ΕΛ EN
Home | People | Makri, Despina
22
Scholarly Works
27%
OA Ratio
18%
Collaborative Ratio
Scholar

Makri, Despina

4
H-index (Scopus)
-
Scholar Citations

Scholarly Works

Research Output

Recent Publications

Marine Oil Spill Monitoring using Sentinel-1 Dual-Pol SAR Data: An Operational GEE Framework

Hadjimitsis, Diofantos G., Iqbal, Muhammad Amjad, Kalogirou, Eleftheria, Makri, Despina, Mettas, Christodoulos
Publishing year: 2026
Journal article
IEEE Geoscience and Remote Sensing Letters, 2026
Repository
Oil Spills (OS) pose serious hazards to economies relying on the sea. Remote sensing facilitates effective offshore OS monitoring. As a result, the Sentinel-1 Synthetic Aperture Radar (SAR) missions regularly deliver open-source cloud-stored data through the Google Earth Engine (GEE) platform. This paper proposed an interactive GEE framework for operational use with Sentinel-1 Ground Range Detected (GRD) Dual-pol SAR data. An Eigen descriptor is developed to examine multiple OS incidents in GEE, and two OS from Cyprus are described in detail. OS signatures extracted from Single Look Complex (SLC) SAR and GRD are compared with the baseline Random Forest (RF) algorithm. Also, OS delineation results are showcased for the three SAR scenes reported in the literature. Qualitative outcomes show effectiveness for the GEE-based framework, whereas quantitative results are outstanding. Collectively, the proposed technique achieved an mIoU of 96%, a recall of 99%, an F1-score of 99% and accuracy of 98%. The novelty of the developed technique is an operational value of the GEE platform for Near-Real-Time (NRT) OS monitoring and actionable remote marine environment safety using SAR images as a “Black-box”.

Seagrass Mapping in Cyprus Using Earth Observation Advances

Poursanidis D., Stylianou N., Mettas C., Traganos D., Makri D., Christofilakos S., Hadjimitsis D.
Publishing year: 2025
Journal article
Remote Sensing, 2025, Volume 17, Issue 21
Scopus
Highlights: What are the main findings? Scalable workflow was developed for local-scale seagrass mapping in Cyprus using Sentinel-2 imagery, cloud computing and machine learning. The workflow maps key Natura 2000 habitats—soft bottoms, hard bottoms, and Posidonia beds—along the Cypriot coastline. The method estimated 10-17 km2 of seagrass with approximately 19,000 Mg C stored in Posidonia oceanica meadows. What is the implication of the main finding? The approach addresses a knowledge gap in the Eastern Mediterranean, providing a replicable, consistent methodology for local- and country-scale mapping. The integration of open-access satellite data and cloud computing supports sustainable blue-carbon management and conservation planning. Seagrass meadows are vital for biodiversity and provide a plethora of ecosystem services, but significant losses due to human activity and climate change have been observed in recent decades. This study aims to evaluate whether the integration of Sentinel-2 composites, cloud computing (Google Earth Engine, GEE), and machine learning (ML) classifiers can produce accurate, scalable maps of seagrass habitats, enabling reliable estimates of associated carbon stocks. In this case study, we developed a methodological workflow for local-scale seagrass mapping in Cyprus, covering a total area of 310 km2. ML techniques, specifically the Random Forest (RF) classifier and Classification And Regression Tree (CART), were employed in the main processing stag...

Oil Spill Detection using Convolutional Neural Networks and Sentinel-1 SAR Imagery

