SMT electronic component Tthree-dimensional reconstruction method for SMT electronic component based on deep learning

Abstract

The invention provides a three-dimensional data reconstruction method based on deep learning after three-dimensional optical measurement of an SMT surface mounted electronic component three-dimensional optical measurement. The three-dimensional data reconstruction method comprises the following steps of step that 1, restoring the surface of a measured object is restored through a phase displacement method; step 2, constructing a neural network, and obtaining a multi-angle 2D image of the detected object through multi-angle shooting to restore a 3D model of the detected object; step 3, synthesizing the final most perfect three-dimensional contour by using the two types of 3D model data obtained in the step 1 and the step 2; and step 4, performing cross-parallel ratio calculation on the model data pushed by the traditional method and the neural network, and when the score is relatively low, updating and training the neural network in the step 2 by using the data in the step 1.

Description

technical field [0001] The invention relates to the field of optical detection of SMT surface mounting, in particular to a method for three-dimensional reconstruction of SMT electronic components based on deep learning. Background technique [0002] SMT, or surface mount technology, is the abbreviation of a series of process processes processed on the basis of PCB. It is the most popular technology and process in the electronic assembly industry. Almost every electronic device we commonly see in life, such as mobile phones, automobiles, etc. Computers, home appliances, electronic communications, etc. will use the circuit system produced by the SMT process. These systems are designed by electronic components such as PCBs, capacitors, resistors, and chips according to the designed circuit diagram. The SMT process has been following the electronics since its birth. The development of technology continues to increase the difficulty of the process. The SMT process requires high r...

AI Technical Summary

Problems solved by technology

[0004] The present invention aims at the technical problem that the three-dimensional reconstruction of the electronic components on the PCB in the existing SMT is uneven in brightness caused by the light source, the color and material of the substrate, and the surface characteristics of the components themselves, which leads to the technical problems of thorns and collapses in the three-dimensional reconstruction data. A method for 3D reconstruction of SMT electronic components based on deep learning. An end-to-end neural network outputs its standard 3D model by training a large number of multi-angle 2D images of electronic components under test. Combine height data with 3D model data obtained through deep neural network to obtain more realistic and delicate 3D contour reconstruction

Images