AUTHOR=Nian Rui , Liu Shasha , Lu Zongcan , Li Xiaoyu , Ren Shidong , Qian Yuqi , Li Qiuying , He Guotong , Shi Kexin , Zhang Guoyao , Zang Lina , Li Luyao , He Bo , Yan Tianhong , Li Xishuang TITLE=Toward the development of smart capabilities for understanding seafloor stretching morphology and biogeographic patterns via DenseNet from high-resolution multibeam bathymetric surveys for underwater vehicles JOURNAL=Frontiers in Marine Science VOLUME=Volume 10 - 2023 YEAR=2023 URL=https://www.frontiersin.org/journals/marine-science/articles/10.3389/fmars.2023.1205142 DOI=10.3389/fmars.2023.1205142 ISSN=2296-7745 ABSTRACT=The increasing use of underwater vehicles, helps facilitate deep sea exploration at a broad range of depths and spatial scales. In this paper, we make an initial attempt to develop online computing strategies to identify seafloor categories and predict biogeographic patterns, with one deep learning based architecture DenseNet, integrated with joint morphological clues, expecting to potentially develop the embedded smart capacities. We utilize high-resolution multibeam bathymetric measurements derived from MBES, and denote a collection of joint morphological clues, to help semantic mapping and localization. We systematically strengthen dominant feature propagation and encourage feature reuse via DenseNet, by employing the channel attention module and the spatial pyramid pooling. It has been illustrated from our experiment results that the seafloor classification accuracy reached up to 89.87%, 82.01%, 73.52% averagely, in terms of PA, MPA, MIoU metrics, achieving comparable performances with the state-of-art deep learning frameworks. We make a This is a provisional file, not the final typeset article preliminary study on potential biogeographic distribution statistics, which permits us to delicately distinguish the functionality of probable submarine benthic habitats. This study demonstrates the premise in deploying underwater vehicles through unbiased means or pre-programmed path planing to quantify and estimate seafloor categories and the exhibiting fine-scale biogeographic patterns.