Page 8 (data 211 to 240 of 351) | Displayed ini 30 data/page
<< PREV 1 2 3 4 5 6 7 8 9 10 11 12 NEXT >>

211	Topic B: Applications of Remote Sensing	ABS-199
Predictive Modelling of Asian Elephant Movement in Endau-Rompin (Peta) National Park, Johor, Malaysia using Maxent and Remote Sensing Approach. Kugan RAJU. 1* and Noordyana HASSAN. 1,2* 1 Department of Geoinformation, Faculty of Built Environment and Surveying, Universiti Teknologi Malaysia (UTM), 81310 UTM Skudai, Johor, Malaysia 2 Geoscience and Digital Earth Centre (INSTeG), Research Institute of Sustainable Environment (RISE), Universiti Teknologi Malaysia (UTM), 81310 UTM Skudai, Johor, Malaysia. Abstract Habitat loss and fragmentation have become major hazards to wide-ranging species such as Asian elephants. In Peninsular Malaysia, the population of Asian elephants is gradually declining due to the degradation of their habitat. Studying the potential movement of an endangered species within a protected rainforest is important, as they are highly significant for ecology. However, spatial data on elephant movement remains limited across Asian elephant range countries, including Peninsular Malaysia. Therefore, in the present study, we estimate the potential Asian elephant movement in Endau-Rompin (Peta) National Park using Maxent and remote sensing while analysing the influence of natural and anthropogenic factors on movement patterns within the landscape. This study used Maximum Entropy Modelling (Maxent) to estimate the potential Asian elephant movement using elephant occurrences and environmental factors. The influence of natural and anthropogenic factors on elephant movement was assessed through statistical analysis and compared with the model output variables contribution. Additionally, the result was validated using supplementary data collected within the study area. The results of this study predicted that 43% of the study area is suitable habitat for Asian elephant movement, with 19% classified as highly suitable and 38% as less suitable. The model training and test data showed satisfactory AUC scores of 0.85 and 0.90, respectively. In conclusion, Maxent is considered an insightful method for estimating the movement patterns of endangered species within a protected rainforest. The Department of Wildlife and National Parks (DWNP) may employ the findings of this study to inform future site-specific conservation and management efforts. Keywords: Asian elephant, maxent, remote sensing. Share Link | Plain Format | Corresponding Author (Kugan Raju)

212	Topic B: Applications of Remote Sensing	ABS-200
Multi-Scale Remote Sensing Analysis for Oil Palm Plantation Mapping: Evaluating the Relative Importance of Spatial and Spectral Resolution Muhammad Nizar Y.P. (a*), Yanuar A.N. (b) , Dika Dika A.J. (c) a) Data Scientist, Professional Services, Esri Indonesia, Jalan Gatot Subroto 6, Kota Jakarta Selatan 12710, Indonesia *ypratama[at]esriindonesia.co.id b) Product and Technology Lead, Solution Engineering, Esri Indonesia, Indonesia c) Account Manager - Enterprise Sales, Esri Indonesia, Indonesia Abstract Accurate oil palm plantation mapping is essential for effective operational management, yet current capabilities remain insufficient. While industrial plantations follow regular patterns enabling straightforward detection, smallholder systems present significant challenges due to irregular shapes, fragmented distribution, mixed cropping systems, and rapid temporal changes, limiting comprehensive monitoring strategies. This study determines optimal remote sensing characteristics for comprehensive oil palm plantation mapping by systematically evaluating the relative importance of spatial and spectral resolution parameters in Southeast Asian tropical environments. We investigate which resolution type provides the most significant contribution to accurate detection of both industrial and smallholder plantation systems. We employ U-Net deep learning architecture to analyze multi-scale oil palm detection across diverse passive remote sensing platforms including drone RGB imagery, Planet/NICFI, Sentinel-2, and Landsat 8/9 datasets. The methodology systematically evaluates spatial resolution thresholds and spectral band combinations through comparative analysis of optical sensing approaches across multiple platforms. Results are expected to establish evidence-based guidelines for optimal pixel size thresholds and critical spectral bands for accurate oil palm mapping. The analysis will demonstrate differential sensitivity between plantation types to each resolution parameter, establishing practical recommendations for cost-effective passive remote sensing monitoring strategies. Keywords: Oil Palm Mapping, Multi-Scale Analysis, Deep Learning, Spatial Resolution, Spectral Resolution Share Link | Plain Format | Corresponding Author (Muhammad Nizar Yoga Pratama)

