The field of quantum computing, especially quantum machine learning (QML), has been the subject of much research in recent years. Leveraging the quantum properties of superposition and entanglement promises exponential decrease in computation costs. With the promises of increased speed and accuracy in the quantum paradigm, many classical machine learning algorithms have been adapted to run on quantum computers, typically using a quantum-classical hybrid model. While some work has been done to compare classical and quantum classification algorithms in the Electro-Optical (EO) image domain, this paper will compare the performance of classical and quantum-hybrid classification algorithms in their applications on Synthetic Aperture Radar (SAR) data using the MSTAR dataset. We find that there is no significant difference in classification performance when training with quantum algorithms in ideal simulators as compared to their classical counterparts. However, the true performance benefits will become more apparent as the hardware matures.
The field of quantum computing, especially quantum machine learning (QML), has been the subject of much research in recent years. Leveraging the quantum properties of superposition and entanglement promises exponential decrease in computation costs. With the promises of increased speed and accuracy in the quantum paradigm, many classical machine learning algorithms have been adapted to run on quantum computers, typically using a quantum-classical hybrid model. While some work has been done to compare classical and quantum classification algorithms in the Electro-Optical (EO) image domain, this paper will compare the performance of classical and quantum-hybrid classification algorithms in their applications on Synthetic Aperture Radar (SAR) data using the MSTAR dataset. We find that there is no significant difference in classification performance when training with quantum algorithms in ideal simulators as compared to their classical counterparts. However, the true performance benefits will become more apparent as the hardware matures.
We present in this paper a method to compare scene classification accuracy of C-band Synthetic aperture radar (SAR) and optical images utilizing both classical and quantum computing algorithms. This REU study uses data from the Sentinel satellite. The dataset contains (i) synthetic aperture radar images collected from the Sentinel-1 satellite and (ii) optical images for the same area as the SAR images collected from the Sentinel-2 satellite. We examine classical neural networks to classify four classes of images. We then explore Quantum Convolutional Neural Networks and deep learning techniques in terms of their training and classification performance. A hybrid Quantum-classical model that is trained on the Sentinel1-2 dataset is proposed, and its performance is then compared against the classical model in terms of classification accuracy.
This paper presents algorithms, simulations, and results using machine learning and quantum image fusion algorithms for radar and remote sensing applications. Previous efforts in the classification of synthetic aperture radar (SAR) images using quantum machine learning provided encouraging results but, nevertheless modest accuracy. In this paper, we propose a novel quantum image fusion technique used for identifying and classifying objects obtained from C-band SAR and optical images. More specifically, we design a four-qubit quantum circuit to process the SAR image dataset. This method enhances the spectral details otherwise not seen when using the raw SAR dataset. In addition to the quantum circuit, we design deep neural networks (NN) to improve classification results. The Visual Geometry Group 16 (VGG16), a convolutional neural network that is sixteen layers deep, is customized and used for classification. The merit of quantum fusion as well as the promising results in improving the overall system and the potential of lowering size, weight, power, and cost (SWaP-C) is described.
This paper presents algorithms, simulations, and results on the use of optical and Synthetic Aperture Radar (SAR) quantum image fusion and machine learning algorithms for radar and remote sensing applications. We propose novel quantum image fusion techniques that extract meaningful features for identifying and classifying scenes obtained from the BigEarthNet dataset. Specifically, this work focuses on multi-label scene classification using quantum image fusion of C-band SAR and optical imagery from the BigEarthNet dataset. The proposed fusion techniques enhance and sharpen structural and textural details that are not readily observable in raw SAR imagery, thereby improving downstream classification performance. To enable quantum image fusion, we design new quantum circuits that employ unitary and Pauli gates for phase encoding of pixel intensity values. These circuits allow optical and SAR image information to be encoded and fused within a quantum processing framework. Three fusion strategies are investigated, resulting in three distinct quantum image fusion models. Comparative evaluation of these models demonstrates that pixel-averaging–based quantum fusion yields improved classification accuracy relative to alternative fusion approaches. The fused images are subsequently processed using deep classical neural networks to perform scene classification. Our neural network study begins by customizing the Visual Geometry Group 16 (VGG16) architecture for multi-label remote sensing classification tasks. In addition, we examine the use of a Residual Network 50 (ResNet-50) architecture and provide comparative performance analysis between the two models. The effects of quantum noise on the fusion and classification pipeline are also studied to assess robustness under realistic quantum processing conditions. To evaluate classification performance, we employ standard metrics such as F1 and F2 scores, and additionally introduce false discovery rate and false negative rate metrics to provide a more comprehensive assessment of classification accuracy, particularly in multi-label settings. Overall, the results demonstrate the effectiveness of optical and SAR quantum image fusion in enhancing scene classification performance and highlight the potential of integrating quantum image processing with deep learning for advanced remote sensing applications.
