Special Issue "Pattern Analysis in Remote Sensing"

A special issue of Remote Sensing (ISSN 2072-4292). This special issue belongs to the section "Engineering Remote Sensing".

Deadline for manuscript submissions: 30 November 2021.

Special Issue Editors

Dr. Mohamed Lamine Mekhalfi
E-Mail Website
Guest Editor
Department of Information Engineering and Computer Science, University of Trento, Trento, Italy
Interests: pattern recognition; computer vision; remote sensing
Prof. Dr. Yakoub Bazi
E-Mail
Guest Editor
Dr. Edoardo Pasolli
E-Mail Website
Guest Editor
Department of Agricultural Sciences, University of Naples Federico II, Via Università 100, 80055 Portici, Naples, Italy
Interests: multi/hyperspectral remote sensing; image processing and analysis; machine learning; pattern recognition; computer vision
Special Issues, Collections and Topics in MDPI journals

Special Issue Information

Dear Colleagues,

Remote sensing constitutes an essential instrument towards the monitoring of earthly changes. It has been effectively adopted in the pre- and post-analysis of various civilian (e.g., urban expansion), environmental (e.g., natural disaster prevention/aftermath), as well as military applications (e.g., target detection and localization). Furthermore, the onset of several European and American missions has brought about voluminous data, often in the form of multispectral and hyperspectral images. In order to avail the most out of these latter ones, their handling and analysis require particular attention. Further, the advent of miniaturized unmanned aerial vehicles (UAV) has enabled the acquisition of high-resolution data, which is prone to facilitate the pinpointing of fine details.

In the last decade, a cutting-edge performance has been witnessed in several remote sensing applications such as object detection and/or segmentation, owing for instance to deep learning architectures. However, it is evident that remote sensing has been lagging behind computer vision, which has quickly moved past traditional tasks such as object recognition and is now already paving the way toward bigger challenges such as image description and annotation, commonly termed as the “next frontier”.

In this regard, this Special Issue encourages the submission of papers that offer challenging applications and innovative solutions in the wide topic of remote sensing image analysis. In particular, topics that fall within (but are not limited to) the following are welcome:

  • Multispectral/hyperspectral remote sensing image classification, segmentation, fusion;
  • Natural disaster analysis (e.g., evolution, aftermath) via remote sensing data;
  • Object detection and/or estimation/counting in remote sensing images;
  • Mapping between natural language and remote sensing images;
  • Deep architectures in remote sensing data;
  • Change detection in remote sensing images;
  • UAV image analysis.

Dr. Mohamed Lamine Mekhalfi
Dr. Yakoub Bazi
Dr. Edoardo Pasolli
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All papers will be peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Remote Sensing is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2400 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Keywords

  • remote sensing image analysis
  • object detection
  • deep learning
  • image description
  • change detection

Published Papers (3 papers)

Order results
Result details
Select all
Export citation of selected articles as:

Research

Article
SFRS-Net: A Cloud-Detection Method Based on Deep Convolutional Neural Networks for GF-1 Remote-Sensing Images
Remote Sens. 2021, 13(15), 2910; https://0-doi-org.brum.beds.ac.uk/10.3390/rs13152910 - 24 Jul 2021
Cited by 2 | Viewed by 467
Abstract
Clouds constitute a major obstacle to the application of optical remote-sensing images as they destroy the continuity of the ground information in the images and reduce their utilization rate. Therefore, cloud detection has become an important preprocessing step for optical remote-sensing image applications. [...] Read more.
Clouds constitute a major obstacle to the application of optical remote-sensing images as they destroy the continuity of the ground information in the images and reduce their utilization rate. Therefore, cloud detection has become an important preprocessing step for optical remote-sensing image applications. Due to the fact that the features of clouds in current cloud-detection methods are mostly manually interpreted and the information in remote-sensing images is complex, the accuracy and generalization of current cloud-detection methods are unsatisfactory. As cloud detection aims to extract cloud regions from the background, it can be regarded as a semantic segmentation problem. A cloud-detection method based on deep convolutional neural networks (DCNN)—that is, a spatial folding–unfolding remote-sensing network (SFRS-Net)—is introduced in the paper, and the reason for the inaccuracy of DCNN during cloud region segmentation and the concept of space folding/unfolding is presented. The backbone network of the proposed method adopts an encoder–decoder structure, in which the pooling operation in the encoder is replaced by a folding operation, and the upsampling operation in the decoder is replaced by an unfolding operation. As a result, the accuracy of cloud detection is improved, while the generalization is guaranteed. In the experiment, the multispectral data of the GaoFen-1 (GF-1) satellite is collected to form a dataset, and the overall accuracy (OA) of this method reaches 96.98%, which is a satisfactory result. This study aims to develop a method that is suitable for cloud detection and can complement other cloud-detection methods, providing a reference for researchers interested in cloud detection of remote-sensing images. Full article
(This article belongs to the Special Issue Pattern Analysis in Remote Sensing)
Show Figures

