QT library and OpenCV image open source processing and based on the VS2019 integrated environment,various image processing algorithms are performed throughout,and the human-computer interaction interfaces are written,and then the assembly inspection platform is built,and the effectiveness and detection accuracy of the assembly inspection system in this paper are verified by experiments.5.Through experiments,it is verified that the visual inspection system of the jacquard machine component assembly designed in this paper,the developed electronic jacquard component platform system is qualified in accordance with the development-related regulations of the enterprise.