Cloud Detection in Multispectral Satellite Images Using Support Vector Machines With Quantum Kernels

TL;DR

This paper introduces a hybrid quantum-classical SVM approach with quantum kernels for cloud detection in multispectral satellite images, demonstrating comparable accuracy to classic SVMs on Landsat-8 data.

Contribution

It presents the design and implementation of quantum kernel-enhanced SVMs for satellite image analysis, a novel application of quantum machine learning in remote sensing.

Findings

Hybrid SVMs achieve accuracy comparable to classic SVMs.

Quantum kernels effectively map satellite data to Hilbert space.

The approach is feasible for cloud detection in multispectral images.

Abstract

Support vector machines (SVMs) are a well-established classifier effectively deployed in an array of pattern recognition and classification tasks. In this work, we consider extending classic SVMs with quantum kernels and applying them to satellite data analysis. The design and implementation of SVMs with quantum kernels (hybrid SVMs) is presented. It consists of the Quantum Kernel Estimation (QKE) procedure combined with a classic SVM training routine. The pixel data are mapped to the Hilbert space using ZZ-feature maps acting on the parameterized ansatz state. The parameters are optimized to maximize the kernel target alignment. We approach the problem of cloud detection in satellite image data, which is one of the pivotal steps in both on-the-ground and on-board satellite image analysis processing chains. The experiments performed over the benchmark Landsat-8 multispectral dataset revealed that the simulated hybrid SVM successfully classifies satellite images with accuracy on par with classic SVMs.