(For USM Staff/Student Only)

EngLib USM > Ω School of Electrical & Electronic Engineering >

Defect and components recognition in printed circuit boards using convolution neural network

Defect and components recognition in printed circuit boards using convolution neural network / Cheong Leong Kean
Pertumbuhan peranti elektronik meningkatkan permintaan pengeluaran papan litar bercetak, “printed circuit coards” (PCB) dalam industri elektronik. Ini menjurus kepada kenaikan kuantiti pengeluaran PCB setiap hari. Oleh itu, pemeriksaan visual automatik menjadi sistem penting untuk dilengkapkan di mana-mana barisan pengeluaran supaya kualiti PCB dihasilkan dapat dipastikan. Matlamat projek ini, ialah membina sistem pengecaman kecacatan dan komponen automatik untuk PCB menggunakan rangkaian neural konvolusi, “convolution neural network” (CNN). Skop projek ini adalah untuk melaksanakan pengesanan dan kecacatan pengesanan komponen PCB. Pada peringkat pertama, model CNN yang sesuai akan dibina untuk membezakan komponenkomponen elektrik di atas papan litar bercetak. Untuk menjimatkan masa, pemindahan pembelajaran dengan model yang sudah terlatih seperti VGG16, DenseNet169 dan InceptionV3 telah dilakukan untuk mengkaji model yang sesuai untuk pengiktirafan komponen. Menggunakan pembelajaran pemindahan dengan VGG-16, hasil terbaik dicapai adalah ketepatan 99% dengan kemampuan untuk mengiktiraf 25 komponen yang berbeza. Selepas itu, penyetempatan objek dilakukan dengan menggunakan rangkaian neural convolutional berasaskan rantau “region based convolution neural network” (RCNN). Pelbagai eksperimen telah dilakukan untuk menentukan kaedah dan parameter latihan yang optimum untuk mencapai sistem yang dapat mengesahkan kecacatan pada PCB dengan ketepatan yang tinggi. “Mean Average Precision” (mAP) terbaik yang dicapai untuk sistem pennempatan kecacatan ialah 96.54%. _______________________________________________________________________________________________________ The growth of electronic devices increases the demands of printed circuit boards (PCB) productions in the electronic industries. This leads to the rise in the quantity of PCB productions every day. Consequently, automated visual inspection becomes an essential system to be equipped in any production line to ensure the quality of the PCB produced which brings us to the aim of this project, building an automated components recognition system for PCB using CNN. In addition to that, localization on the defects of the PCB components will also be performed. In the first stage, a simple CNN-based component recognition classifier will be developed. Since training a CNN from scratch is expensive, transfer learning with ImageNet pre-trained models is performed instead. Pre-trained models such as VGG16, DenseNet169 and InceptionV3 are used to investigate which model suits the best for components recognition. Using transfer learning with VGG-16, the best result achieved is 99% accuracy with the capability of recognizing up to 25 different components. Following that, object localization is performed using faster region-based convolutional neural network (R-CNN). Multiple experiments have been performed to determine the optimum method and training parameters to achieve a system that is able to localize defects on the PCB with high accuracy and precision. The best mean average precision (mAP) achieved for the defects localization system is 96.54%.
Contributor(s):
Cheong Leong Kean - Author
Primary Item Type:
Final Year Project
Identifiers:
Accession Number : 875007657
Language:
English
Subject Keywords:
electronic devices; printed circuit boards (PCB); productions
First presented to the public:
6/1/2018
Original Publication Date:
8/7/2018
Previously Published By:
Universiti Sains Malaysia
Place Of Publication:
School of Electrical & Electronic Engineering
Citation:
Extents:
Number of Pages - 92
License Grantor / Date Granted:
  / ( View License )
Date Deposited
2018-08-07 16:55:44.555
Date Last Updated
2019-01-07 11:24:32.9118
Submitter:
Mohd Jasnizam Mohd Salleh

All Versions

Thumbnail Name Version Created Date
Defect and components recognition in printed circuit boards using convolution neural network1 2018-08-07 16:55:44.555