Author(s):
Divya Bharati, Sunil Kumar Pandey
Email(s):
8divya2000@gmail.com , sunilpnd73@gmail.com
Address:
Divya Bharati1, Sunil Kumar Pandey2
1Assistant Professor, MATS School of Information Technology, MATS University, Arang(Raipur),Chhattisgarh
2Assistant Professor and Head of Department Computer Science,
Century Cement College, Baikunth, Pt. Ravishankar Shukla University, Raipur (Chhattisgarh)
Published In:
Book, Multimodal Artificial Intelligence for Intelligent Quality Assessment
Year of Publication:
August, 2026
Online since:
August 12, 2026
DOI:
10.52711/book.anv.9788199786417-04 
ABSTRACT:
This chapter focuses on the importance of Explainable Artificial Intelligence (XAI) in developing trustworthy multimodal quality-assessment systems. It examines how AI models can combine images, sensor measurements, environmental information, text, and spectral data while providing understandable evidence for their predictions. The chapter discusses explainability at different levels and techniques such as Grad-CAM, SHAP, LIME, attention visualization, feature importance, and counterfactual analysis. It highlights how explanations can identify influential image regions, sensor measurements, environmental conditions, and modality contributions. Applications in agriculture, food processing, manufacturing, and healthcare are discussed, along with challenges such as explanation complexity, modality reliability, misleading explanations, and domain-specific interpretation.
Cite this article:
Divya Bharati, Sunil Kumar Pandey. Explainable Multimodal AI for Trustworthy Quality Assessment. Multimodal Artificial Intelligence for Intelligent Quality Assessment. 2026; 1(1): 23-31. DOI: 10.52711/book.anv.9788199786417-04DOI: https://doi.org/10.52711/book.anv.9788199786417-04
References not available.