Graphical abstract

Article
Early Identification of Root Rot Disease by Using Hyperspectral Reflectance: The Case of Pathosystem Grapevine/Armillaria
Remote Sens. 2021, 13(13), 2436; https://0-doi-org.brum.beds.ac.uk/10.3390/rs13132436 - 22 Jun 2021
Viewed by 861
Abstract
Armillaria genus represents one of the most common causes of chronic root rot disease in woody plants. Prompt recognition of diseased plants is crucial to control the pathogen. However, the current disease detection methods are limited at a field scale. Therefore, an alternative [...] Read more.
Armillaria genus represents one of the most common causes of chronic root rot disease in woody plants. Prompt recognition of diseased plants is crucial to control the pathogen. However, the current disease detection methods are limited at a field scale. Therefore, an alternative approach is needed. In this study, we investigated the potential of hyperspectral techniques to identify fungi-infected vs. healthy plants of Vitis vinifera. We used the hyperspectral imaging sensor Specim-IQ to acquire leaves’ reflectance data of the Teroldego Rotaliano grapevine cultivar. We analyzed three different groups of plants: healthy, asymptomatic, and diseased. Highly significant differences were found in the near-infrared (NIR) spectral region with a decreasing pattern from healthy to diseased plants attributable to the leaf mesophyll changes. Asymptomatic plants emerged from the other groups due to a lower reflectance in the red edge spectrum (around 705 nm), ascribable to an accumulation of secondary metabolites involved in plant defense strategies. Further significant differences were observed in the wavelengths close to 550 nm in diseased vs. asymptomatic plants. We evaluated several machine learning paradigms to differentiate the plant groups. The Naïve Bayes (NB) algorithm, combined with the most discriminant variables among vegetation indices and spectral narrow bands, provided the best results with an overall accuracy of 90% and 75% in healthy vs. diseased and healthy vs. asymptomatic plants, respectively. To our knowledge, this study represents the first report on the possibility of using hyperspectral data for root rot disease diagnosis in woody plants. Although further validation studies are required, it appears that the spectral reflectance technique, possibly implemented on unmanned aerial vehicles (UAVs), could be a promising tool for a cost-effective, non-invasive method of Armillaria disease diagnosis and mapping in-field, contributing to a significant step forward in precision viticulture. Full article
(This article belongs to the Special Issue Pattern Analysis in Remote Sensing)
Show Figures

Graphical abstract

Article
Damage-Map Estimation Using UAV Images and Deep Learning Algorithms for Disaster Management System
Remote Sens. 2020, 12(24), 4169; https://0-doi-org.brum.beds.ac.uk/10.3390/rs12244169 - 19 Dec 2020
Cited by 2 | Viewed by 1354
Abstract
Estimating the damaged area after a forest fire is important for responding to this natural catastrophe. With the support of aerial remote sensing, typically with unmanned aerial vehicles (UAVs), the aerial imagery of forest-fire areas can be easily obtained; however, retrieving the burnt [...] Read more.
Estimating the damaged area after a forest fire is important for responding to this natural catastrophe. With the support of aerial remote sensing, typically with unmanned aerial vehicles (UAVs), the aerial imagery of forest-fire areas can be easily obtained; however, retrieving the burnt area from the image is still a challenge. We implemented a new approach for segmenting burnt areas from UAV images using deep learning algorithms. First, the data were collected from a forest fire in Andong, the Republic of Korea, in April 2020. Then, the proposed two-patch-level deep-learning models were implemented. A patch-level 1 network was trained using the UNet++ architecture. The output prediction of this network was used as a position input for the second network, which used UNet. It took the reference position from the first network as its input and refined the results. Finally, the final performance of our proposed method was compared with a state-of-the-art image-segmentation algorithm to prove its robustness. Comparative research on the loss functions was also performed. Our proposed approach demonstrated its effectiveness in extracting burnt areas from UAV images and can contribute to estimating maps showing the areas damaged by forest fires. Full article
(This article belongs to the Special Issue Pattern Analysis in Remote Sensing)
Show Figures

