Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

IEEE Geoscience and Remote Sensing Letters

Quantum-Enhanced Water Quality Monitoring: Exploiting $\Phi$ Sat-2 Data With Quanvolution

Francesco Mauro, Francesca Razzano, Pietro Di Stasio, Alessandro Sebastianelli, Gabriele Meoni, Gilda Schirinzi, Paolo E. Gamba, Silvia Liberata Ullo

Published Jan 1, 2025• 9 citations
Publisher

Abstract

Abstract—Coastal water quality monitoring is crucial for environmental sustainability and public health. This work introduces a very cutting-edge methodology, using ΦSat-2 multispectral data and quanvolutional neural networks to explore quantum-enhanced machine learning for water contaminant assessment. By integrating quantum preprocessing into a classical regression model, it is possible to achieve a significant reduction in model parameters while maintaining high predictive accuracy. Additionally, this work introduces an innovative dataset that integrates simulated ΦSat-2 spectral data with Copernicus Marine Service bio-geochemical products, ensuring a strong alignment between satellite observations and reference turbidity measurements. Our results show that quantum models use up to 98% fewer parameters than their classical counterparts, while achieving a 6.9% improvement in the Pearson correlation coefficient between the ΦSat-2 pre-processed bands and the ground-truth turbidity values, compared to the case without quantum pre-processing. Additionally, the Root Mean Square Error (RMSE) improves by 7.3% over the classical baseline. These findings highlight the potential of quantum-assisted remote sensing to enable more efficient and scalable analysis of large-scale water contaminant data, paving the way for advanced big data approaches in water quality monitoring.

DOI
10.1109/lgrs.2025.3576677

Related papers

Ground Water Quality Assessment using Quantum Machine Learning

Mohit Malu, Qigang Dai, Xurxo Rigueira +4 more

Groundwater contamination by arsenic, nitrate, and uranium poses significant risks to public health across the United States. Accurate prediction of exceedance of contaminants relative to maximum contaminants levels (MCLs) is critical for risk assessment and monitoring, particularly under sparse and heterogeneous sampling conditions. In this work, we develop a hybrid quantum-classical machine learning framework for classifying contaminant presence above or below certain thresholds (80% MCL) using a large-scale, curated dataset of groundwater observations and environmental covariates. The proposed approach employs dimensionality reduction via Principal Component Analysis (PCA) to select 12–14 informative features, which are subsequently encoded into quantum states using parameterized feature maps (e.g., ZZ and Y-ZZ embeddings). These quantum representations are integrated with classical features and used to train a neural network classifier. We evaluate the hybrid quantum model against classical machine learning baselines on nationwide data. Results demonstrate comparable predictive performance between the quantum-enhanced hybrid models and the classical models. While performance gains are modest, the results indicate that quantum feature embeddings can achieve competitive accuracy while offering a pathway toward richer feature representations and improved modeling of nonlinear environmental interactions. This study provides the first large-scale evaluations of hybrid quantum ML for groundwater quality assessment and establishes a foundation for future quantum-enhanced environmental decision systems.

International Conference on Information, Intelligence, Systems & Applications (IISA)•Jul 1, 2026•0 citations

Quanv4EO: Empowering Earth Observation by Means of Quanvolutional Neural Networks

Alessandro Sebastianelli, Francesco Mauro, Giulia Ciabatti +4 more

A significant amount of remotely sensed data is generated daily by many Earth observation (EO) spaceborne and airborne sensors over different countries of our planet. Different applications use those data, such as natural hazard monitoring, global climate change, urban planning, and more. Many challenges are brought by the use of these big data in the context of remote sensing (RS) applications. In recent years, the employment of machine learning (ML) and deep learning (DL)-based algorithms has allowed a more efficient use of these data, but the issues in managing, processing, and efficiently exploiting them have even increased as classical computers have reached their limits. This article highlights a significant shift toward leveraging quantum computing (QC) techniques in processing large volumes of RS data. The proposed Quanv4EO framework introduces a quanvolution method for (pre)processing multidimensional EO data. Its effectiveness was first demonstrated on standard image classification datasets (MNIST and FashionMNIST), achieving accuracies of 99.84% and 96.81%, respectively, with a significantly reduced model size of 42 k parameters and 16 frozen qubits. Its capabilities were then checked on EO datasets, such as EuroSAT, with a mean accuracy of 96% using balanced iterative reducing and clustering using hierarchies (BIRCHs) clustering and 93% using automated DL (AutoDL), surpassing or matching state-of-the-art (SOTA) classical nonquantum models. Applying the framework to synthetic aperture radar (SAR) data, the QSPeckleFilter demonstrates notable improvements in speckle noise reduction, achieving a peak signal-to-noise ratio (PSNR) of 21.72 and a structural similarity index measure (SSIM) of 0.81, surpassing all tested classical counterparts. The proposed results underscore the potential of quantum-enhanced approaches in RS data analysis, paving the way for more efficient and effective solutions for wide geographical area EO data exploitation.