Christofi, Konstantinos, Hadjimitsis, Diofantos G., Iqbal, Muhammad Amjad, Kalogirou, Eleftheria, La Pegna, Valeria, Makri, Despina, Mettas, Christodoulos, Tzouvaras, Marios
Publishing year: 2025
Conference paper
ISPRS Geospatial Week 2025 “Photogrammetry & Remote Sensing for a Better Tomorrow, 2025, Volume 48, Issue XLVIII-G-2025
Repository
Pages 757-764
Oil spills impose significant environmental challenges, leading to critical consequences for marine ecosystems and sea habitant’s health. Early delineatin and efficient surveillance are absolutely important to prevent more contamination and support quick hazards reduction. This study focuses on detecting oil spills using satellite imagery and deep learning models, specifically Convolutional Neural Networks (CNN). The dataset used to train the CNN comprised 695 images extracted from Sentinel-1 Synthetic Aperture Radar (SAR) data over the Mediterranean Sea. In particular, 486 images (70%) were allocated for training, 139 images (20%) for validation, and 70 images (10%) for testing. Preprocessing involved a thresholding technique to enhance feature extraction and improve classification precision. The CNN model achieved a high test accuracy of 98.57%, with perfect precision (1.0000), recall of 96.43%, and F1 score of 0.9818, demonstrating strong performance and reliability. These high accuracy levels underscore the model’s efficiency in identifying oil spills and its soundness in handling unseen data. The significance of this work is in using satellite-based deep learning models for scalable and automated oil spill detection, therefore providing a reliable and effective substitute for more traditional monitoring systems. The model may be applied over large oceanic areas by using satellite images, thereby supporting marine ecosystem preservation and enhancing environmental risk ma...

Habitat types mapping using remote sensing techniques in Cyprus

Hadjimitsis, Diofantos G., Iqbal, Muhammad Amjad, Kalogirou, Eleftheria, Makri, Despina, Mettas, Christodoulos
Publishing year: 2025
Conference paper
2025, Volume 13816
Repository
Pages 1381606-1381607
Oceans cover 75% of the Earth's surface and account for 99% of the planet's living space by volume. They are home to around 200,000 known species, with Posidonia oceanica meadows being among the oldest. These habitats are very important coastal ecosystems, providing vital services such as habitat for marine life, coastal protection, and oxygen generation. Despite their importance, seagrass meadows, including Posidonia oceanica, face significant threats, including degradation and loss, with slow regeneration rates. As a result, their monitoring and conservation are essential for maintaining biodiversity and the health of marine ecosystems. This study presents an innovative approach for mapping and monitoring seagrass meadows in western Cyprus using the Google Earth Engine (GEE) platform. Leveraging freely available satellite data from the Copernicus Program, we demonstrate how Earth observation tools can be utilized to map the spatial extent of seagrass meadows and other significant habitat types. This approach supports the monitoring efforts of the European Union’s conservation frameworks and aligns with the United Nations' Sustainable Development Goal (SDG) 14, focusing on the conservation of marine life. The integration of GEE for seagrass mapping offers a scalable and efficient method to track habitat changes, contributing to more effective conservation strategies and ensuring the sustainable use of coastal ecosystems.

Comparison of Novel, Classical and Developed AI Techniques for SAR Oil Spill Detection

Hadjimitsis, Diofantos G., Iqbal, Muhammad Amjad, Kalogirou, Eleftheria, Makri, Despina, Mettas, Christodoulos, Tzouvaras, Marios
Publishing year: 2025
Conference paper
International Conference on Machine Intelligence for Geoanalytics and Remote Sensing, 2025
Repository

CFAR-based automatic ship detection exploiting polarimetric correlation descriptors

Tzouvaras M., Mettas C., Makri D., Iqbal M.A., Kalogirou E., Hadjimitsis D.
Publishing year: 2025
Conference paper
Proceedings of SPIE the International Society for Optical Engineering, 2025, Volume 13816
Scopus
Synthetic Aperture Radar (SAR) provides a reliable remote sensing solution for detecting ships over vast oceanic regions under all-weather and day-night conditions. However, the normalized radar cross-section (NRCS) often encounters ambiguities that hinder accurate detection. To address this, we utilized the polarimetric correlation of SAR channels, including co-, cross-, and dual-polarization, to effectively mitigate or suppress these ambiguities (or clutter). By integrating these polarimetric correlation descriptors with a constant false alarm rate (CFAR) algorithm, we achieve automatic and robust ship detection. Experimental results demonstrate that the proposed approach yields high detection accuracy with minimal root mean square error (RMSE), outperforming conventional manual image interpretation methods. The framework was applied to SAR data for the Mediterranean Sea near Cyprus, showcasing its potential for detecting and monitoring ships, preventing illegal fishing, and addressing maritime security challenges.

22
Unique publications
27.3%
Open access (6 publications; 3 without information)
18.2%
With other institutions (4 publications; 4 without information)

H-Index Reports

Index type Scopus Google Scholar Web of Science
H-index 4 0 0
H5-index 1 0 0
H10-index 4 0 0

Publications per year and source

No description yet.