Synthetic Aperture Radar (SAR) data is inherently complex-valued, while quantum machine learning (QML) models operate in complex Hilbert spaces. This similarity suggests that using both the magnitude and phase of SAR data in quantum encoding should help automatic target recognition in SAR images. In this study, we test this assumption by comparing five encoding strategies for quantum models: magnitude-only encoding, joint magnitude-phase encoding, in-phase and quadrature encoding, preprocessed phase encoding, and a purely quantum architecture. All approaches are evaluated under a unified experimental setup on the MSTAR benchmark dataset.
Surprisingly, we find that magnitude-only encoding performs better than phase-inclusive encodings in hybrid quantum-classical models. It achieves 99.57 percent accuracy on the 3-class task and 71.19 percent accuracy on the 8-class task, outperforming complex-valued alternatives under the same framework. Adding phase information provides little or no improvement and can sometimes degrade performance. However, in purely quantum models with only 184 to 224 trainable parameters and no classical neural-network layers, phase information becomes much more important, improving accuracy by up to 21.65 percentage points.
These findings show that the usefulness of phase information depends not only on the data, but also on the architecture used to process it. Hybrid models can compensate for missing phase information through their classical components, while pure quantum models rely more strongly on phase information for class discrimination. The results provide practical guidance for encoding complex-valued data in quantum machine learning and highlight the importance of jointly designing encoding strategies and model architectures for current quantum systems.
Synthetic Aperture Radar (SAR) data, characterised by its high-dimensionality and complex spatial correlations, poses significant challenges in terms of efficient processing and meaningful interpretation. Classical algorithms, while effective, often struggle with the sheer volume and intricacy of the data. This paper introduces a novel approach employing tensor quantum machine learning (QML) to tackle the intricacies of SAR data. By harnessing the computational advantages of quantum mechanics and the representational efficiency of tensor networks, we try to achieve enhanced feature extraction and pattern recognition. We look at various Tensor decomposition schemes to reduce data dimensionality as well as Tensor based quantum circuits to perform land-cover classification. Preliminary results, based on simulations, demonstrate the potential of our tensor QML framework. For the scope of this research so far, we worked with simulated amplitude and phase data, but we will be applying the same for real world data in the future. This interdisciplinary study not only opens avenues for improved SAR data analysis but also enriches the burgeoning field of quantum machine learning by highlighting its applicability in remote sensing domains.
Illegal, unreported, and unregulated (IUU) fishing causes global economic losses of 10-25 billion USD annually and undermines marine sustainability and governance. Synthetic Aperture Radar (SAR) provides reliable maritime surveillance under all weather and lighting conditions, but classifying small maritime objects in SAR imagery remains challenging. We investigate quantum machine learning for this task, focusing on quantum kernel methods (QKMs) applied to real and complex SAR chips extracted from the SARFish dataset. We tackle two binary classification problems, the first for distinguishing vessels from non-vessels, and the second for distinguishing fishing vessels from other types of vessels. We compare QKMs applied to real and complex SAR chips against classical Laplacian, RBF, and linear kernels applied to real SAR chips. We restrict the comparison to be between just kernel based models so that the comparison is as fair and meaningful as possible. Using noiseless numerical simulations of the quantum kernels, we find that with the real SAR chips, QKMs are capable of obtaining equal or better performance than the classical kernels in the best case. However, the specific quantum kernel used to encode the complex SAR data overfits and performs poorly. This work presents the first application of QKMs to maritime classification in SAR imagery and offers insight into the potential and current limitations of quantum-enhanced learning for maritime surveillance.