Quantum computers with hundreds of noisy qubits are already available for the research community. They have the potential to run complex quantum computations well beyond the computational capacity of any classical device. It is natural to ask the question, what application these devices could be useful for? Land Use and Land Cover classification of multispectral Earth observation data collected from the earth observation satellite mission is one such problem that is hard for classical methods due to its unique characteristics. In this work, we compare the performance of several classical machine learning algorithms on the stilted re-labeled dataset of the Copernicus Sentinel-2 mission, when the algorithm has access to Projected Quantum Kernel (PQK) features. We show that the classification accuracy increases drastically when the model has access to PQK features. We then naively study the performance of these algorithms with and without access to PQK features on the original Copernicus Sentinel-2 mission data set. This study provides key evidence that shows the potential of quantum machine learning methods for Earth Observation data.
Quantum computers with hundreds of noisy qubits are already available for the research community. They have the potential to run complex quantum computations well beyond the computational capacity of any classical device. It is natural to ask the question, what application these devices could be useful for? Land Use and Land Cover classification of multispectral Earth observation data collected from the earth observation satellite mission is one such problem that is hard for classical methods due to its unique characteristics. In this work, we compare the performance of several classical machine learning algorithms on the stilted re-labeled dataset of the Copernicus Sentinel-2 mission, when the algorithm has access to Projected Quantum Kernel (PQK) features. We show that the classification accuracy increases drastically when the model has access to PQK features. We then naively study the performance of these algorithms with and without access to PQK features on the original Copernicus Sentinel-2 mission data set. This study provides key evidence that shows the potential of quantum machine learning methods for Earth Observation data.
Quantum computers consisting of hundreds of noisy qubits are already available and can run specific quantum algorithms although a large-scale fully error-corrected quantum computer is decades away. It is important to study their application to real-life computational problems. One such problem is Land Use and Land Cover classification of Earth Observation data set collected from the earth observation satellite mission using quantum machine learning methods. In this work, we compare the performance of the classical neural network on the re-labeled dataset of the Copernicus Sentinel-2 mission, when the model has access to Projected Quantum Kernel features. We show that classical neural net-work training accuracy increases drastically when the model has access to Projected Quantum Kernel features. This study shows the potential for quantum machine learning methods to Earth Observation data and provides key evidence for further investigation.
Exploiting machine learning techniques to automatically classify multispectral remote sensing imagery plays a significant role in deriving changes on the Earth’s surface. However, the computation power required to manage large Earth observation data and apply sophisticated machine learning models for this analysis purpose has become an intractable bottleneck. Leveraging quantum computing provides a possibility to tackle this challenge in the future. This article focuses on land cover classification by analyzing Sentinel-2 images with quantum computing. Two hybrid quantum-classical deep learning frameworks are proposed. Both models exploit quantum computing to extract features efficiently from multispectral images and classical computing for final classification. As proof of concept, numerical simulation results on the LCZ42 dataset through the TensorFlow Quantum platform verify our models' validity. The experiments indicate that our models can extract features more effectively compared with their classical counterparts, specifically, the convolutional neural network (CNN) model. Our models demonstrated improvements, with an average test accuracy increase of 4.5% and 3.3%, respectively, in comparison to the CNN model. In addition, our proposed models exhibit better transferability and robustness than CNN models.
Land cover classification using remote sensing data is essential for environmental monitoring and agricultural management. This study evaluates the feasibility of applying a Variational Quantum Classifier (VQC) to distinguish soybean fields from forest formations using Sentinel-2 imagery. Spectral bands B2, B3, B4 (10 m resolution) and annual mean NDVI were used as input features. The quantum model, implemented in Qiskit’s simulator, was compared with a classical neural network trained on the same dataset. While the classical model achieved higher accuracy (above 90% for both classes), the quantum classifier demonstrated competitive performance, particularly for forest identification. The results highlight current limitations of quantum classifiers in complex agricultural landscapes, while indicating future potential as quantum hardware and algorithms evolve.
Exploiting deep learning techniques to automatically analyze multi-spectral remote sensing imagery plays an essential role in urban land cover and land use classification. However, the computation power required to analyze large earth observation data with complex machine learning models for this task becomes an intractable bottleneck. Leveraging quantum computing might tackle this challenge. In this paper, we present two hybrid quantum-classical deep learning frameworks. They both exploit quantum computing to extract features from multi-spectral images efficiently and classical computing for final classification. The effectiveness of our models is verified with the LCZ42 dataset through the TensorFlow Quantum platform.
If you read that quantum machine learning applications solve some traditional machine learning problems at an amazing speed, be sure to check if they return quantum results. Quantum outputs, such as magnitude-encoded vectors, limit the use of applications and require additional specifications on how to extract practical and useful results. In satellite images, degradation of image contrast and color quality is a common problem, which brings great difficulties for information extraction and visibility of image objects. This is due to atmospheric conditions, which can cause a loss of color and contrast in the image. Regardless of the band, image enhancement plays a vital role in presenting details more clearly. The research includes new computational algorithms that accelerate multispectral satellite imagery to improve land cover mapping and study feature extraction. The first method is a quantum Fourier transform algorithm based on quantum computing, and the second method is a parallel modified data algorithm based on parallel computing. These calculation algorithms are applied to multispectral images for improvement. The quantum-processed image is compared with the original image, and better results are obtained in terms of visual interpretation of the extracted information. Quantitatively evaluate the performance of the improved method and evaluate the applicability of the improved technology to different land types.
Quantum computing (QC) promises to revolutionize data-intensive fields due to its potential for exponential speedups and novel computational paradigms. Satellite image processing (also called Earth Observation, EO) is one such domain, where large volumes of high-dimensional data (e.g., multispectral or hyperspectral images) challenge classical methods in terms of computation, memory, and latency. In this review, we survey the state-of-the-art at the intersection of quantum computing and satellite image processing. We first outline quantum image representations, key quantum algorithms proposed for image processing (e.g., edge detection, segmentation, classification), and quantum machine learning (QML) models. We then examine recent studies that target Earth observation data, including resource estimation and hybrid quantum-classical architectures. Next, we delineate the major challenges (hardware, encoding, noise, scalability) and propose potential solutions, focusing on suitable quantum algorithms and hybrid strategies. Finally, we discuss future directions and research opportunities.