NORMA eResearch @NCI Library

Identification and Classification of Electrical Components on Printed Circuit Boards Using Transfer Learning

Pattanayak, Pratyush Kumar (2022) Identification and Classification of Electrical Components on Printed Circuit Boards Using Transfer Learning. Masters thesis, Dublin, National College of Ireland.

[thumbnail of Master of Science]
PDF (Master of Science)
Download (487kB) | Preview
[thumbnail of Configuration manual]
PDF (Configuration manual)
Download (237kB) | Preview


Electronic circuit boards are becoming part of most modern-day equipment such as computers, mobile phones, autonomous cars, manufacturing equipment, etc. With the advancement of technology, printed circuit boards (PCBs) are becoming smaller and densely packed with a variety of electronic components. Given their central part in the operation of all digitally enabled equipment it is extremely necessary to inspect them for any manufacturing defect before they are released to the end consumer. It is important to accurately identify and classify the components on the PCBs before they can be inspected for any defect. Manual or mechanical inspection is both time-consuming and inefficient. Existing automated visual inspection methods are trained on datasets that do not capture real-world scenarios such as illumination and scale variation. The aim of this research is to develop a model which can accurately identify the electrical components on a printed circuit board in real-world scenarios. To achieve this, a challenging dataset has been used which contains images of PCB components from three variable aspects. They are illumination, image scale, and image sensor. Five modern transfer learning models have been implemented and their performance is evaluated. The models are InceptionV3, EfficientNetB1, EfficientNetB2, Xception and ResNet152V2. The models are optimized and evaluated on the basis of accuracy, precision, recall and F1-score.

Item Type: Thesis (Masters)
Uncontrolled Keywords: Electrical components; Transfer Learning; Image classification; Printed Circuit Boards
Subjects: Q Science > QA Mathematics > Electronic computers. Computer science
T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science
T Technology > TK Electrical engineering. Electronics. Nuclear engineering
Divisions: School of Computing > Master of Science in Data Analytics
Depositing User: Tamara Malone
Date Deposited: 28 Feb 2023 15:00
Last Modified: 28 Feb 2023 15:00

Actions (login required)

View Item View Item