The rapid development of sensor technology has made multi-modal remote sensing data valuable for land cover classification due to its diverse and complementary information. Many feature extraction methods for multi-modal data, combining light detection and ranging (LiDAR) and hyperspectral imaging (HSI), have recognized the importance of incorporating multiple spatial scales. However, effectively capturing both long-range global correlations and short-range local features simultaneously on different scales remains a challenge, particularly in large-scale, complex ground scenes. To address this limitation, we propose a multi-scale graph encoder–decoder network (MGEN) for multi-modal data classification. The MGEN adopts a graph model that maintains global sample correlations to fuse multi-scale features, enabling simultaneous extraction of local and global information. The graph encoder maps multi-modal data from different scales to the graph space and completes feature extraction in the graph space. The graph decoder maps the features of multiple scales back to the original data space and completes multi-scale feature fusion and classification. Experimental results on three HSI-LiDAR datasets demonstrate that the proposed MGEN achieves considerable classification accuracies and outperforms state-of-the-art methods.

To solve the problem of low demodulation accuracy of conventional peak-to-peak algorithm for fiber-optic Fabry-Perot (FP) sensors due to failure of determining the interference order, a novel cavity length sequence matching demodulation algorithm based on a combined valley peak positioning is proposed. Firstly, a pair of a peak and its neighboring valley in the reflection spectrum is selected and positioned, and two groups of interference orders are supposed to generate two groups of cavity length sequences. Finally, these cavity lengths are compared to find the real interference order of the peak and valley for the extraction of the accurate cavity length. In order to verify the feasibility and performance of the proposed algorithm, simulations and experiments were carried out for fiber-optic FP sensors with cavity lengths in the range of 15–115 μm. A demodulation accuracy better than 8.8 nm was found. The proposed algorithm can achieve highly accurate cavity length demodulation of fiber-optic FP sensors.

High-sensitivity interferometers are one of the basic tools for precision measurement, and their sensitivity is limited by their shot noise limit (SNL), which is determined by vacuum fluctuations of the probe field. The quantum interferometer with novel structures can break the SNL and measure the weak signals, such as the direct observation of gravity wave signal. Combining classical interferometers and the optical parametric amplifier (OPA) can enhance the signal; meanwhile, the quantum noise is kept at the vacuum level, so that the sensitivity of the nonlinear interferometer beyond the SNL can be achieved. By analyzing in detail the influence of system parameters on the precision of quantum metrology, including the intensity of optical fields for phase sensing, the gain factor of OPA, and the losses inside and outside the interferometers, the application conditions of high-sensitivity nonlinear quantum interferometers are obtained. Quantum interferometer-based OPAs provide the direct references for the practical development of quantum precise measurement.