Search

The use of deep learning techniques for semantic segmentation in remote sensing has been increasingly prevalent. Effectively modeling remote contextual information and integrating high-level abstract features with low-level spatial features are critical challenges for semantic segmentation tasks. This article addresses these challenges by constructing a graph space reasoning (GSR) module and a dual-channel cross-attention upsampling (DCAU) module. Meanwhile, a new domain-incremental learning (DIL) framework is designed to alleviate catastrophic forgetting when the deep learning model is used in cross-domain. This framework makes a balance between retaining prior knowledge and acquiring new information through the use of frozen feature layers and multifeature joint loss optimization. Based on this, a new DIL of remote sensing semantic segmentation with multifeature constraints in graph space (GSMF-RS-DIL) framework is proposed. Extensive experiments, including ablation experiments on the ISPRS and LoveDA datasets, demonstrate that the proposed method achieves superior performance and optimal computational efficiency in both single-domain and cross-domain tasks. The code is publicly available at https://github.com/Huang WBill/GSMF-RS-DIL.

The leakage of phase-change materials presents a significant challenge that impedes their application. Loading porous materials onto phase-change materials is an effective approach to addressing this issue. In this study, porous expanded dickite as a carrier was utilized to load decanoic acid and create a composite phase-change material. The loading content of decanoic acid was varied to obtain different composite phase-change materials. Fourier transform infrared spectroscopy analysis confirmed the formation of hydrogen bonds between the expanded dickite carrier and decanoic acid. Scanning electron microscopy images and energy-dispersive X-ray spectroscopy mapping results demonstrated that decanoic acid was evenly dispersed on the expanded dickite carrier without any agglomeration. The expanded dickite carrier effectively immobilized decanoic acid through hydrogen bonding, thereby preventing leakage, as long as the loading content of decanoic acid did not exceed 60%. The higher thermal conductivity of the expanded dickite carrier promoted the thermal conductivity of the expanded dickite/decanoic acid composite phase-change materials, enhancing the responsiveness of the composite phase-change materials to ambient temperature. The composite phase-change material containing 60 wt% decanoic acid exhibited excellent endothermic/exothermic cycle stability, and after six cycles, its latent heat remained stable.