Mingzhong Pan

Hyperspectral imaging acquires rich scene information by capturing hundreds of contiguous bands across ultra-violet, visible, infrared, and other spectral ranges, making it valuable for remote sensing and material analysis. However, the resulting data volumes demand efficient compression to reduce storage and transmission costs while preserving reconstruction quality. Implicit neural representations (INRs) can represent HSIs compactly but struggle to model their spatial locality, often incurring high GPU memory cost and slow encoding. LoR-SGS addresses these limitations by coupling low-rank spectral decomposition with 2-D Gaussian splatting, yet two performance-critical bottlenecks remain. First, Gaussian rasterization introduces local coefficient inconsistency across neighboring pixels. Second, the fixed NMF-initialized spectral basis causes spectral mismatch that cannot be corrected during optimization. To address these two bottlenecks, we propose LoR-SGS++, which introduces lightweight, zero-initialized correction modules. Spatial Coefficient Refinement (SCR) applies a convolutional residual to the rasterized coefficient map to improve local spatial consistency, while Low-Rank Endmember Calibration (LEC) introduces a scene-adaptive low-rank update to the spectral basis. Both modules are gated by learnable scalars initialized to zero, so that the codec starts from the baseline behavior and activates corrections only as optimization requires. Evaluations on four benchmark datasets show that LoR-SGS++ achieves consistently strong performance in PSNR, MS-SSIM, and SAM across all tested scenes. At matched bitrates, the proposed method surpasses LoR-SGS by over 3 dB in PSNR on the most challenging scene while adding fewer than 0.001 M parameters.

To address asynchronous acquisition, heterogeneous data dimensions, and the difficulty of directly applying multi-source measurements to fusion modeling in online water quality monitoring, this paper presents a multimodal water quality detection system for data fusion. The system uses a 940 nm light-emitting diode (LED) near-infrared scattering channel for turbidity and particulate-matter characterization, and integrates 200–1000 nm broadband spectral detection, pH detection, and electrical conductivity detection. An embedded node performs data buffering and preliminary processing, while PC-based software manages calibration, spectral preprocessing, timestamp recording, and nearest-neighbor temporal matching. The resulting multimodal data frames combine scalar electrochemical and scattering features with high-dimensional spectral features under explicit temporal correspondence. Experimental results show that the pH, electrical conductivity, and near-infrared scattering turbidity modules provide calibratable and repeatable scalar outputs, while the spectral module outputs transmittance and absorbance features. In an asynchronous alignment experiment using the electrical conductivity stream as the reference, the system matched 77 out of 77 frames, achieving a 100% alignment success rate and a mean maximum temporal deviation of 0.437 s. A 72-sample preliminary fusion verification further indicates that adding pH and suspended-solids-related auxiliary information to spectral features reduces the mean five-fold cross-validation relative error for all tested indicators. These results demonstrate that the proposed system provides a practical data acquisition foundation for optoelectronic water quality detection, spectral compensation, and multimodal data fusion.

Cross-scene hyperspectral band selection must reduce spectral redundancy while retaining channels that remain useful beyond the source scene. We propose Adaptive Band Gating (ABG), a source-only selector that combines frequency-domain decoupling enhancement, global and sample-specific gating, source-side spectral, spatial, morphology-inspired, and sensor-noise perturbations, and a dual-head evaluator. The selector is trained with source data and frozen before downstream evaluation. Selected bands are assessed with a radial-basis-function support vector machine on Pavia Center and HyRANK under fixed band budgets and target-label fractions from 0% to 10%. At 5% target labels, 15 selected bands achieve 96.05% overall accuracy on Pavia Center, compared with 95.41% using all 102 bands; 20 selected bands achieve 81.49% on HyRANK, compared with 79.20% using all 176 bands. Across evaluated band budgets, ABG is comparable to XGBS on Pavia Center and provides stronger results on HyRANK. Ablation experiments show that adaptive gating, frequency-domain enhancement, and source-side expansion each contribute to performance. Together, these results demonstrate that ABG learns compact and traceable original-band subsets with strong downstream transfer utility across the evaluated cross-scene settings.

Traditional hyperspectral cameras transmit full data cubes to host computers, creating severe bandwidth and storage bottlenecks that impede real-time analysis. We present a Zynq-7035-based intelligent camera using hardware–software codesign to enable on-board processing and transmit only actionable results. This intelligent camera is designed for high-throughput edge-sensing tasks, prioritizing rapid detection and information extraction over exhaustive raw data acquisition. The processing system (PS) handles command scheduling while the programmable logic (PL) implements a row-parallel pipeline for image acquisition, preprocessing, and spectral matching; all modules are decoupled through a unified DDR3 interface to support flexible algorithm integration. Push-broom experiments on leaf samples demonstrate Euclidean distance-based spectral matching executed entirely within the camera. Raw data and classification maps are uploaded via User Datagram Protocol (UDP). Results confirm accurate identification of diseased regions with two orders of magnitude data reduction, validating the architecture for real-time hyperspectral processing.

Brain tumor is a common neurological surgical disease, where surgical resection is the primary treatment method. Neurosurgeons need to accurately determine the location of the tumor during tumor resection surgery, but existing clinical tumor identification technologies face numerous challenges, such as high equipment costs, long processing times, a certain degree of invasiveness, and insufficient image clarity. In this work, we propose a hyperspectral image detection algorithm based on the fusion of multiple features to maximize the determination of tumor boundaries. The algorithm establishes the machine learning models of Support Vector Machine (SVM) and Random Forest (RF) by integrating data features from optimal wavelengths, spectral indices, and textural features. Experimental results show that on different datasets, the classification accuracy of the three-feature fusion model is significantly higher than that of models using only two features or a single feature. Hyperspectral tumor image recognition can effectively help distinguish the tumors from the surrounding tissue, thereby enhancing the safety and thoroughness of tumor surgery.

The primary measurement involves detecting tiny (picometer-level) pathlength fluctuations between satellites using heterodyne laser interferometry for space-based gravitational wave detection. The interference of two laser beams with a MHz-level frequency difference produces a MHz beat-note, in which the gravitational wave signal is encoded in the phase of the beat-note. The phasemeter then performs micro-radian accuracy phase measurement and communication information demodulation for this beat-note. To mitigate the impact of phase modulation, existing solutions mostly alleviate it by reducing the modulation depth and optimizing the structure of the pseudo-random noise (PRN) codes. Since the phase modulation is not effectively separated from the phase of the beat-note phase measurement, it has a potential impact on the phase extraction of the micro-radian accuracy of the beat-note. To solve this problem, this paper analyzes the influence mechanism of phase modulation on beat-note phase measurement and proposes a method to separate the modulated phase based on complex rotation. The beat-note is processed by complex conjugate rotation, which can effectively eliminate the PRN modulated phase. Simulation and analysis results demonstrate that this method can significantly enhance the purity of the measured phase in the beat-note while maintaining the ranging and communication functions. Targeting the application of the micro-radian phasemeter in space-based gravitational wave detection, this study presents the reconstruction and separation method of the ranging and communication phase in beat-note, which also provides a new direction for the final selection of modulation depth in the future.