arXiv:2407.17108•IEEE Transactions on Geoscience and Remote Sensing•Jan 1, 2025•26 citations

Quantum-Inspired Feature Extraction for Earth Observation: The Quanv4eo Framework

Francesco Mauro, Alessandro Sebastianelli, Paolo E. Gamba +1 more

The exponential growth of Earth Observation (EO) data challenges the scalability of traditional Deep Learning (DL) models due to their high computational and storage demands. To address this issue, Quanv4EO is presented as a quantum-inspired framework that leverages simulated quantum circuits for efficient feature extraction. The framework is validated across diverse operational scenarios, including land-use classification (Sentinel-2), building segmentation (Sentinel-1), speckle filtering (Sentinel-1), and turbidity regression (ΦSat-2). Experimental results demonstrate extreme model compression, with parameter reductions ranging from 94% to 99.8%, while maintaining near-baseline performance in classification and segmentation tasks and improving accuracy in turbidity regression and SAR speckle filtering. These findings highlight the potential of hybrid quantum–classical pipelines for next-generation, resource-constrained remote sensing analytics.

IEEE International Geoscience and Remote Sensing Symposium (IGARSS)•Aug 1, 2026•0 citations

Advancing Earth Observation with Trainable Quanvolutional Neural Networks for Classification Tasks

Francesco Mauro, Francesca De Falco, Andrea Ceschini +5 more

The growing complexity of Earth observation data demands innovative approaches for efficient processing and analysis. Quantum convolutional neural networks integrate quantum computing principles into classical deep learning architectures, offering enhanced feature extraction capabilities. Building on the Quanv4EO framework, this study introduces a trainable quanvolutional layer within a hybrid neural network, enabling dynamic parameter optimization during training. Experiments using the EuroSAT dataset demonstrate that the trainable quanvolution surpasses both its non-trainable quantum and the classical counterparts, achieving a higher accuracy value with fewer parameters. These findings underscore the potential of trainable quantum models for earth observation applications.

IEEE International Geoscience and Remote Sensing Symposium (IGARSS)•Aug 3, 2025•2 citations

AI-DRIVEN QUANTUM SENSING FOR HIGH-PRECISION ENVIRONMENTAL MONITORING

Narayana Gaddam

On the integration of artificial intelligence with quantum sensing technologies for environmental monitoring, these make it much more precise, efficient and real time information end-user.In order to combat climate change markers, insulation effectiveness, and environmental hazard immediate response systems, this research dictates an AI based quantum sensing framework.Quantum augmented machine learning methods, i.e., Quantum Bayesian Networks and Quanvolutional Neural Networks, are used to improve accuracy of measurements from sensors and furthermore sensor calibration.In addition, the proposed system incorporates Q-SCALE for advanced sensor calibration to eliminate noise and improve sensor performance.The Quantum Wireless Sensing framework, in the form of quantum-based sensors, has an advantage in sensing dynamic environmental factors with an increased sensitivity.Some key findings include that AI quantum AI sensors are better than traditional sensors at alerting to abrupt environmental shifts, monitoring pollutants, testing greenhouse conditions, etc.Quantum entanglement was not used, and quantum sensors also achieved secure data transmission of distances beyond 50 kilometers.The proposed system achieves real-time data acquisition using lightweight models optimized for such resource constrained environments and reduces both the energy consumption.This major step in research will lead towards development of more accurate as well as scalable, sustainable environmental monitoring solutions with enhanced data security.As such, recent innovations such as AI coupled with nanomaterials and IoT-instigated quantum surveillance systems indicate the extent of this field's capability.

