Quantum Enchanced Multi-Scale CNN with Bi-directional Mamba for Crop Field Analysis
Researchers have developed a novel framework for crop field analysis using hyperspectral imagery, combining a multi-scale Convolutional Neural Network (CNN) with a Bi-directional Mamba module. This approach enhances spatial-spectral feature learning by fusing information across different resolutions and modeling long-range dependencies in the spectral data. The framework also incorporates spectral attention and quantum-inspired learning to improve accuracy and address challenges like class imbalance and limited labeled samples. Experiments on the UAVHSI-Crop dataset showed the method achieved an 84.83% overall accuracy, demonstrating its potential for various remote sensing applications. AI