Search

Vegetation cover in the Loess Plateau region is an important component of ecological protection in the Yellow River Basin, and this study provides a scientific reference for further vegetation restoration. Based on Landsat images and related data, we utilized the dimidiate pixel model and Geodetector method to study the vegetation cover in the Wuding River Basin from 2000 to 2022. The results indicated the spatial and temporal distribution of the vegetation cover and its changes over the study period. Additionally, the driving factors influencing its spatial changes were also uncovered. We also propose a land use shift vegetation cover contribution formula to quantify the effect of land type change on the FVC. The study showed that (1) the overall vegetation cover of the watershed increased significantly, and the FVC showed an increasing trend from 2000 to 2013 and a slow decline from 2013 to 2022, with the gradual transformation of low-graded FVC into a higher graded one. (2) The FVC increased spatially from northwest to southeast, and the trend of future changes is mainly decreasing. (3) The strongest explanatory power for the FVC change is the land use type and its interactive combination with rainfall. (4) The conversion of grassland to cropland contributes the most to the vegetation cover at 1.52%, and the increase in the cropland area is more conducive to the increase in the vegetation cover. [ABSTRACT FROM AUTHOR]

The performance of deep neural networks depends on the accuracy of labeled samples, as they usually contain label noise. This study examines the semantic segmentation of remote sensing images that include label noise and proposes an anti-label-noise network framework, termed Labeled Noise Robust Network in Remote Sensing Image Semantic Segmentation (NRN-RSSEG), to combat label noise. The algorithm combines three main components: network, attention mechanism, and a noise-robust loss function. Three different noise rates (containing both symmetric and asymmetric noise) were simulated to test the noise resistance of the network. Validation was performed in the Vaihingen region of the ISPRS Vaihingen 2D semantic labeling dataset, and the performance of the network was evaluated by comparing the NRN-RSSEG with the original U-Net model. The results show that NRN-RSSEG maintains a high accuracy on both clean and noisy datasets. Specifically, NRN-RSSEG outperforms UNET in terms of PA, MPA, Kappa, Mean_F1, and FWIoU in the presence of noisy datasets, and as the noise rate increases, each performance of UNET shows a decreasing trend while the performance of NRN-RSSEG decreases slowly and some performances show an increasing trend. At a noise rate of 0.5, the PA (−6.14%), MPA (−4.27%) Kappa (−8.55%), Mean_F1 (−5.11%), and FWIOU (−9.75%) of UNET degrade faster; while the PA (−2.51%), Kappa (−3.33%), and FWIoU of NRN-RSSEG (−3.26) degraded more slowly, MPA (+1.41) and Mean_F1 (+2.69%) showed an increasing trend. Furthermore, comparing the proposed model with the baseline method, the results demonstrate that the proposed NRN-RSSEG anti-noise framework can effectively help the current segmentation model to overcome the adverse effects of noisy label training. [ABSTRACT FROM AUTHOR]