Soufyane Bouchelaghem

Dr. Soufyane Bouchelaghem

orcid

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

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

Representative Publications

View All Publications
6 March 2025 in Remote Sensing
Verified
Journal Article
Open access
PDF file
Marco BalsiMarco Balsi
Marco Balsi
MMMonica Moroni
Monica Moroni
SBSoufyane Bouchelaghem
Soufyane Bouchelaghem
Plastic waste has become a critical environmental issue, necessitating effective methods for detection and monitoring. This article presents a machine-learning-based methodology and embedded solution to detect plastic waste in the environment using an airborne hyperspectral sensor operating in the short-wave infrared (SWIR) band. Experimental data were obtained from drone flights in several case studies in natural and controlled environments. Data were preprocessed to simply equalize the spectra across the whole band and across different environmental conditions, and machine learning techniques were applied to detect plastics even in real-time. Several algorithms for spectrum calibration, feature selection, and classification were optimized and compared to obtain an optimal solution that has high-quality results under cross-validation. This way, deploying the system in different environments without requiring complicated manual adjustments or re-learning is possible. The results of this work prove the feasibility of the proposed plastic litter detection approach using high-definition aerial remote sensing, with high specificity to plastic polymers that are not obtained using visible and NIR data.

1 May 2025 in Waste Management
Verified
Journal Article
Closed access
MMMonica Moroni
Monica Moroni
Marco BalsiMarco Balsi
Marco Balsi
SBSoufyane Bouchelaghem
Soufyane Bouchelaghem

15 April 2024
Verified
Conference Paper
Closed access
SBSoufyane Bouchelaghem
Soufyane Bouchelaghem
Imad Eddine TIBERMACINEImad Eddine TIBERMACINE
Imad Eddine TIBERMACINE
MBMarco Balsi
Marco Balsi
MMMonica Moroni
Monica Moroni
CNChristian Napoli
Christian Napoli
Plastics detection and polymer separation are important tasks for environmental monitoring and waste recycling. The authors previously used hyperspectral data in the lower bands of the SWIR range from 900nm to 1700 nm to perform plastics classification, by applying supervised machine learning methods to data gathered by a custom-assembled stand-alone push-broom system in indoor and outdoor conditions. In this work, we explored the application of two unsupervised learning algorithms: K-Means clustering and Gaussian Mixture Models (GMM), in order to avoid the use of a manually-labeled test set, and to obtain more reliable and flexible application in different environmental conditions. The results suggest that GMM performs well indoors but poorly outside, with GMM exhibiting subtle segmentation and K-Means displaying broader, more generalist categorization. The cross-domain adaptation is difficult and models that balance granularity and generalization are needed to detect plastic in varied settings. This effort advances garbage management and recycling and establishes hyperspectral imaging in real-world environments.

5 January 2026 in Remote Sensing
Verified
Journal Article
Open access
PDF file
SBSoufyane Bouchelaghem
Soufyane Bouchelaghem
Marco BalsiMarco Balsi
Marco Balsi
MMMonica Moroni
Monica Moroni
The proliferation of plastic waste across natural ecosystems has created a global environmental and public health crisis. Monitoring plastic litter using remote sensing remains challenging due to the significant variability in terrain, lighting, and weather conditions. Although earlier approaches, including classical supervised machine learning techniques such as Linear Discriminant Analysis (LDA) and Support Vector Machine (SVM), applied to hyperspectral and multispectral data have shown promise in controlled settings, they often may face challenges in generalizing across diverse environmental conditions encountered in real-world scenarios. In this work, we present a deep learning framework for pixel-wise segmentation of plastic waste in short-wave infrared (900–1700 nm) hyperspectral imagery acquired from an Unmanned Aerial Vehicle (UAV). Our architecture integrates attention gates and residual connections within a U-Net backbone to enhance contextual modeling and spatial-spectral consistency. We introduce a multi-flight dataset spanning over 9 UAV missions across varied environmental settings, consisting of hyperspectral cubes with centimeter-level resolution. Using a leave-one-out cross-validation protocol, our model achieves test accuracy of up to 96.8% (average 90.5%) and a 91.1% F1 score, demonstrating robust generalization to unseen data collected in different environments. Compared to classical models, the deep network captures richer semantic representations, particularly under challenging conditions. This work offers a scalable and deployable tool for automated plastic waste monitoring and represents a significant advancement in remote environmental sensing.

