Hybrid Quantum Deep Learning with Superpixel Encoding for Earth Observation Data Classification
This paper introduces a hybrid quantum deep learning model (SEQNN) that effectively encodes and analyzes EO data for classification tasks. The proposed model utilizes an efficient encoding approach called superpixel encoding, which reduces the quantum resources required for large image representation by incorporating the concept of superpixels. To validate its effectiveness, we conducted evaluations on multiple EO benchmarks, including Overhead-MNIST, So2Sat LCZ42, and SAT-6 datasets. The experimental results suggest the validity of our model for accurate classification of EO data.
The experimental data can be found: