Matches in SemOpenAlex for { <https://semopenalex.org/work/W4387355985> ?p ?o ?g. }
Showing items 1 to 51 of
51
with 100 items per page.
- W4387355985 abstract "Current AI-based methods do not provide comprehensible physical interpretations of the utilized data, extracted features, and predictions/inference operations. As a result, deep learning models trained using high-resolution satellite imagery lack transparency and explainability and can be merely seen as a black box, which limits their wide-level adoption. Experts need help understanding the complex behavior of AI models and the underlying decision-making process. The explainable artificial intelligence (XAI) field is an emerging field providing means for robust, practical, and trustworthy deployment of AI models. Several XAI techniques have been proposed for image classification tasks, whereas the interpretation of image segmentation remains largely unexplored. This paper offers to bridge this gap by adapting the recent XAI classification algorithms and making them usable for muti-class image segmentation, where we mainly focus on buildings' segmentation from high-resolution satellite images. To benchmark and compare the performance of the proposed approaches, we introduce a new XAI evaluation methodology and metric based on Entropy to measure the model uncertainty. Conventional XAI evaluation methods rely mainly on feeding area-of-interest regions from the image back to the pre-trained (utility) model and then calculating the average change in the probability of the target class. Those evaluation metrics lack the needed robustness, and we show that using Entropy to monitor the model uncertainty in segmenting the pixels within the target class is more suitable. We hope this work will pave the way for additional XAI research for image segmentation and applications in the remote sensing discipline." @default.
- W4387355985 created "2023-10-05" @default.
- W4387355985 creator A5036799594 @default.
- W4387355985 creator A5060978257 @default.
- W4387355985 creator A5083709105 @default.
- W4387355985 date "2023-10-03" @default.
- W4387355985 modified "2023-10-06" @default.
- W4387355985 title "Extending CAM-based XAI methods for Remote Sensing Imagery Segmentation" @default.
- W4387355985 doi "https://doi.org/10.48550/arxiv.2310.01837" @default.
- W4387355985 hasPublicationYear "2023" @default.
- W4387355985 type Work @default.
- W4387355985 citedByCount "0" @default.
- W4387355985 crossrefType "posted-content" @default.
- W4387355985 hasAuthorship W4387355985A5036799594 @default.
- W4387355985 hasAuthorship W4387355985A5060978257 @default.
- W4387355985 hasAuthorship W4387355985A5083709105 @default.
- W4387355985 hasBestOaLocation W43873559851 @default.
- W4387355985 hasConcept C106301342 @default.
- W4387355985 hasConcept C119857082 @default.
- W4387355985 hasConcept C121332964 @default.
- W4387355985 hasConcept C124101348 @default.
- W4387355985 hasConcept C153180895 @default.
- W4387355985 hasConcept C154945302 @default.
- W4387355985 hasConcept C41008148 @default.
- W4387355985 hasConcept C62520636 @default.
- W4387355985 hasConcept C89600930 @default.
- W4387355985 hasConceptScore W4387355985C106301342 @default.
- W4387355985 hasConceptScore W4387355985C119857082 @default.
- W4387355985 hasConceptScore W4387355985C121332964 @default.
- W4387355985 hasConceptScore W4387355985C124101348 @default.
- W4387355985 hasConceptScore W4387355985C153180895 @default.
- W4387355985 hasConceptScore W4387355985C154945302 @default.
- W4387355985 hasConceptScore W4387355985C41008148 @default.
- W4387355985 hasConceptScore W4387355985C62520636 @default.
- W4387355985 hasConceptScore W4387355985C89600930 @default.
- W4387355985 hasLocation W43873559851 @default.
- W4387355985 hasOpenAccess W4387355985 @default.
- W4387355985 hasPrimaryLocation W43873559851 @default.
- W4387355985 hasRelatedWork W2070902896 @default.
- W4387355985 hasRelatedWork W2501551404 @default.
- W4387355985 hasRelatedWork W2799953226 @default.
- W4387355985 hasRelatedWork W2961085424 @default.
- W4387355985 hasRelatedWork W4298131179 @default.
- W4387355985 hasRelatedWork W4324315429 @default.
- W4387355985 hasRelatedWork W4366829857 @default.
- W4387355985 hasRelatedWork W4379231730 @default.
- W4387355985 hasRelatedWork W4385583601 @default.
- W4387355985 hasRelatedWork W4224009465 @default.
- W4387355985 isParatext "false" @default.
- W4387355985 isRetracted "false" @default.
- W4387355985 workType "article" @default.