Aug 2026· Measurement science and technology· Vol 37, pp. 356304· 0 citations· 50 references
Physics
TL;DR
The Navigable Terrain Segmentation Network (NTSNet), which integrates two core modules: the Gaussian Fourier Attention Module (GFAM) and the Entropy-Aware Fusion Module (EAFM).
Abstract
Accurately identifying navigable areas in wild autonomous driving environments using onboard visual sensors remains a challenging task due to highly complex and dynamic terrain. Existing methods often struggle to simultaneously balance global discrimination and local consistency, leading to insufficient terrain understanding. To address this, we propose the Navigable Terrain Segmentation Network (NTSNet), which integrates two core modules: the Gaussian Fourier Attention Module (GFAM) and the Entropy-Aware Fusion Module (EAFM). The GFAM introduces Gaussian Fourier-driven spatial priors to model long-range dependencies, enabling robust extraction of deep global semantic features and enhancing global discrimination. The EAFM, comprising the attention entropy modulation Unit and weight entropy aggregation Unit, leverages Shannon entropy to uncover latent advantageous information among neighboring features and adaptively fuse multi-scale representations, improving local consistency. Quantitative and qualitative experiments on two wild terrain datasets demonstrate that the proposed NTSNet achieves competitive segmentation performance against state-of-the-art methods, providing reliable pixel-level terrain perception for wild autonomous driving systems. The code for NTSNet will be made available at: https://github.com/lv881314/NTSNet.
Terrain matching localization provides an effective passive navigation solution for unmanned aerial vehicles. However, traditional methods often rely on manually designed descriptors, raw elevation comparisons, or iterative optimization, making them sensitive to measurement noise, terrain ambiguity, and rotation distur...
Yao Jiang, Q. Miao, De-Wei Wu et al.· International Conference on...· 0 citations
Autonomous navigation of Mars rovers requires accurate terrain perception, as surface properties directly influence traversability and wheel slip. However, semantic segmentation of Martian terrain remains challenging due to the high visual similarity and low contrast between surface classes, particularly sand and rocks...
Yi-Fan Dai, Ke-Hao Du, A. V. Bobkov· 2026 International Russian A...· 0 citations
Timely and accurate delineation of flooded regions from uncrewed aerial vehicle (UAV) imagery is critical for disaster response, damage assessment, and resource allocation. Yet flood segmentation remains difficult: water surfaces exhibit specular reflections, low-texture homogeneity, and diffuse, irregular boundaries a...
Hassan Ali, Muhammad Asghar Khan, Gordana Barb et al.· IEEE Journal of Selected Top...· 0 citations
MCMB-UNet is proposed, an effective dual-path encoder model with multi-attention mechanisms that achieves a favorable accuracy-efficiency trade-off compared to mainstream Transformer-based models, and demonstrates good applicability on the Inria Aerial Labeling and Massachusetts Buildings datasets.
Dong-Dong Huang, Yu-Hong Ding· Signal, Image and Video Proc...· 0 citations
This work proposes RDAQ, a density-aware dynamic detection framework, which explicitly activates tiny-object details through top-down guidance, thereby bridging the semantic-spatial gap and proposing a density-driven dynamic Filter (DDDF), which generates sample-specific convolutional weights according to density prior...
Ling-Feng Lin, Bo-Xiang Xie, W. Luo et al.· IEEE Geoscience and Remote S...· 0 citations
GFE-Net is rigorously benchmarked on two widely adopted large-scale datasets—S3DIS and SensatUrban—yielding OA/mIoU of 89.6%/73.1% and 93.3%/61.1%, respectively.