Towards Ontology Reasoning for Topological Cluster Labeling

Chahdi, Hatim; Grozavu, Nistor; Mougenot, Isabelle; Bennani, Younès; Berti‐Équille, Laure

doi:10.1007/978-3-319-46675-0_18

Cited by 5 publications

(5 citation statements)

References 17 publications

(23 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For these reasons, as many scientists working in the field of remote sensing have done [6][7][8], in this paper we will focus mostly on unsupervised learning algorithms, and more specifically unsupervised neural networks. While they still need a lot of data to be trained, they do not require these data to be labeled.…”

Section: Applications Of Artificial Intelligence To Satellite Images mentioning

confidence: 99%

“…In the case of this paper, the application of unsupervised AI techniques to the survey and mapping of damages caused tsunamis presents the extra difficulty that it would not be applied to one remote-sensing image, but rather 2 images (before and after the disaster) to assess the difference between them and deduce the extent of the damages. While clustering techniques as simple as the K-Means algorithm are relatively successful with remote-sensing images [2,7,8], analyzing the differences between two remote-sensing images before and after a geohazard with unsupervised techniques presents some extra difficulties [9]:…”

Section: Applications Of Artificial Intelligence To Satellite Images mentioning

confidence: 99%

See 1 more Smart Citation

Automatic Post-Disaster Damage Mapping Using Deep-Learning Techniques for Change Detection: Case Study of the Tohoku Tsunami

2019

View full text Add to dashboard Cite

Post-disaster damage mapping is an essential task following tragic events such as hurricanes, earthquakes, and tsunamis. It is also a time-consuming and risky task that still often requires the sending of experts on the ground to meticulously map and assess the damages. Presently, the increasing number of remote-sensing satellites taking pictures of Earth on a regular basis with programs such as Sentinel, ASTER, or Landsat makes it easy to acquire almost in real time images from areas struck by a disaster before and after it hits. While the manual study of such images is also a tedious task, progress in artificial intelligence and in particular deep-learning techniques makes it possible to analyze such images to quickly detect areas that have been flooded or destroyed. From there, it is possible to evaluate both the extent and the severity of the damages. In this paper, we present a state-of-the-art deep-learning approach for change detection applied to satellite images taken before and after the Tohoku tsunami of 2011. We compare our approach with other machine-learning methods and show that our approach is superior to existing techniques due to its unsupervised nature, good performance, and relative speed of analysis.

show abstract

Section: Applications Of Artificial Intelligence To Satellite Images mentioning

confidence: 99%

Section: Applications Of Artificial Intelligence To Satellite Images mentioning

confidence: 99%

Automatic Post-Disaster Damage Mapping Using Deep-Learning Techniques for Change Detection: Case Study of the Tohoku Tsunami

2019

View full text Add to dashboard Cite

show abstract

“…Chahdi et al [3] defines an ontology using two conceptual building blocks: TBox (Terminology-Box), composed of concepts (classes), axioms and relations; and ABox (Assertion-Box), to describe individuals and its relationships.…”

Section: Obaa Metadata Ontologymentioning

confidence: 99%

Enhancing Learning Object Repositories with Ontologies

Behr

Mendes

Cascalho

et al. 2021

Advances in Intelligent Systems and Computing

View full text Add to dashboard Cite

In this paper, we present a review on the use of ontologies in learning object repositories systems for searching and suggestion purposes, considering its adoption for the seaThings project that aims to promote the ocean literacy. We also describe the use case of the Cognix system and Agent-based Learning Objects -OBAA metadata standard for learning objects which is being implemented on a new learning objects repository. This system includes concepts from artificial intelligence such as agents and ontologies that aim to improve the search and so making the system more responsive. This paper also sugests how an ontology can be implemented, using metadata in learning object repositories to provide relevant aspects, such as interoperability, reuse, and searching.

show abstract

“…A first possible solution consists of using semi-supervised approaches instead of fully-unsupervised ones: In the case of pixel-based analysis of VHR images, a solution proposed in the literature is to guide the clustering process using ontologies [26,27], a tool commonly used in supervised process. The results achieved using these methods are promising, but seem limited to a very low number of clusters/classes.…”

Section: Unsupervised Analysis Of the Segmentsmentioning

confidence: 99%

Multi-Scale Analysis of Very High Resolution Satellite Images Using Unsupervised Techniques

2017

View full text Add to dashboard Cite

Abstract:This article is concerned with the use of unsupervised methods to process very high resolution satellite images with minimal or little human intervention. In a context where more and more complex and very high resolution satellite images are available, it has become increasingly difficult to propose learning sets for supervised algorithms to process such data and even more complicated to process them manually. Within this context, in this article we propose a fully unsupervised step by step method to process very high resolution images, making it possible to link clusters to the land cover classes of interest. For each step, we discuss the various challenges and state of the art algorithms to make the full process as efficient as possible. In particular, one of the main contributions of this article comes in the form of a multi-scale analysis clustering algorithm that we use during the processing of the image segments. Our proposed methods are tested on a very high resolution image (Pléiades) of the urban area around the French city of Strasbourg and show relevant results at each step of the process.

show abstract

Towards Ontology Reasoning for Topological Cluster Labeling

Cited by 5 publications

References 17 publications

Automatic Post-Disaster Damage Mapping Using Deep-Learning Techniques for Change Detection: Case Study of the Tohoku Tsunami

Automatic Post-Disaster Damage Mapping Using Deep-Learning Techniques for Change Detection: Case Study of the Tohoku Tsunami

Enhancing Learning Object Repositories with Ontologies

Multi-Scale Analysis of Very High Resolution Satellite Images Using Unsupervised Techniques

Contact Info

Product

Resources

About