Karima Hadj-Rabah
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Karima Hadj-Rabah has a total of 25 co-authors. Here are the top co-authors by number of shared publications:
- Gilda Schirinzi
University of Naples Parthenope
10 shared publications - Nabil Haddad
Antennas and Microwave Devices Laboratory, Ecole Militaire Polytechnique, Algiers, Algeria
6 shared publications - Alessandra Budillon
Department of Engineering, Universit degli Studi di Napoli Parthenope, Naples, Italy
5 shared publications - Azzedine Bouaraba
Radar Laboratory, Ecole Militaire Polytechnique, Algiers, Algeria
4 shared publications Faiza Hocine
Department of Geoinformatics, Paris-Lodron University of Salzburg, Salzburg, Austria
4 shared publications- Aichouche Belhadj-Aissa
University of Sciences and Technology Houari Boumediene
3 shared publications
Latest Publications
- Open Access
Assessment of Dual-Polarization Sentinel-1 SAR Data for Improved Wildfire Burned Area Mapping: A Case Study of the Palisades Region, USA
Wildfires have become more frequent and intense worldwide due to climate change and anthropogenic activities, which is why accurate and timely burned area mapping is essential for estimating damage and effective post-fire recovery planning. Synthetic Aperture Radar (SAR) data, which operates under all weather conditions and day-night cycles, offers a reliable source for burned area mapping. In this context, several studies have explored the use of dual-polarization SAR imagery and machine learning, yet the influence of multi-date, dual-orbit pass data and texture features remained unexplored. Therefore, this study aims to assess the Sentinel-1 acquisition configurations, varying in temporal depth and orbital direction, for wildfire burned area mapping, considering the recent Palisades wildfire event as a study area. A comparative study was conducted across different scenarios to evaluate the effectiveness of using single-date versus multi-date SAR imagery, the integration of ascending and descending orbit passes, and the contribution of Grey-Level Co-occurrence Matrix texture features. The performance of Random Forest (RF) and Extreme Gradient Boosting classifiers was analyzed through the scenarios mentioned above. The single-date configuration using RF achieved an accuracy of 82.34%, F1-score of 81.43%, precision of 83.07%, recall of 80.84%, and ROC-AUC of 90.88%, whereas the multi-date approach reached 85.78%, 85.15%, 86.45%, 84.56%, and 93.28%, respectively. Our study highlights the importance of acquisition configuration and texture information for reliable SAR-based wildfire burned area assessment.
Open AccessAssessment of Dual-Polarization Sentinel-1 SAR Data for Improved Wildfire Burned Area Mapping: A Case Study of the Palisades Region, USA
Wildfires have become more frequent and intense worldwide due to climate change and anthropogenic activities, which is why accurate and timely burned area mapping is essential for estimating damage and effective post-fire recovery planning. Synthetic Aperture Radar (SAR) data, which operates under all weather conditions and day-night cycles, offers a reliable source for burned area mapping. In this context, several studies have explored the use of dual-polarization SAR imagery and machine learning, yet the influence of multi-date, dual-orbit pass data and texture features remained unexplored. Therefore, this study aims to assess the Sentinel-1 acquisition configurations, varying in temporal depth and orbital direction, for wildfire burned area mapping, considering the recent Palisades wildfire event as a study area. A comparative study was conducted across different scenarios to evaluate the effectiveness of using single-date versus multi-date SAR imagery, the integration of ascending and descending orbit passes, and the contribution of Grey-Level Co-occurrence Matrix texture features. The performance of Random Forest (RF) and Extreme Gradient Boosting classifiers was analyzed through the scenarios mentioned above. The single-date configuration using RF achieved an accuracy of 82.34%, F1-score of 81.43%, precision of 83.07%, recall of 80.84%, and ROC-AUC of 90.88%, whereas the multi-date approach reached 85.78%, 85.15%, 86.45%, 84.56%, and 93.28%, respectively. Our study highlights the importance of acquisition configuration and texture information for reliable SAR-based wildfire burned area assessment.
