Search

The task of camouflaged object segmentation (COS) is to extract objects with camouflage characteristics that are difficult to detect in an image. The current mainstream COS method mainly relies on the general object segmentation framework using the encoder-decoder structure to predict the objects. However, the current COS framework suffers from over-reliance on the labeled ground truth and the output results. To this end, we propose a bilateral diffusion model for camouflaged object segmentation (BiDiCOS). The main innovation of BiDiCOS is that we design a new bilateral diffusion paradigm, transforming the segmentation and depth estimation tasks into denoising tasks. At the same time, the depth estimation branch can provide semantic information to the segmentation branch. Specifically, we design the adaptive feature fusion module for two branches to provide the required features adaptively. We design the bilateral feature fusion module to fuse the depth information with the mask information to reduce the semantic and logical errors during segmentation. Finally, we develop the specialized training strategy and inference strategy for BiDiCOS, which provide a solid foundation for realizing rational training and accurate prediction. Comprehensive experiments on four challenging COS datasets attest to the superior performance of BiDiCOS compared to the existing 20 COS methods. Our BiDiCOS achieves state-of-the-art with different sizes of backbones, where the MAE is reduced by 4.3 % on average. We have also performed comparative experiments on depth estimation. The experiments demonstrate that BiDiCOS can achieve a win–win situation for both tasks. The code will be available at https://github.com/jiangxinhao2020/BiDiCOS. [ABSTRACT FROM AUTHOR]

• Multiple sources of DOM were investigated in Daning River tributary of the TGR. • Hydrological management of TGR enhanced the terrestrial signal of DOM in Daning River tributary. • Hydrological management enhanced organic carbon burial in Daning River tributary. With the linkage between dissolved organic matter (DOM) and the characteristics of natural ecosystem assessed extensively, the properties of DOM in reservoirs, the typical human interrupted ecosystems, have been focused on in recent years, which is critical for the understanding of human impacts on watershed ecosystems and carbon cycling. This study aims to analyze the effect of hydrological management on the DOM chemistry and organic carbon burial in Daning River tributary of the world's largest Three Gorges Reservoir (TGR). Based on the application of a combined approach including bulk geochemical analyses, optical spectroscopy, and ultrahigh-resolution Fourier transform ion cyclotron resonance mass spectrometry, various sources of DOM (terrestrial, anthropogenic, and autochthonous sources) were revealed. An increasing trend of terrestrial and recalcitrant DOM was observed along the upstream to downstream transect of Daning River tributary, which was mainly caused by the water intrusion with a higher terrestrial and recalcitrant signature from mainstream to tributary resulted from hydrological management of TGR. Integrated with the analysis of sedimentary organic matter in Daning River tributary in the past decade (after the construction of TGR), our work suggests that organic carbon burial in the reservoir could be enhanced by hydrological management-induced variation in DOM chemistry. Further studies are needed to better constrain the effects of damming reservoirs on carbon cycling considering their booming all over the world. [Display omitted] [ABSTRACT FROM AUTHOR]