To address the limitation of traditional deep learning algorithms in fully utilizing contextual information in multispectral remote sensing (RS) images, this paper proposes an improved vegetation cover classification algorithm called LO-MLPRNN, which integrates Large Selective Kernel Network (LSK) and Omni-Dimensional Dynamic Convolution (ODC) with a Multi-Layer Perceptron Recurrent Neural Network (MLPRNN). The algorithm employs parallel-connected ODC and LSK modules to adaptively adjust convolution kernel parameters across multiple dimensions and dynamically optimize spatial receptive fields, enabling multi-perspective feature fusion for efficient processing of multispectral band information. The extracted features are mapped to a high-dimensional space through a Gate Recurrent Unit (GRU) and fully connected layers, with nonlinear characteristics enhanced by activation functions, ultimately achieving pixel-level land cover classification. Experiments conducted on GF-2 (0.75 m) and Sentinel-2 (10 m) multispectral RS images from Liucheng County, Liuzhou City, Guangxi Province, demonstrate that LO-MLPRNN achieves overall accuracies of 99.11% and 99.43%, outperforming Vision Transformer (ViT) by 2.61% and 3.98%, respectively. Notably, the classification accuracy for sugarcane reaches 99.70% and 99.67%, showcasing its superior performance.
Due to the impact of global warming, drought has caused serious damage to China’s ecological environment and social status. This article selects Laibin City in the Guangxi Zhuang Autonomous Region as the research area, utilizing multispectral remote sensing data as the data source and Landsat series image data for relevant preprocessing. It calculates the monthly normalized vegetation index (NDVI) and surface temperature (LST) data for Laibin City. Based on the ecological environment and surface coverage conditions of the research area, the ratio vegetation index (RVI), normalized vegetation moisture index (NDWI), temperature vegetation drought index (TVDI), and conditional vegetation temperature drought index (VTCI) were selected to calculate and invert the drought monitoring results of Laibin City. The drought monitoring results were obtained and overlaid with the vegetation area map to generate the vegetation drought monitoring results of Laibin City. Based on the climate, geography, and ecological characteristics of the monitored area in Laibin City, a specific analysis will be conducted to develop an appropriate TVDI index drought level, and generate vegetation drought level result maps for Laibin City in 2021, 2022, and 2023. Then, a detailed analysis of the vegetation drought situation in Laibin City is conducted according to time and space. Among them, in the past three years, the vegetation areas in Laibin City have experienced drought seasons mostly in summer and autumn. The interannual drought is mainly mild drought, and the proportion of areas with mild drought shows a relatively stable trend. In conclusion, TVDI proves to be a valuable tool for monitoring vegetation drought in Laibin City, offering insights for efficient water resource management strategies.
17 March 2026 in IEEE Transactions on Geoscience and Remote Sensing
In response to the challenges faced by traditional deep learning models in adaptively extracting multi-scale features from remote sensing images and establishing long-range dependencies across both channels and spatial dimensions—issues that contribute to a decline in classification performance—this study introduces a novel network architecture that integrates dynamic multi-scale attention with spatiotemporal fusion, designated as (Multi-scale Dynamic Convolution and Attention Fusion Network) MDCA-Net.The research presents a Dynamic Multi-scale Large Kernel Attention, termed D-MLKA, which is engineered to dynamically weight and amalgamate multi-scale convolutional features derived from 7×7 and 5×5 kernels, thereby facilitating a spectral-spatial adaptive perception mechanism. Subsequently, a Spectral–Spatial Convolution and Attention Fusion Module, SS-CAFM, is developed to leverage depthwise separable convolutions for the extraction of cross-channel spectral correlations, in conjunction with multi-head attention to effectively capture long-range spatial dependencies. Moreover, a pseudo-temporal enhancement unit is introduced to bolster the model’s robustness against interannual variations by generating virtual temporal features. Comparative experiments conducted on the Indian Pines and Houston University public datasets yielded overall accuracies (OA) of 77.38% and 83.34%, respectively. Furthermore, when validated against the 2024 multispectral dataset from Rong’an County, Guangxi Autonomous Region, the model achieved a classification accuracy of 98.66%, representing a 2.15% improvement over existing methodologies such as Morphformer. Ultimately, the model demonstrated Average Accuracy (AA) accuracies of 98.12% and 97.79% on datasets from 2020 and 2022, respectively, thereby sustaining high performance levels while mitigating the risk of performance degradation.