- Open Access
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Geoscience and Remote Sensing Letters
- Vol. 23, 2505605
Geometry-Based Forest Structure Extraction From SAR Tomographic Data for Terrain and Canopy Modeling
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Geoscience and Remote Sensing Letters
- Vol. 23, 2505605
Open AccessGeometry-Based Forest Structure Extraction From SAR Tomographic Data for Terrain and Canopy Modeling
- Open Access
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
- Vol. 19, 21849-21862
Gridless GLRT Approach for Polarimetric SAR Tomography
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
- Vol. 19, 21849-21862
Open AccessGridless GLRT Approach for Polarimetric SAR Tomography
- Open Access
- 25 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 63 (01962892) , 1-13
Adaptive Coherent Multilook GLRT for SAR Tomography Detection
In recent years, generalized likelihood ratio test (GLRT) scatterers’ detection in the context of synthetic aperture radar tomography (TomoSAR) has gained great interest from the remote sensing scientific community. This is due to its effectiveness in identifying scatterers within each single azimuth–range resolution cell, particularly in urban areas. The multilook GLRT (M-GLRT) variant offers more satisfactory results at the expense of spatial resolution deterioration, by jointly exploiting the neighboring information of the pixel to be reconstructed. In this context, coherent and incoherent formulations can be adopted. The former provides better performance in the assumption of constant reflectivity of the pixels in the considered neighborhood, while the latter is much more robust with respect to the violation of this assumption. In this article, an adaptive formulation of the coherent and incoherent GLRT is presented with the aim of improving scatterers’ detection and their height estimation. The method is based on an adaptive window setting, to select adjacent pixels with similar height and reflectivity characteristics. A detailed study and analysis of the proposed adaptive coherent M-GLRT (ACM-GLRT) detector has been conducted and validated through simulations along with comparison to both standard and adaptive formulations of incoherent M-GLRT. Experimental findings from a real dataset acquired by the German TerraSAR-X (TSX) over the city of Naples (Italy) demonstrate the performance improvement of our proposed approach.
- 25 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 63 (01962892) , 1-13
Open AccessAdaptive Coherent Multilook GLRT for SAR Tomography Detection
In recent years, generalized likelihood ratio test (GLRT) scatterers’ detection in the context of synthetic aperture radar tomography (TomoSAR) has gained great interest from the remote sensing scientific community. This is due to its effectiveness in identifying scatterers within each single azimuth–range resolution cell, particularly in urban areas. The multilook GLRT (M-GLRT) variant offers more satisfactory results at the expense of spatial resolution deterioration, by jointly exploiting the neighboring information of the pixel to be reconstructed. In this context, coherent and incoherent formulations can be adopted. The former provides better performance in the assumption of constant reflectivity of the pixels in the considered neighborhood, while the latter is much more robust with respect to the violation of this assumption. In this article, an adaptive formulation of the coherent and incoherent GLRT is presented with the aim of improving scatterers’ detection and their height estimation. The method is based on an adaptive window setting, to select adjacent pixels with similar height and reflectivity characteristics. A detailed study and analysis of the proposed adaptive coherent M-GLRT (ACM-GLRT) detector has been conducted and validated through simulations along with comparison to both standard and adaptive formulations of incoherent M-GLRT. Experimental findings from a real dataset acquired by the German TerraSAR-X (TSX) over the city of Naples (Italy) demonstrate the performance improvement of our proposed approach.
- Open Access
- 8 July 2025
- Published by MDPI AG in Remote Sensing
- Vol. 17 (14) , 2334
Beyond the Grid: GLRT-Based TomoSAR Fast Detection for Retrieving Height and Thermal Dilation
The Tomographic Synthetic Aperture Radar (TomoSAR) technique is widely used for monitoring urban infrastructures, as it enables the mapping of individual scatterers across additional dimensions such as height (3D), thermal dilation (4D), and deformation velocity (5D). Retrieving this information is crucial for building management and maintenance. Nevertheless, accurately extracting it from TomoSAR data poses several challenges, particularly the presence of outliers due to uneven and limited baseline distributions. One way to address these issues is through statistical detection approaches such as the Generalized Likelihood Ratio Test, which ensures a Constant False Alarm Rate. While effective, these methods face two primary limitations: high computational complexity and the off-grid problem caused by the discretization of the search space. To overcome these drawbacks, we propose an approach that combines a quick initialization process using Fast-Sup GLRT with local descent optimization. This method operates directly in the continuous domain, bypassing the limitations of grid-based search while significantly reducing computational costs. Experiments conducted on both simulated and real datasets acquired with the TerraSAR-X satellite over the Spanish city of Barcelona demonstrate the ability of the proposed approach to maintain computational efficiency while improving scatterer localization accuracy in the third and fourth dimensions.
- 8 July 2025
- Published by MDPI AG in Remote Sensing
- Vol. 17 (14) , 2334
Open AccessBeyond the Grid: GLRT-Based TomoSAR Fast Detection for Retrieving Height and Thermal Dilation
The Tomographic Synthetic Aperture Radar (TomoSAR) technique is widely used for monitoring urban infrastructures, as it enables the mapping of individual scatterers across additional dimensions such as height (3D), thermal dilation (4D), and deformation velocity (5D). Retrieving this information is crucial for building management and maintenance. Nevertheless, accurately extracting it from TomoSAR data poses several challenges, particularly the presence of outliers due to uneven and limited baseline distributions. One way to address these issues is through statistical detection approaches such as the Generalized Likelihood Ratio Test, which ensures a Constant False Alarm Rate. While effective, these methods face two primary limitations: high computational complexity and the off-grid problem caused by the discretization of the search space. To overcome these drawbacks, we propose an approach that combines a quick initialization process using Fast-Sup GLRT with local descent optimization. This method operates directly in the continuous domain, bypassing the limitations of grid-based search while significantly reducing computational costs. Experiments conducted on both simulated and real datasets acquired with the TerraSAR-X satellite over the Spanish city of Barcelona demonstrate the ability of the proposed approach to maintain computational efficiency while improving scatterer localization accuracy in the third and fourth dimensions.
- Open Access
- 24 May 2025
- Published by Copernicus GmbH in The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
- Vol. XLVIII-M-7, 89-96
Structuring Thesis Topics within the Copernicus Master in Digital Earth: Preparing Alumni for Future-Oriented Careers in the Space Industry
The Copernicus Master in Digital Earth (CDE), a joint European Master of Science, co-funded by the Erasmus Mundus programme, addresses the growing demand for geospatial experts by equipping students with advanced skills in Earth observation, geoinformatics, geovisualisation and geodata science. We propose a unified master’s thesis framework that encompasses development strategy, geospatial datasets, implementation mode, geographic scope, and application domains, aiming to bridge theory and practice while fostering innovative and research-driven education. Compulsory internships linked to thesis projects strengthen academia-industry collaboration, enabling students to address societal challenges such as climate change, green transition and digital transformation. The CDE programme has produced numerous joint publications, dedicated industry partnerships and even start-ups, with alumni excelling in academia, industry and public service. By uniting research, education and industry, the CDE exemplifies a role model for educating geospatial professionals in the broader scope of global challenges.
- 24 May 2025
- Published by Copernicus GmbH in The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
- Vol. XLVIII-M-7, 89-96
Open AccessStructuring Thesis Topics within the Copernicus Master in Digital Earth: Preparing Alumni for Future-Oriented Careers in the Space Industry
The Copernicus Master in Digital Earth (CDE), a joint European Master of Science, co-funded by the Erasmus Mundus programme, addresses the growing demand for geospatial experts by equipping students with advanced skills in Earth observation, geoinformatics, geovisualisation and geodata science. We propose a unified master’s thesis framework that encompasses development strategy, geospatial datasets, implementation mode, geographic scope, and application domains, aiming to bridge theory and practice while fostering innovative and research-driven education. Compulsory internships linked to thesis projects strengthen academia-industry collaboration, enabling students to address societal challenges such as climate change, green transition and digital transformation. The CDE programme has produced numerous joint publications, dedicated industry partnerships and even start-ups, with alumni excelling in academia, industry and public service. By uniting research, education and industry, the CDE exemplifies a role model for educating geospatial professionals in the broader scope of global challenges.
- Open Access
- 8 May 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 63 (01962892) , 1-12
Contextual Tomographic SAR Denoising Approach for Estimating Scatterers’ Height and Deformation Velocity
The reliability of Tomographic SAR (TomoSAR) products depends on the quality of complex-valued tomographic interferograms. The latters are unfortunately affected by noise from various sources. In order to reduce their impact and hence improve the outcome, filtering methods can be implemented at different levels of TomoSAR process. In this paper, we propose the application of a contextual denoising approach in transformed domains, based on Subbands decomposition and non-linear weighting, with the aim to study its influence on TomosAR height and deformation velocity estimation using Generalized Likelihood Ratio Test detection. Both spatial and spatiotemporal arrangements of overlapping blocks are considered in wavelet domains. The non-linear filtering parameter is estimated from the coherence and/or pseudo-correlation indicators. In order to show the effectiveness of the approach, the obtained findings have been compared to the state-of-the-art methods namely Goldstein and Baran filters. The assessment of the results with respect to denoising and TomoSAR evaluation metrics was carried out using both simulated and real data acquired by TerraSAR-X satellite over the city of Naples.
- 8 May 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 63 (01962892) , 1-12
Open AccessContextual Tomographic SAR Denoising Approach for Estimating Scatterers’ Height and Deformation Velocity
The reliability of Tomographic SAR (TomoSAR) products depends on the quality of complex-valued tomographic interferograms. The latters are unfortunately affected by noise from various sources. In order to reduce their impact and hence improve the outcome, filtering methods can be implemented at different levels of TomoSAR process. In this paper, we propose the application of a contextual denoising approach in transformed domains, based on Subbands decomposition and non-linear weighting, with the aim to study its influence on TomosAR height and deformation velocity estimation using Generalized Likelihood Ratio Test detection. Both spatial and spatiotemporal arrangements of overlapping blocks are considered in wavelet domains. The non-linear filtering parameter is estimated from the coherence and/or pseudo-correlation indicators. In order to show the effectiveness of the approach, the obtained findings have been compared to the state-of-the-art methods namely Goldstein and Baran filters. The assessment of the results with respect to denoising and TomoSAR evaluation metrics was carried out using both simulated and real data acquired by TerraSAR-X satellite over the city of Naples.
