Performance Analysis of Deep Learning Models for Classification of Surface-Mount Device (SMD) Components

Performance Analysis of Deep Learning Models for Classification of Surface-Mount Device (SMD) Components
  IJETT-book-cover           
  
© 2024 by IJETT Journal
Volume-72 Issue-10
Year of Publication : 2024
Author : Pratik J Gohel, Brijesh R Solanki, Amit C Rathod, Devang G Jani
DOI : 10.14445/22315381/IJETT-V72I10P123

How to Cite?
Pratik J Gohel, Brijesh R Solanki, Amit C Rathod, Devang G Jani, "Performance Analysis of Deep Learning Models for Classification of Surface-Mount Device (SMD) Components," International Journal of Engineering Trends and Technology, vol. 72, no. 10, pp. 235-245, 2024. Crossref, https://doi.org/10.14445/22315381/IJETT-V72I10P123

Abstract
Accurate classification of Surface-Mount Device (SMD) components is important for a range of electronics manufacturing and assembly applications. Impressive results have been achieved using recent deep learning models when applied to image-classification problems. This study provides a comprehensive examination of the categorization of Surface-Mounted Device (SMD) components using advanced learning models. Four cutting-edge Deep Learning (DL) models–ResNet50, VGG16, AlexNet, and MobileNet–were utilized to categorize SMD components into eight classes: capacitors, diodes, Electrolytic Capacitors (EC), Integrated Circuits (IC), LED, resistors, supercapacitors, and Zener diodes. Our approach encompasses the training of these models on a dataset containing SMD component images and the assessment of their performance in terms of accuracy, precision, recall, and F1-score. The findings indicate that MobileNet achieved the highest classification accuracy, reaching up to 98%, surpassing the other models. Through a comprehensive comparative analysis, we discern the strengths and limitations of each model in this categorization task. Our results suggest that MobileNet is the most effective deep-learning framework for SMD component classification, underscoring its potential applications in automated electronic assembly and quality control processes. This study contributes to the progress of automated electronic component classification and guides future research in selecting suitable deep learning models for similar tasks.

Keywords
Object detection, SMD components, Machine learning, VGG16, ResNet50, Alexnet, MobileNet.

References
[1] SeungGeun Youn, YounAe Lee, and TaeHyung Park, “Automatic Classification of SMD Packages using Neural Network,” IEEE/SICE International Symposium on System Integration, Tokyo, Japan, pp. 790-795, 2014.
[CrossRef] [Google Scholar] [Publisher Link]
[2] D.U. Lim, Y.G. Kim, and T.H. Park, “SMD Classification for Automated Optical Inspection Machine Using Convolution Neural Network,” Third IEEE International Conference on Robotic Computing, Naples, Italy, pp. 395-398, 2019.
[CrossRef] [Google Scholar] [Publisher Link]
[3] Stanisław Hożyń, “Convolutional Neural Networks for Classifying Electronic Components in Industrial Applications,” Energies, vol. 16, no. 2, pp. 1-22, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[4] Emel Soylu, and İbrahim Kaya, “Classification of Electronics Components Using Deep Learning,” Sakarya University Journal of Computer and Information Sciences, vol. 7, no. 1, pp. 36-45, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[5] Longfei Zhou, and Lin Zhang, “A Novel Convolutional Neural Network for Electronic Component Classification with Diverse Backgrounds,” International Journal of Modeling, Simulation, and Scientific Computing, vol. 13, no. 1, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[6] Qiang Zhang et al., “Image Defect Classification of Surface Mount Technology Welding Based on the Improved Resnet Model,” Journal of Engineering Research, vol. 12, no. 2, pp. 154-162, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[7] Van-Truong Nguyen, and Huy-Anh Bui, “A Real-Time Defect Detection in Printed Circuit Boards Applying Deep Learning,” EUREKA: Physics and Engineering, no. 2, pp. 143-153, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[8] Mee Chun Loo et al., “CNN Aided Surface Inspection for SMT Manufacturing,” 15th International Conference on Developments in eSystems Engineering, Baghdad & Anbar, Iraq, pp. 328-332, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[9] Rui Huang et al., “A Rapid Recognition Method for Electronic Components Based on the Improved Yolo-V3 Network,” Electronics, vol. 8, no. 8, pp. 1-18, 2019.
[CrossRef] [Google Scholar] [Publisher Link]
[10] Sifundvolesihle Dlamini, Chung-Feng Jeffrey Kuo, and Shin-Min Chao, “Developing a Surface Mount Technology Defect Detection System for Mounted Devices on Printed Circuit Boards Using a Mobilenetv2 with Feature Pyramid Network,” Engineering Applications of Artificial Intelligence, vol. 121, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[11] Haihong Pan et al., “A New Image Recognition and Classification Method Combining Transfer Learning Algorithm and Mobilenet Model for Welding Defects,” IEEE Access, vol. 8, pp. 119951-119960, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[12] Aqsa Hassan et al., “An Empirical Analysis of Deep Learning Architectures for Vehicle Make and Model Recognition,” IEEE Access, vol. 9, pp. 91487-91499, 2021.
[CrossRef] [Google Scholar] [Publisher Link]
[13] Jason Adam, Resnet-50 API, 2019. [Online]. Available: https://jason-adam.github.io/resnet50/
[14] Aqeel Anwar, Difference between Alexnet, Vggnet, Resnet and Inception, Medium, 2019. [Online]. Available: https://towardsdatascience.com/the-w3h-of-alexnetvggnet-resnet-and-inception-7baaaecccc96
[15] Mohanad A. Deif et al., “Diagnosis of Oral Squamous Cell Carcinoma Using Deep Neural Networks And Binary Particle Swarm Optimization On Histopathological Images: An AIoMT Approach,” Computational Intelligence and Neuroscience, vol. 2022, pp. 1-13, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[16] Young-Gyu Kim et al., “SMD Defect Classification by Convolution Neural Network and PCB Image Transform,” IEEE 3rd International Conference on Computing, Communication and Security, Kathmandu, Nepal, pp. 180-183, 2018.
[CrossRef] [Google Scholar] [Publisher Link]