31 October 2023
Verified
Conference Paper
Closed access
MBMarco Balsi
Marco Balsi
SBSoufyane Bouchelaghem
Soufyane Bouchelaghem
LCLivio Conti
Livio Conti
MMMonica Moroni
Monica Moroni
RSRiccardo Scalia
Riccardo Scalia
A hyperspectral sensing system operating in the SWIR band (900-1700nm) was previously developed for drone-based detection of plastic litter in the environment. The detection software is based on a linear classifier trained on examples, that detects plastics and transmits information to a ground station that visualizes on a map the location of points surveyed, labeled as plastics where applicable. In this paper, we report on implementation of plastics litter detection in real time, while the drone is flying.

31 March 2026 in Frontiers in Marine Science
Verified
Journal Article
Open access
PDF file
SBSoufyane Bouchelaghem
Soufyane Bouchelaghem
LMLahcene Mamen
Lahcene Mamen
MBMarco Balsi
Marco Balsi
ATAhmed Tibermacine
Ahmed Tibermacine
MMMonica Moroni
Monica Moroni
ITImad Eddine Tibermacine
Imad Eddine Tibermacine
Introduction: Marine debris detection from satellite imagery is challenged by two major factors: extreme class imbalance, with debris pixels accounting for less than 0.01% of image content, and the need for robust generalization across diverse geographic and temporal domains for operational deployment. Although existing methods often report strong within-dataset performance, cross-dataset generalization, where models trained on one dataset are applied to entirely different geographic regions, remains insufficiently investigated. Methods: To address this limitation, we conducted rigorous bidirectional cross-dataset validation experiments using the MARIDA and MADOS datasets. The problem was reformulated as a binary segmentation task and addressed using a standard U-Net architecture combined with a composite imbalance-aware loss and a rarity-aware sampling strategy. Two experimental settings were considered: training on MARIDA and testing on MADOS, and training on MADOS and testing on MARIDA. Results: The experiments revealed asymmetric cross-dataset generalization. Models trained on the geographically diverse MADOS dataset achieved an F1-score of 0.890 when tested on MARIDA, corresponding to only a 1.25% decrease from the within-dataset baseline of 0.901. In contrast, models trained on MARIDA achieved an F1-score of 0.833 on MADOS, representing a 7.55% decrease. The average cross-dataset degradation was 4.38%, which is substantially lower than the typical 10--25% performance drops reported in remote sensing domain-shift scenarios. Despite comparable patch counts (2,529 for MADOS versus 2,173 for MARIDA), the superior transferability of MADOS-trained models indicates that geographic diversity across globally distributed tiles is more beneficial than exhaustive annotation within concentrated regions. Moreover, the MADOS-to-MARIDA cross-dataset F1-score of 0.890 exceeded MAP-Mapper's within-dataset F1-score of 0.880 and closely approached MariNeXt's reported performance of 0.891. Discussion: These findings show that careful data formulation and training design can enable standard architectures to achieve strong cross-domain performance under extreme class imbalance, approaching or even surpassing more specialized models in realistic deployment conditions. The results provide practical guidance for operational marine debris monitoring systems: spatially stratified sampling across diverse marine environments should be prioritized, F1-scores in the range of 0.86--0.89 can be expected when deploying on previously unseen regions without fine-tuning, and a two-stage strategy should be considered in which models are first trained on geographically diverse data and then optionally adapted for region-specific applications. To the best of our knowledge, this is the first systematic cross-dataset validation study involving both MARIDA and MADOS, demonstrating that binary reformulation supports generalization-preserving marine debris detection across geographic and temporal domain shifts.