- 30 April 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Geoscience and Remote Sensing Letters
- Vol. 22 (1545598X) , 1-5
Covariance Matrix Estimation via Geometric Median in Highly Heterogeneous PolSAR Images
The Wishart distribution is a well-established statistical model for characterizing the density of random variables in Polarimetric SAR (PolSAR) data, particularly within homogeneous regions where Gaussian assumptions hold. However, as PolSAR applications expand into heterogeneous environments, alternative statistical models have been developed to better capture the complexity of such areas, playing an important role in tasks such as classification. In this study, we examine the effectiveness of covariance matrix estimation using the median matrix, a technique grounded in optimal transport theory and validated in prior research for its effectiveness. Building on this foundation, we propose the application of a statistical model tailored for heterogeneous regions, i.e., following the G0P distribution, addressing the limitations of traditional assumptions. This method is particularly suitable for high-resolution PolSAR datasets, where the homogeneity hypothesis often does not hold. The experimental results obtained using L-band PolSAR images acquired over Foulum in Denmark demonstrate the robustness of our proposed variant.
- 30 April 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Geoscience and Remote Sensing Letters
- Vol. 22 (1545598X) , 1-5
Covariance Matrix Estimation via Geometric Median in Highly Heterogeneous PolSAR Images
The Wishart distribution is a well-established statistical model for characterizing the density of random variables in Polarimetric SAR (PolSAR) data, particularly within homogeneous regions where Gaussian assumptions hold. However, as PolSAR applications expand into heterogeneous environments, alternative statistical models have been developed to better capture the complexity of such areas, playing an important role in tasks such as classification. In this study, we examine the effectiveness of covariance matrix estimation using the median matrix, a technique grounded in optimal transport theory and validated in prior research for its effectiveness. Building on this foundation, we propose the application of a statistical model tailored for heterogeneous regions, i.e., following the G0P distribution, addressing the limitations of traditional assumptions. This method is particularly suitable for high-resolution PolSAR datasets, where the homogeneity hypothesis often does not hold. The experimental results obtained using L-band PolSAR images acquired over Foulum in Denmark demonstrate the robustness of our proposed variant.
- 8 January 2025
- Published by Taylor & Francis in International Journal of Remote Sensing
- Vol. 46 (5) , 2115-2135
Adaptive quality indicator for subband multi-baseline TomoSAR filter
- 8 January 2025
- Published by Taylor & Francis in International Journal of Remote Sensing
- Vol. 46 (5) , 2115-2135
Adaptive quality indicator for subband multi-baseline TomoSAR filter
- 24 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Geoscience and Remote Sensing Letters
- Vol. 21 (1545598X) , 1-5
Gridless GLRT for Tomographic SAR Detection Using Particle Swarm Optimization Algorithm
The detection of multiple scatterers within each resolution cell is an open research subject in Synthetic Aperture Radar tomography. For over a decade, the Generalised Likelihood Ratio Test (GLRT) detector has been implemented along with its variants, allowing the generation of height maps and 3D point clouds with good precision. However, they are limited by the grid search during the optimization of the Maximum Likelihood function. In order to mitigate this, we propose a gridless version of GLRT where the Particle Swarm Optimization (PSO) method is used to locate the minima. The conducted analysis of the proposed detector with respect to the state-of-the-art methods behavior on simulated and real datasets proved the effectiveness of PSO-GLRT in terms of height accuracy and computational cost. The evaluation metrics Root Mean Square Error, Accuracy, and Completeness have been used as a quantitative improvement indicator for estimated height assessment.
- 24 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Geoscience and Remote Sensing Letters
- Vol. 21 (1545598X) , 1-5
Gridless GLRT for Tomographic SAR Detection Using Particle Swarm Optimization Algorithm
The detection of multiple scatterers within each resolution cell is an open research subject in Synthetic Aperture Radar tomography. For over a decade, the Generalised Likelihood Ratio Test (GLRT) detector has been implemented along with its variants, allowing the generation of height maps and 3D point clouds with good precision. However, they are limited by the grid search during the optimization of the Maximum Likelihood function. In order to mitigate this, we propose a gridless version of GLRT where the Particle Swarm Optimization (PSO) method is used to locate the minima. The conducted analysis of the proposed detector with respect to the state-of-the-art methods behavior on simulated and real datasets proved the effectiveness of PSO-GLRT in terms of height accuracy and computational cost. The evaluation metrics Root Mean Square Error, Accuracy, and Completeness have been used as a quantitative improvement indicator for estimated height assessment.