Geo-Spatial Information Science

#ai #geo-spatial #location Call for papers：GSIS' new Special Issue entitled"Future Trends in Geo-spatial Information Science: The role of Location Intelligence"(Guest Editor: John Van Genderen Zohreh Masoumi ). Welcome to submit!

🌍🚀 🌠 Are you passionate about leveraging AI and machine learning to tackle global challenges? The University of Manchester is seeking ambitious researchers to join and make a real-world impact on sustainable development goals. 🌟 Don't miss this stellar opportunity—reach out to Prof Robert Beswick and Dr Jack Radcliffe, and apply before January 8th, 2025. 🌌 #FSEPhD #FutureTech #ResearchOpportunity #phd

🗺️ [Latest article]Single target tracking in high-resolution satellite videos: a comprehensive #review By Xin Huang, Ding Wang, Qiqi Zhu, Ying Zheng, and Qingfeng (Gene) Guan https://lnkd.in/gTJbtTXf Recent advancements in #satellitevideo data acquisition technology have significantly improved #targettracking, which is crucial for monitoring fast-changing events. However, unlike ordinary videos, #satellite #videos present challenges such as small and dim targets due to factors like occlusion and illumination variation. This paper addresses the gap in comprehensive satellite video object tracking by providing an overview, meta-analysis, and performance evaluation of various algorithms, while also compiling open-source datasets and suggesting future research directions.

✨ [Latest article]Integrating street view imagery and taxi trajectory for identifying urban function of street space By Gao, Mianxin, Haijing Guo, Longwei Liu, Yongyu Zeng, Wenkai Liu, Yefei Liu, and Hanfa Xing https://lnkd.in/gXxzUpAM This study addresses the oversight of non-traffic functions in #streetspace research by introducing a classification method that combines taxi trajectory data and street view imagery. The method successfully classifies street spaces into five functional types in #Shenzhen's Futian District, aiding in the assessment of street space quality and informing #urbanplanning and management.

🗺️ [Latest article]Land use/land cover (LULC) classification using hyperspectral images: a review https://lnkd.in/gUba7f83 By Lou, Chen, Mohammed A. A. Al-qaness, Dalal AL-Alimi, Abdelghani Dahou, Mohamed Abd Elaziz, Laith Abualigah, and Ahmed A. Ewees. This #review study addresses the critical challenge of classifying #Hyperspectral Images (HSIs) for Land Use/Land Cover (LULC) by providing a systematic overview of the field, analyzing specific datasets, and exploring both traditional and advanced methods including #deeplearning and #spectral decomposition techniques. It aims to serve as a foundational resource, offering insights into current practices and future research directions in #HSI classification, with a comprehensive analysis of potential challenges and developmental trajectories.

🎇 [Latest article]New method for landslide susceptibility evaluation in alpine valley regions that considers the suitability of InSAR monitoring and introduces deformation rate grading By Zhou, Dingyi, Zhifang Zhao, Wenfei Xi, Xin Zhao, and Jiangqin Chao.  https://lnkd.in/gMDn3H3S This study presented a novel approach for #landslide susceptibility assessment in alpine valley regions , which incorporates Interferometric Synthetic Aperture Radar (#InSAR) data and introduces a grading system based on deformation rates, significantly improving the accuracy and efficiency of susceptibility evaluation. The study demonstrates that the proposed method, which includes a Particle Swarm Optimization-Back Propagation (PSO-BP) algorithm and high-resolution image integration, enhances grading accuracy by 37.84% to 60.91% and provides a reliable and effective model for evaluating landslide susceptibility in regions like the Dongchuan district in southwest China .

🌏 [Latest article]A cross-stage features fusion network for building extraction from remote sensing images By Zuo, Xiaolong, Zhenfeng Shao, Jiaming Wang, Xiao Huang, and Yu Wang https://lnkd.in/gDsuYpPw This study proposed a new Cross-stage Features Fusion Network (CFF-Net) for building extraction from #remotesensing images, which addresses the limitations of traditional #deeplearning methods by fusing features from different network stages and employing an #attention mechanism to focus on scale-specific important information. The CFF-Net demonstrated superior performance over existing algorithms on the #Inria and #WHU datasets, achieving a balance between accuracy and efficiency with faster convergence and higher robustness.

✨ [Latest article]Evaluation of optical and microwave-derived vegetation indices for monitoring aboveground biomass over China By Chang, Zhongbing, Lei Fan, Jean-Pierre Wigneron, Ying-Ping Wang, Xiaojun Li, Mengjia Wang, Xiangzhuo Liu https://lnkd.in/gnF8r5CK This study evaluated the effectiveness of four #microwave-derived #Vegetation Optical Depth (VOD) products and optical vegetation indices (VI) for monitoring aboveground #biomass (AGB) across China, finding that the tree cover product had the highest spatial agreement with reference AGBs, followed by L-VOD, outperforming other VIs and VODs. The research suggests that L-VOD is a particularly effective proxy for large-scale AGB monitoring in China due to its higher sensitivity to the vegetation characteristics of the entire #canopy, including #woody components .

✨ [Latest article]Ecological health assessment of Tibetan alpine grasslands in Gannan using remote sensed ecological indicators By Du, Zeyu, Xibin Ji, Jane Liu, Wenyue Zhao, Zhibin He, Jiachang Jiang, Qiyue Yang, Liwen Zhao, and Jinlong Gao https://lnkd.in/gwx7GxjF This study evaluates the #ecosystem health of the Gannan alpine grassland in China using the #RemoteSensing-based Ecological Index (RSEI), providing a comparative analysis with #GrossPrimaryProductivity (GPP) and highlighting the strong comparability between RSEI and GPP in reflecting ecosystem health. The results indicate a good overall health of the Gannan alpine grassland ecosystem with a slow, fluctuating upward trend from 2000 to 2020, while also identifying regional variations and degradation patterns, offering valuable data for ecological restoration efforts in the area.

✨ [Latest article]ForestSemantic: a dataset for semantic learning of forest from close-range sensing By Liang, Xinlian, Hanwen Qi, Xuejie Deng, Jianchang Chen, Shangshu Cai, Qingjun Zhang, Yunsheng Wang, Antero Kukko, and Juha Hyyppä. https://lnkd.in/gMJ_mmjH This paper introduces a new open point cloud dataset ForestSemantic dataset that addresses the critical need for high-quality annotated forest data in #DeepLearning (DL) by providing detailed #pointcloud data at both individual tree and plot levels, supporting instance and semantic studies. This comprehensive dataset, created from TLS data, includes #semantic information down to visible branches and offers structural tree traits, enabling advanced research and development in forest ecosystem services assessment and DL algorithm benchmarking.