XAI-Enhanced Semantic Segmentation Models for Visual Quality Inspection

Clement T, Nguyen Truong Thanh H, Abdelaal M, Cao H (2024)

Publication Type: Conference contribution

Publication year: 2024

Journal

Digest of Technical Papers - IEEE International Conference on Consumer Electronics Institute of Electrical and Electronics Engineers

Publisher: Institute of Electrical and Electronics Engineers Inc.

Conference Proceedings Title: Digest of Technical Papers - IEEE International Conference on Consumer Electronics

Event location: Las Vegas, NV US

ISBN: 9798350324136

DOI: 10.1109/ICCE59016.2024.10444225

Abstract

Visual quality inspection systems, crucial in sectors like manufacturing and logistics, employ computer vision and machine learning for precise, rapid defect detection. However, their unexplained nature can hinder trust, error identification, and system improvement. This paper presents a framework to bolster visual quality inspection by using CAM-based explanations to refine semantic segmentation models. Our approach consists of 1) Model Training, 2) XAI-based Model Explanation, 3) XAI Evaluation, and 4) Annotation Augmentation for Model Enhancement, informed by explanations and expert insights. Evaluations show XAI-enhanced models surpass original DeepLabv3-ResNet101 models, especially in intricate object segmentation.

Authors with CRIS profile

Tobias Clement Lehrstuhl für Wirtschaftsinformatik, insbesondere IT-Management Hung Nguyen Truong Thanh Lehrstuhl für Wirtschaftsinformatik, insbesondere IT-Management

Involved external institutions

University of New Brunswick (UNB) CA Canada (CA)

How to cite

APA:

Clement, T., Nguyen Truong Thanh, H., Abdelaal, M., & Cao, H. (2024). XAI-Enhanced Semantic Segmentation Models for Visual Quality Inspection. In Digest of Technical Papers - IEEE International Conference on Consumer Electronics. Las Vegas, NV, US: Institute of Electrical and Electronics Engineers Inc..

MLA:

Clement, Tobias, et al. "XAI-Enhanced Semantic Segmentation Models for Visual Quality Inspection." Proceedings of the 2024 IEEE International Conference on Consumer Electronics, ICCE 2024, Las Vegas, NV Institute of Electrical and Electronics Engineers Inc., 2024.

BibTeX: Download