README.md

Quantum Support Vector Machine Algorithms for Remote Sensing Data Classification

General information

🗃️ This repository contains Python functions and processing pipelines documented in Jupyter notebook for pixel-wise binary classification of remote sensing multispectral images with the D-Wave Advantage quantum annealer.

Current publication

More information can be found in the conference paper connected to this repository

📜 Amer Delilbasic, Gabriele Cavallaro, Madita Willsch, Farid Melgani, Morris Riedel and Kristel Michielsen, “Quantum Support Vector Machine Algorithms for Remote Sensing Data Classification”, in Proceedings of the IEEE International Geoscience and Remote Sensing Symposium (IGARSS), 2021 (accepted).

Recent developments in Quantum Computing (QC) have paved the way for an enhancement of computing capabilities. Quantum Machine Learning (QML) aims at developing Machine Learning (ML) models specifically designed for quantum computers. The availability of the first quantum processors enabled further research, in particular the exploration of possible practical applications of QML algorithms. In this work, quantum formulations of the Support Vector Machine (SVM) are presented. Then, their implementation using existing quantum technologies is discussed and Remote Sensing (RS) image classification is considered for evaluation.

Previous publications

📃 D. Willsch, M. Willsch, H. De Raedt, and K. Michielsen, “Support Vector Machines on the D-Wave Quantum Annealer” in Computer Physics Communications, vol. 248, 2020, https://doi.org/10.1016/j.cpc.2019.107006

📃 G. Cavallaro, D. Willsch, M. Willsch, K. Michielsen, and M. Riedel, “Approaching Remote Sensing Image Classification with Ensembles of Support Vector Machines on the D-Wave Quantum Annealer,” in Proceedings of the IEEE International Geoscience and Remote Sensing Symposium (IGARSS), pp. 1973-1976, 2020, https://doi.org/10.1109/IGARSS39084.2020.9323544

D-Wave Leap

👌Everyone can make a free account to run on the D-Wave Advantage quantum annealer:

• Make a free account to run on the D-Wave through 👉 (https://www.dwavesys.com/take-leap

• Install Ocean Software with 'pip install dwave-ocean-sdk' 👉 https://docs.ocean.dwavesys.com/en/latest/overview/install.html

• Configuring the D-Wave System as a Solver with 'dwave config create' 👉 https://docs.ocean.dwavesys.com/en/stable/overview/sapi.html

Experiments

Praparation of the binary classification problem

The binary classification problem is constructed from the SemCity Toulouse multispectral benchmark data set, that is publicly available 👉 https://doi.org/10.5194/isprs-annals-V-5-2020-109-2020. More information about the dataset can be found in the publication below

R. Roscher, M. Volpi, C. Mallet, L. Drees, and J. D. Wegner, “Semcity toulouse: a benchmark for building instance segmentation in satellite images,” ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, vol. V-5-2020, p. 109–116, 2020.

The processing workflow to build the classification problem is in this Jupyter Notebook 👉 Build_train_and_test_sets.ipynb

The specific training and test sets that we used for the experiments in the paper are in the folder 👉 experiments

📐 Now you can proceed in two was: