The field of multi-source remote sensing observation is becoming increasingly dynamic through the integration of various remote sensing data sources. However, existing deep learning methods face challenges in differentiating between internal and external relationships and capturing fine spatial features. These models often struggle to effectively capture comprehensive information across remote sensing data bands, and they have inherent differences in the size, structure, and physical properties of different remote sensing datasets. To address these challenges, this paper proposes a novel geometric-algebra-based spectral–spatial hierarchical fusion network (GASSF-Net), which uses geometric algebra for the first time to process multi-source remote sensing images, enabling a more holistic approach to handling these images by simultaneously leveraging the real and imaginary components of geometric algebra to express structural information. This method captures the internal and external relationships between remote sensing image features and spatial information, effectively fusing the features of different remote sensing data to improve classification accuracy. GASSF-Net uses geometric algebra (GA) to represent pixels from different bands as multivectors, thus capturing the intrinsic relationships between spectral bands while preserving spatial information. The network begins by deeply mining the spectral–spatial features of a hyperspectral image (HSI) using pairwise covariance operators. These features are then extracted through two branches: a geometric-algebra-based branch and a real-valued network branch. Additionally, the geometric-algebra-based network extracts spatial information from light detection and ranging (LiDAR) to complement the elevation data lacking in the HSI. Finally, a genetic-algorithm-based cross-fusion module is introduced to fuse the HSI and LiDAR data for improved classification. Experiments conducted on three well-known datasets, Trento, MUUFL, and Houston, demonstrate that GASSF-Net significantly outperforms traditional methods in terms of classification accuracy and model efficiency.

Hyperspectral anomaly detection (HAD) is of great significance in remote sensing by identifying spectrally distinct pixels as anomalies without prior information, while the majority of pixels with similar spectral characteristics are classified as background. While existing approaches combine the strengths of deep learning (DL) in feature extraction and background suppression with the effectiveness of low-rank representation (LRR) in background modeling, they fail to fully exploit rich spatial information and long-range spectral dependencies due to the limitations of conventional convolutional operations. To address this issue, we propose a geometric algebra (GA)-based deep neural network for HAD, named GA-HAD. The network constructs an encoder-decoder architecture using GA convolutional layers to extract spectral-spatial features, capturing multidimensional spatial information while preserving the intricate spectral characteristics of hyperspectral images (HSIs). A specialized reconstruction error is designed to train the GA feature extraction network with high efficiency and accuracy. The encoder features are integrated with the LRR algorithm and a Gaussian mixture model (GMM)-based dictionary to improve background modeling and robustness. Finally, the detection maps output by the low-rank detection module are refined through an edge-preserving filter to produce the final detection result. Extensive experiments on four datasets demonstrate the superiority of our proposed GA-HAD framework in overall detection effect compared to thirteen representative baselines, implying our potential application in the field of HAD.

Although the multiresolution analysis (MRA)-based pansharpening methods are able to generate high-resolution multispectral (HRMS) images with good spectral retention, they are prone to spatial distortion. To address this problem, this letter combines deep learning (DL) with the MRA methods and proposes a novel detail-injected-based multiscale asymmetric residual network. The difference strategy is combined with MRA methods and the corresponding injection coefficients are obtained using the multiscale residual block (MSRB) to effectively map spatial information to each waveband of the multispectral (MS) images. In addition, asymmetric convolution block (ACB) is embedded in the residual network to obtain more robust features, and an inception feature pyramid network (FPN) is designed to enrich the spatial information of the fusion results while fusing features at different levels. Experimental results show that the method proposed in this letter outperforms the state-of-the-art pansharpening methods.

Satellite remote sensing, with its wide coverage, long time series, and high revisit frequency, has become an indispensable tool for CO₂ monitoring. However, mainstream XCO₂ datasets derived from satellite observations typically feature low spatial resolution, which limits the practical applications of these valuable satellite data. To overcome this limitation, we propose a multifactor Deep Forest Regression Stepwise Downscaling (DFRSD) Framework using the existing global monthly and Gap-Free column-averaged dry-air mole fraction of CO₂ (GF-XCO₂) resolution of 0.1°. This approach introduces an intermediate resolution level (0.05°) between the initial resolution (0.1°) and target resolution (1km). Specifically, column-averaged dry-air mole fraction of CO₂ (XCO₂) at resolution of 0.1° is first downscaled to that of 0.05°, and then further refined to 1 km resolution. The Deep Forest Regression (DFR) model is used to establish relationships between XCO2 and auxiliary variables at each downscaling step, including near-surface air pollutant, Normalized Difference Vegetation Index (NDVI), Temperature (TMP), and Precipitation (PRE). We successfully enhance the XCO₂ spatial resolution of 0.1° to that of 1 km, generating high-resolution monthly spatial resolution products of 1 km for the period from January 2016 to December 2020. To validate the accuracy of the reconstructed high-resolution XCO₂, we compare it with field measurements from ground monitoring stations. The results demonstrate a strong agreement between the downscaled XCO₂ and field observations, with R² of 0.912 and RMSE of 1.10ppm, highlighting its reliability and precision. The results demonstrate the effectiveness of using SRF and Multifactor methods for XCO₂ downscaling. The proposed method not only significantly enhances spatial resolution but also preserves spatial distribution integrity, which satisfies the precision requirements of various research applications.