Entropy-Centric Explainable AI for Remote Sensing Image Segmentation

Source

Source name: arXiv
Source kind: paper
Authors / organization: Ali Saleh, Abdul Karim Gizzini, Mohamad Ghassany, Ali J. Ghandour
Venue: arXiv: cs.CV
Published: 2026-08-11
Collected: 2026-08-12
URL: http://arxiv.org/abs/2608.11064v1
PDF: https://arxiv.org/pdf/2608.11064v1
DOI: UNKNOWN
Registry group: academic_apis_and_preprints
Evidence role: canonical_academic_index
Language: en
Country / region: UNKNOWN

Plain-Language Brief

The provided text discusses a new Explainable AI (XAI) method for semantic image segmentation in remote sensing, which aims to increase transparency and trust in AI models. The proposed method is based on entropy and outperforms other XAI methods according to experimental results.

What the Document Says

AI has become powerful in critical domains but concerns arise about its models' decision-making process, especially for deep neural networks.
In remote sensing, the lack of transparency limits trust and adoption of these models due to their black-box nature.
Explainable AI (XAI) is proposed as a solution to bridge this gap by providing insights into AI models' decision-making process.
Significant progress has been made in explaining image classification tasks, but image segmentation still offers room for improvement.
The paper proposes an entropy-centric XAI method for semantic segmentation.
A new XAI evaluation methodology is proposed to measure the relevance of highlighted regions by the XAI method.
Experimental results show the superiority of the proposed XAI method compared with other recent XAI methods for semantic segmentation.

Key Concepts Mentioned

Artificial Intelligence
Explainable AI
XAI
Remote Sensing
Semantic Segmentation
Image Classification

Claims Made by the Document

Deep neural networks outperform their peers but lack transparency in feature extraction and prediction.
In remote sensing, the lack of transparency limits trust and adoption of AI models.
Explainable AI (XAI) aims to provide insights into how and why certain decisions are made by AI models.
The proposed entropy-centric XAI method for semantic segmentation outperforms other recent XAI methods for this task.

Evidence Used

The text mentions the concerns about deep neural networks' ambiguity in feature extraction and prediction.
The text states that in remote sensing, high-resolution imagery needs to be analyzed using black-box models, limiting trust and adoption.
The text explains that XAI aims to provide insights into AI models' decision-making process.
The text mentions that experimental results demonstrate the superiority of the proposed XAI method.

Methods / Approach

UNKNOWN

Limitations Stated by the Authors

UNKNOWN

Limitations Noted by the Agent

Agent interpretation — not from the source.

UNKNOWN

Open Questions

UNKNOWN

Notes for Future Q&A

May include interpretation. Labeled where it does.

UNKNOWN

Raw Abstract / Excerpt

Artificial intelligence (AI) has become a powerful approach to solving complex problems in critical domains. Many concerns arise regarding the decision-making process of its models, mainly due to deep neural networks outperforming their peers at the cost of ambiguity in feature extraction and prediction. Consequently, in critical domains such as remote sensing, where high-resolution imagery must be analyzed using black-box models, the lack of transparency limits trust in these models and, thus, their adoption. In light of this reality, explaining and understanding the complex decision-making process of AI models has become essential. Explainable AI (XAI) aims to bridge this gap by providing insights into how and why certain decisions are made. While significant progress has been achieved in explaining image classification tasks, image segmentation still offers considerable room for improvement. In this context, this paper proposes an entropy-centric XAI method for semantic segmentation. Moreover, a new XAI evaluation methodology is proposed to efficiently measure the relevance of the regions highlighted by the proposed XAI method. Experimental results demonstrate the superiority of the proposed XAI method compared with recently adapted XAI methods for semantic segmentation.