•Jan 18, 2025•1 citation

Comparative Analysis of Neural Network and Quantum Neural Network for Enhanced Air Quality Forecasting

Noha A. Sobhi, Doaa A. Shoieb, Sherin Moustafa Youssef

Air pollution forecasting is a critical challenge in environmental science. Nitrogen Dioxide (NO2) and Carbon Monoxide (CO) are two of the most harmful gases, contributing significantly to respiratory illnesses and environmental degradation. Accurate predictions of NO2and CO levels help mitigate the adverse effects of pollutants on public health. A hybrid quantum-classical machine learning model is developed, integrating a feature map, variational ansatz, and quantum regressor optimized with multiple optimizers. The results are compared with classical machine learning model, such as Multi-Layer Perceptron (MLP), demonstrating the potential of quantum computing in predictive modeling. Experimental results using the air quality dataset reveal that the proposed model achieves superior performance compared to traditional approaches, as measured by MSE, RMSE, MAE, and R2. The findings suggest that quantum-enhanced neural networks provide a promising avenue for real-time air pollution forecasting, enabling proactive pollution control measures.

International Conference on Machine Intelligence and Smart Innovation (ICMISI)•May 10, 2025•2 citations
Quantum-Enhanced Water Quality Monitoring: Exploiting Φ\PhiΦ Sat-2 Data With Quanvolution
Abstract—Coastal water quality monitoring is crucial for environmental sustainability and public health. This work introduces a very cutting-edge methodology, using ΦSat-2 multispectral data and quanvolutional neural networks to explore quantum-enhanced machine learning for water contaminant assessment. By integrating quantum preprocessing into a classical regression model, it is possible to achieve a significant reduction in model parameters while maintaining high predictive accuracy. Additionally, this work introduces an innovative dataset that integrates simulated ΦSat-2 spectral data with Copernicus Marine Service bio-geochemical products, ensuring a strong alignment between satellite observations and reference turbidity measurements. Our results show that quantum models use up to 98% fewer parameters than their classical counterparts, while achieving a 6.9% improvement in the Pearson correlation coefficient between the ΦSat-2 pre-processed bands and the ground-truth turbidity values, compared to the case without quantum pre-processing. Additionally, the Root Mean Square Error (RMSE) improves by 7.3% over the classical baseline. These findings highlight the potential of quantum-assisted remote sensing to enable more efficient and scalable analysis of large-scale water contaminant data, paving the way for advanced big data approaches in water quality monitoring.
Groundwater contamination by arsenic, nitrate, and uranium poses significant risks to public health across the United States. Accurate prediction of exceedance of contaminants relative to maximum contaminants levels (MCLs) is critical for risk assessment and monitoring, particularly under sparse and heterogeneous sampling conditions. In this work, we develop a hybrid quantum-classical machine learning framework for classifying contaminant presence above or below certain thresholds (80% MCL) using a large-scale, curated dataset of groundwater observations and environmental covariates. The proposed approach employs dimensionality reduction via Principal Component Analysis (PCA) to select 12–14 informative features, which are subsequently encoded into quantum states using parameterized feature maps (e.g., ZZ and Y-ZZ embeddings). These quantum representations are integrated with classical features and used to train a neural network classifier. We evaluate the hybrid quantum model against classical machine learning baselines on nationwide data. Results demonstrate comparable predictive performance between the quantum-enhanced hybrid models and the classical models. While performance gains are modest, the results indicate that quantum feature embeddings can achieve competitive accuracy while offering a pathway toward richer feature representations and improved modeling of nonlinear environmental interactions. This study provides the first large-scale evaluations of hybrid quantum ML for groundwater quality assessment and establishes a foundation for future quantum-enhanced environmental decision systems.
A significant amount of remotely sensed data is generated daily by many Earth observation (EO) spaceborne and airborne sensors over different countries of our planet. Different applications use those data, such as natural hazard monitoring, global climate change, urban planning, and more. Many challenges are brought by the use of these big data in the context of remote sensing (RS) applications. In recent years, the employment of machine learning (ML) and deep learning (DL)-based algorithms has allowed a more efficient use of these data, but the issues in managing, processing, and efficiently exploiting them have even increased as classical computers have reached their limits. This article highlights a significant shift toward leveraging quantum computing (QC) techniques in processing large volumes of RS data. The proposed Quanv4EO framework introduces a quanvolution method for (pre)processing multidimensional EO data. Its effectiveness was first demonstrated on standard image classification datasets (MNIST and FashionMNIST), achieving accuracies of 99.84% and 96.81%, respectively, with a significantly reduced model size of 42 k parameters and 16 frozen qubits. Its capabilities were then checked on EO datasets, such as EuroSAT, with a mean accuracy of 96% using balanced iterative reducing and clustering using hierarchies (BIRCHs) clustering and 93% using automated DL (AutoDL), surpassing or matching state-of-the-art (SOTA) classical nonquantum models. Applying the framework to synthetic aperture radar (SAR) data, the QSPeckleFilter demonstrates notable improvements in speckle noise reduction, achieving a peak signal-to-noise ratio (PSNR) of 21.72 and a structural similarity index measure (SSIM) of 0.81, surpassing all tested classical counterparts. The proposed results underscore the potential of quantum-enhanced approaches in RS data analysis, paving the way for more efficient and effective solutions for wide geographical area EO data exploitation.
The exponential growth of Earth Observation (EO) data challenges the scalability of traditional Deep Learning (DL) models due to their high computational and storage demands. To address this issue, Quanv4EO is presented as a quantum-inspired framework that leverages simulated quantum circuits for efficient feature extraction. The framework is validated across diverse operational scenarios, including land-use classification (Sentinel-2), building segmentation (Sentinel-1), speckle filtering (Sentinel-1), and turbidity regression (ΦSat-2). Experimental results demonstrate extreme model compression, with parameter reductions ranging from 94% to 99.8%, while maintaining near-baseline performance in classification and segmentation tasks and improving accuracy in turbidity regression and SAR speckle filtering. These findings highlight the potential of hybrid quantum–classical pipelines for next-generation, resource-constrained remote sensing analytics.
The growing complexity of Earth observation data demands innovative approaches for efficient processing and analysis. Quantum convolutional neural networks integrate quantum computing principles into classical deep learning architectures, offering enhanced feature extraction capabilities. Building on the Quanv4EO framework, this study introduces a trainable quanvolutional layer within a hybrid neural network, enabling dynamic parameter optimization during training. Experiments using the EuroSAT dataset demonstrate that the trainable quanvolution surpasses both its non-trainable quantum and the classical counterparts, achieving a higher accuracy value with fewer parameters. These findings underscore the potential of trainable quantum models for earth observation applications.
On the integration of artificial intelligence with quantum sensing technologies for environmental monitoring, these make it much more precise, efficient and real time information end-user.In order to combat climate change markers, insulation effectiveness, and environmental hazard immediate response systems, this research dictates an AI based quantum sensing framework.Quantum augmented machine learning methods, i.e., Quantum Bayesian Networks and Quanvolutional Neural Networks, are used to improve accuracy of measurements from sensors and furthermore sensor calibration.In addition, the proposed system incorporates Q-SCALE for advanced sensor calibration to eliminate noise and improve sensor performance.The Quantum Wireless Sensing framework, in the form of quantum-based sensors, has an advantage in sensing dynamic environmental factors with an increased sensitivity.Some key findings include that AI quantum AI sensors are better than traditional sensors at alerting to abrupt environmental shifts, monitoring pollutants, testing greenhouse conditions, etc.Quantum entanglement was not used, and quantum sensors also achieved secure data transmission of distances beyond 50 kilometers.The proposed system achieves real-time data acquisition using lightweight models optimized for such resource constrained environments and reduces both the energy consumption.This major step in research will lead towards development of more accurate as well as scalable, sustainable environmental monitoring solutions with enhanced data security.As such, recent innovations such as AI coupled with nanomaterials and IoT-instigated quantum surveillance systems indicate the extent of this field's capability.
Air pollution forecasting is a critical challenge in environmental science. Nitrogen Dioxide (NO2) and Carbon Monoxide (CO) are two of the most harmful gases, contributing significantly to respiratory illnesses and environmental degradation. Accurate predictions of NO2and CO levels help mitigate the adverse effects of pollutants on public health. A hybrid quantum-classical machine learning model is developed, integrating a feature map, variational ansatz, and quantum regressor optimized with multiple optimizers. The results are compared with classical machine learning model, such as Multi-Layer Perceptron (MLP), demonstrating the potential of quantum computing in predictive modeling. Experimental results using the air quality dataset reveal that the proposed model achieves superior performance compared to traditional approaches, as measured by MSE, RMSE, MAE, and R2. The findings suggest that quantum-enhanced neural networks provide a promising avenue for real-time air pollution forecasting, enabling proactive pollution control measures.