Figure 1

Planned Papers

The below list represents only planned manuscripts. Some of these manuscripts have not been received by the Editorial Office yet. Papers submitted to MDPI journals are subject to peer-review.

Title: Evolving neural architectures for semantic segmentation of high resolution satellite images.
Authors: Marcin Pietron; Radoslaw Szostak
Affiliation: AGH - University of Science and Technology

Title: DPA-KB: A Structural Drainage Pattern Recognition Method and Knowledge-based System
Authors: Evangelos Roussos; Demetre Argialas
Affiliation: Ronin Institute for Independent Scholarship, Montclair, NJ, USA National Technical University of Athens, Athens, Greece
Abstract: This paper presents DPA-KB, a method and associated software for the analysis and recognition of a category of complex, tree-like natural objects called drainage networks. The system models geomorphological knowledge of drainage network patterns (DPs) and captures photointerpretation knowledge using structural pattern recognition theory combined with concepts from knowledge-based systems, to effectively recognise DPs in an automated manner. Regarding domain knowledge representation, a hierarchical structural pattern representation scheme was used, in order to express composition of DPs from their simpler subpatterns at various levels of complexity. The higher-level, semantic objects are used to model the geomorphological qualitative ``definitions'' (descriptions) of the DP classes. The arcs of the graph correspond to an aggregation relation, meaning here that of objects combining to form composite objects. For the quantitative description of DPs, a variety of topological and geometric attributes were selected. From the point of view of attribute interaction and functional abstraction, these can be classified as `simple' object attributes, aggregate attributes, and attributes of the relations among the various objects. These also form a hierarchy. The final, highest-level attributes are compatible with domain knowledge and include: straightness, uniformity, and elongation measures of objects, various angular attributes between objects, branching ratios, and branching types. With the addition of the above relations, the tree-like hierarchical description of the patterns is generalised to a relational graph. The digitisation procedure of the pattens was a kind of depth-first node encoding, which allows the reconstruction of the topology of the DPs. For the hierarchical description of the DPs, their subparts are identified by a conceptual grouping process, in a bottom-up manner. The aggregation criteria included: proximity, similarity, and continuation of an uninterrupted attribute or relationship. To cope with the inherent complexity and variation of the patterns, and to incorporate qualitative expert knowledge, in the form of rules, a method for handling uncertainty is needed. In particular, in this system numerical-valued attributes were converted to symbolic attributes through appropriate thresholds. `Linguistic hedges', such as `nearly' (as in ``nearly right angle''), were employed through the use of mean values and standard deviation thresholds. The critical values of all the above variables have been determined by empirical analysis. Furthermore, quantities reminiscent of symbolic probabilities were assigned to the attribute values, to express their contribution to the degree of certainty in the classification of the DPs. These can be thought of as symbolic confidence-like measures of the form C(Y|X,I), where Y is the pattern class, X is a set of pattern attributes, and I is the context, i.e. the part of the decision tree that the decision process is at. Finally, a hierarchical conceptual organisation of pattern classes, which is a `is-a' hierarchy representing class inclussion relationship, was designed. For the classification of DPs, a decision tree classifier approach was selected, in order to test for the occurences or non-occurences of certain pattern attributes and combinations of them in a top-down strategy. Each decision node contains a set of rules and combines the uncertainties of its constituent knowledge in a way that is appropriate to the context. That is, each level of the hierarchy combines the constituent pieces of evidence, to form a symbolic uncertainty measure for the next level. The system was tested using several drainage patterns, and the results indicate that the computer can effectively assist in the photointerpretation of such patterns.

Back to TopTop