Research And Realization Of Polar Electronics Identification Towards Printed Circuit Diagram

Before mounting the printed circuit board,a special quality inspector needs to check the electronic devices in the corresponding printed circuit diagram one by one.The content of the check is mainly the electronic device type,size,and orientation.Polar electronics type.The manual verification method is more reliable for PCBs that use fewer electronic devices.However,with the advancement of related technologies,the number of components that can be mounted on printed circuit boards has increased exponentially.In the past,manual verification methods not only required Consume a lot of manpower and material resources,and the efficiency is extremely low.In order to improve the detection efficiency and save the production cost of the enterprise,this paper proposes a method for identifying polar electronic devices in printed circuit diagrams based on machine learning.For identification of polar electronics.This article first creates a polar electronic device image library to solve the problem of lack of data sets in the research;improves the directional gradient histogram algorithm and proposes the HOG-Gabort fusion feature,in which the HOG algorithm mainly uses the contours of polar electronic devices.The Gabor algorithm is mainly used to characterize the texture distribution of polar electronic devices at different scales and directions.The HOG-Gabor fusion feature can effectively make up for the shortcomings of HOG in the image rotation or scaling changes,and improve the smaller or larger individuals and The detection rate of the polar electronic devices that have undergone rotation is reduced by using the principal component analysis method to reduce the characteristics,reduce time consumption,and improve real-time performance.Support vector machine is used as a classification algorithm,in which the kernel function uses a radial basis kernel and relaxation variables are set to prevent the optimal hyperplane from moving due to individual outliers,thereby obtaining a larger geometric interval.A penalty factor C is introduced.A value makes the classifier more generalizable.In order to improve the classification effect of the classifier,the genetic algorithm and K-fold cross-validation are used to optimize the parameters to obtain the best C value and the parameters of the kernel function.Finally,the prepared test data is used to verify the performance of the obtained model.The experimental results show that the identification method proposed in this paper can effectively solve the identification problem of polar electronic devices.Next,using the obtained model as the underlying classification algorithm and Vue and Django as the front-end and back-end,an online identification system for polar electronic devices was built.This system can help quality inspectors to quickly calculate the information of polar electronics and improve placement efficiency.