Comparative Performance of Finetuned ImageNet Pre-trained Models for Electronic Component Classification

2025-06-24

AI Summary

This study addresses the fine-grained classification of electronic components in industrial manufacturing. We systematically evaluate twelve ImageNet-pretrained models—including lightweight CNNs such as MobileNet-V2 and EfficientNet-B0—under identical transfer learning fine-tuning protocols. MobileNet-V2 achieves the highest accuracy (99.95%), substantially outperforming alternatives like EfficientNet-B0 (92.26%), demonstrating superior adaptability and generalization under resource-constrained production-line conditions. Our key contributions are threefold: (1) the first multi-model benchmark on a real-world electronic component dataset; (2) empirical characterization of the trade-offs among model capacity, inference efficiency, and classification accuracy; and (3) evidence-based guidelines for selecting pretrained models in industrial visual inspection systems. The results provide actionable insights for deploying efficient, high-accuracy vision solutions in practical manufacturing environments.

Technology Category

• Computer Vision: Large Vision Models
• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Web Mining and Content Analysis: Large pretrained models with web data
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems

Abstract

Electronic component classification and detection are crucial in manufacturing industries, significantly reducing labor costs and promoting technological and industrial development. Pre-trained models, especially those trained on ImageNet, are highly effective in image classification, allowing researchers to achieve excellent results even with limited data. This paper compares the performance of twelve ImageNet pre-trained models in classifying electronic components. Our findings show that all models tested delivered respectable accuracies. MobileNet-V2 recorded the highest at 99.95%, while EfficientNet-B0 had the lowest at 92.26%. These results underscore the substantial benefits of using ImageNet pre-trained models in image classification tasks and confirm the practical applicability of these methods in the electronics manufacturing sector.

Problem

• Compare performance of ImageNet models for electronic component classification
• Evaluate accuracy of 12 pre-trained models in manufacturing applications
• Assess practical use of ImageNet models in electronics industry

Innovation

• Finetuned ImageNet models for component classification
• MobileNet-V2 achieves highest accuracy 99.95%
• EfficientNet-B0 shows lowest accuracy 92.26%