Feature-Based Transfer Learning for IoT-Enabled Defect Detection for Quality Control in Industrial Manufacturing Processes: A DenseNet Evaluation

Anwar P. P. Abdul Majeed, Muhammad Ateeq*, Bintao Hu, Wan Hasbullah Mohd Isa, Zaid Omar, Wei Chen

*Corresponding author for this work

Research output: Chapter in Book or Report/Conference proceedingConference Proceedingpeer-review

Abstract

Surface defects in industrial products can lead to performance issues or early failure. Automated optical inspection using computer vision and deep learning provides an efficient approach for quality control. This study investigated a feature-based transfer learning technique for classifying surface defects in images. The resampled KolektorSDD dataset containing 50 defect and 50 non-defect images of electrical commutator surfaces was utilised. Pre-trained DenseNet convolutional neural network (CNN) models were used to extract discriminative features from the images. A support vector machine (SVM) classifier was then trained on these features to classify images as either defective or non-defective. Three DenseNet architectures were tested, namely DenseNet121, DenseNet169, and DenseNet201. All three pipelines achieved 100% accuracy on the test and validation sets when used with the SVM classifier. Owing to the size and the overall performance of the pipeline, it was established that the DenseNet169-SVM was suitable for the given dataset. The high accuracy demonstrates that transfer learning can provide an effective approach for classifying surface defects compared to training a deep neural network from scratch. This allows for the development of automated quality control and surface inspection systems powered by computer vision and deep learning.

Original languageEnglish
Title of host publicationAdvances in Intelligent Manufacturing and Robotics - Selected Articles from ICIMR 2023
EditorsAndrew Tan, Fan Zhu, Haochuan Jiang, Kazi Mostafa, Eng Hwa Yap, Leo Chen, Lillian J. A. Olule, Hyun Myung
PublisherSpringer Science and Business Media Deutschland GmbH
Pages443-449
Number of pages7
ISBN (Print)9789819984978
DOIs
Publication statusPublished - 2024
EventInternational Conference on Intelligent Manufacturing and Robotics, ICIMR 2023 - Suzhou, China
Duration: 22 Aug 202323 Aug 2023

Publication series

NameLecture Notes in Networks and Systems
Volume845
ISSN (Print)2367-3370
ISSN (Electronic)2367-3389

Conference

ConferenceInternational Conference on Intelligent Manufacturing and Robotics, ICIMR 2023
Country/TerritoryChina
CitySuzhou
Period22/08/2323/08/23

Keywords

  • Deep learning
  • Feature-based transfer learning
  • Industrial IoT
  • Industry 4.0
  • IoT-enabled defect detection
  • Machine learning
  • Manufacturing processes
  • Smart manufacturing

Fingerprint

Dive into the research topics of 'Feature-Based Transfer Learning for IoT-Enabled Defect Detection for Quality Control in Industrial Manufacturing Processes: A DenseNet Evaluation'. Together they form a unique fingerprint.

Cite this