Katerina Kikaki

Dr. Katerina Kikaki


Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
7 January 2022 in PLOS ONE
Verified
Journal Article
Open access
PDF file
Katerina KikakiKaterina Kikaki
Katerina Kikaki
Ioannis KakogeorgiouIoannis Kakogeorgiou
Ioannis Kakogeorgiou
PMParaskevi Mikeli
Paraskevi Mikeli
Dionysios RaitsosDionysios Raitsos
Dionysios Raitsos
KKKonstantinos Karantzalos
Konstantinos Karantzalos
Currently, a significant amount of research is focused on detecting Marine Debris and assessing its spectral behaviour via remote sensing, ultimately aiming at new operational monitoring solutions. Here, we introduce a Marine Debris Archive (MARIDA), as a benchmark dataset for developing and evaluating Machine Learning (ML) algorithms capable of detecting Marine Debris. MARIDA is the first dataset based on the multispectral Sentinel-2 (S2) satellite data, which distinguishes Marine Debris from various marine features that co-exist, includingSargassum macroalgae,Ships,Natural Organic Material,Waves,Wakes,Foam, dissimilar water types (i.e.,Clear,Turbid Water,Sediment-Laden Water,Shallow Water), andClouds. We provide annotations (georeferenced polygons/ pixels) from verified plastic debris events in several geographical regions globally, during different seasons, years and sea state conditions. A detailed spectral and statistical analysis of the MARIDA dataset is presented along with well-established ML baselines for weakly supervised semantic segmentation and multi-label classification tasks. MARIDA is an open-access dataset which enables the research community to explore the spectral behaviour of certain floating materials, sea state features and water types, to develop and evaluate Marine Debris detection solutions based on artificial intelligence and deep learning architectures, as well as satellite pre-processing pipelines.

7 March 2024 in ISPRS Journal of Photogrammetry and Remote Sensing
Verified
Journal Article
Open access
Katerina KikakiKaterina Kikaki
Katerina Kikaki
Ioannis KakogeorgiouIoannis Kakogeorgiou
Ioannis Kakogeorgiou
IHIbrahim Hoteit
Ibrahim Hoteit
KKKonstantinos Karantzalos
Konstantinos Karantzalos
1

7 August 2021 in Water
Verified
Journal Article
Open access
PDF file
DVDimitris Vandarakis
Dimitris Vandarakis
Ioannis Polykarpos PanagiotopoulosIoannis Polykarpos Panagiotopoulos
Ioannis Polykarpos Panagiotopoulos
VLVassiliki Loukaidi
Vassiliki Loukaidi
GHGeorgios Angelos Hatiris
Georgios Angelos Hatiris
PDPARASKEVI DRAKOPOULOU
PARASKEVI DRAKOPOULOU
Vasilios KapsimalisVasilios Kapsimalis
Vasilios Kapsimalis
The foreseeable acceleration of global sea level rise could potentially pose a major threat to the natural charm and functional integrity of the world-renowned tourist coastal attractions of Rhodes Island, as a result of the anticipated increasing frequency of flooding and erosion events. Hence, this study aims to determine the most vulnerable segments (in terms of physical impact) of the Rhodes coastline through the widely accepted coastal vulnerability index (CVI), applying a combination of well-known, broadly used approaches and methods. The frequency distribution of the current CVI along the island’s coastline suggests a rather worrying high to very high vulnerability of 40%. In addition, a CVI projection to the end of the 21st century (based on the Intergovernmental Panel on Climate Change predictive scenarios) indicates an enhancement of the total vulnerability by 48%, mainly focused on the majority of the western coastline. Hence, a considerable number of popular coastal destinations in the island shall remain under unignorable threat and, therefore, coastal managers and decision-makers need to hatch an integrated plan to minimize economic and natural losses, private property damage and tourism infrastructure deterioration from flooding and erosion episodes, which will most likely be intensified in the future.

23 June 2022 in Water
Verified
Journal Article
Open access
PDF file
DMDimitra I. Malliouri
Dimitra I. Malliouri
Stelios PetrakisStelios Petrakis
Stelios Petrakis
DVDimitris Vandarakis
Dimitris Vandarakis
Katerina KikakiKaterina Kikaki
Katerina Kikaki
GHGeorgios Angelos Hatiris
Georgios Angelos Hatiris
Vasilios KapsimalisVasilios Kapsimalis
Vasilios Kapsimalis
Wave-induced morphodynamic processes that cause formation, preservation, and destruction of the Prasonisi tombolo in Rhodes Island are investigated, based on satellite image analysis and numerical modeling. A new method is developed for extracting wave events that consist of successive wave data of similar characteristics. The wave events refer either to wind seas or swell seas. This process combined with the satellite image analysis is then utilized for the derivation of the most representative wave scenarios that affect tombolo and salient formation. In particular, the main factors that play a significant role in tombolo and salient evolution are the offshore wave conditions, the location and width of the surf zone, the maximum value of the wave breaking index in the study area, and the initial bottom bathymetry before the study area is exposed to a new sea state. In general, the proposed method provides a realistic insight into tombolo morphodynamics and can be used to provide a cost-effective approach and a wave data-reduction technique for coastal engineering studies.

27 December 2023
Verified
Book Chapter
Closed access
PDF file
DVDimitris Vandarakis
Dimitris Vandarakis
Serafeim E. PoulosSerafeim E. Poulos
Serafeim E. Poulos
Katerina KarditsaKaterina Karditsa
Katerina Karditsa
PSPetrakis Stelios
Petrakis Stelios
George AlexandrakisGeorge Alexandrakis
George Alexandrakis
Vasilios KapsimalisVasilios Kapsimalis
Vasilios Kapsimalis

12 June 2026 in Remote Sensing
Verified
Journal Article
Open access
PDF file
GGGeorge Giannopoulos
George Giannopoulos
MKMaria Kremezi
Maria Kremezi
Vassilia KarathanassiVassilia Karathanassi
Vassilia Karathanassi
Vassilios K. AndronisVassilios K. Andronis
Vassilios K. Andronis
DBDimitris Bliziotis
Dimitris Bliziotis
AMAriane Müting
Ariane Müting
Synthetic Aperture Radar (SAR) is widely used for marine oil spill surveillance due to its all-weather capabilities and sensitivity to sea surface roughness. However, oil slicks often appear as dark formations that can be confounded with visually similar “look-alikes”, making automated detection and boundary delineation challenging. This study proposes a two-stage deep learning framework for oil spill mapping in Sentinel-1 SAR imagery. First, a ConvNeXt-T classifier screens image patches for likely slick presence, reducing the search space for dense prediction. Second, spill boundaries are extracted with a domain-adapted Segment Anything Model (SAM) configured for prompt-free, single-shot segmentation. The input representation is enhanced by combining preprocessed Sentinel-1 VV backscatter with Gray-Level Co-occurrence Matrix (GLCM) texture measures (homogeneity and variance) to better separate oil from heterogeneous background sea at the segmentation level. Quantitative evaluation against established segmentation baselines demonstrates that our adapted SAM achieves the highest overall accuracy, reaching an F1-score of 0.86. This outperforms traditional models such as UNet and CBDNet (0.83), as well as DeepLabV3, SegNeXt, and OFCNet (all at 0.82). Furthermore, an analysis of the wind speed on the test set shows that wind speed affects detectability but does not by itself determine segmentation quality. The results indicate that combining transformer-based screening with efficient foundation-model adaptation can provide accurate and scalable oil spill mapping for operational SAR monitoring.