213	Topic B: Applications of Remote Sensing	ABS-206
GNSS Positioning Environment Assessment in Urban Rivers for Autonomous Boats using Polarimetric SAR data Takuto Nagaoka(a*), Kyogo Noda(a), Tetsu Yamaguchi(a), Nobuaki Kudo(b), Etsuro Shimizu(b), Masahumi Nakagawa(a) a) Shibaura Institute of Technology, Japan *ah22021[at]shibaura-it.ac.jp b) Tokyo University of Marine Science and Technology, Japan Abstract The increasing growing need for autonomous boat navigation on urban rivers has revealed a major challenge: the substantial reduction in Global Navigation Satellite System (GNSS) accuracy caused by surrounding high-rise buildings and bridges. To ensure safe autonomous navigation requires, identifying non-GNSS areas by assessing the positioning environment over a wide area, however, conventional methods lack efficiency and spatial coverage. This study proposes a novel method of estimating the GNSS positioning environment over a wide area using full-polarimetric synthetic aperture radar (SAR) data. We validated its effectiveness using in-situ data from a waterborne mobile mapping system (MMS). We used ALOS-2/PALSAR-2 data and applied, Pauli decomposition to classify microwave scattering mechanisms as surface, double-bounce, or volume scattering. We hypothesized that surface scattering would correspond to favorable GNSS conditions, while double-bounce and volume scattering would indicate poor environments prone to multipath and signal blockage. To validate the estimation, we conducted a qualitative analysis by comparing the Pauli decomposition results with three data sources: 1) 3D point clouds acquired by the waterborne MMS, 2) the 3D city model(Project PLATEAU), and 3) the measured GNSS positioning solutions (RTK-FIX/FLOAT). The analysis revealed a strong correlation between the SAR-derived scattering map and the actual GNSS performance. Specifically, open sky areas such as the Shiomi canal exhibited surface scattering and stable RTK-FIX solutions. In contrast, the sections of the Nihonbashi River under the expressway were dominated by double-bounce and volume scattering, resulting in unstable RTK-FLOAT solutions. This study demonstrates that Pauli decomposition of polarimetric SAR data can effectively evaluate the GNSS positioning environment across an entire area. These results can be applied to optimize navigation routes and generate positioning environment maps to enable seamless switching between GNSS and non-GNSS positioning. Keywords: SAR, ALOS, RTK-GNSS, waterborne MMS Share Link | Plain Format | Corresponding Author (Takuto Nagaoka)

214	Topic B: Applications of Remote Sensing	ABS-208
Investigation of Spatial and Temporal Trends in Lake Toba Water Level Based on ICESat-2 Data Widyani Galih Pangersa (a*), Ni Made Rai Ratih Cahya Perbani, Ir., M. Si (b) a) Undergraduate Program in Geodetic Engineering, Institut Teknologi Nasional Bandung, Indonesia b) Doctoral Program in Geodesy and Geomatics, Institut Teknologi Bandung, Indonesia *glhpngrs[at]gmail.com Abstract Lake Toba is the largest volcanic lake in Indonesia and one of the 15 lakes prioritized by the Indonesian government. As the largest volcanic lake in Southeast Asia, it serves as a critical freshwater reservoir, supports fisheries and tourism, and influences local hydrological and climatic conditions. It also plays an important role in climate change, so fluctuations in its water level need to be monitored continuously. ICESat-2 is a laser based measurement satellite equipped with the Advanced Topographic Laser Altimeter System (ATLAS), which works by counting photons and enables periodic monitoring of water levels. This study focuses on identifying spatial and temporal trends in water levels in Lake Toba using ICESat-2 ATL13 data from 2018 to 2024. The parameters analyzed were segment latitude (segmen_lat), segment longitude (segmen_long), and water surface height (ht_water_surf) from Tracks 385, 825, and 1269. Spatial analysis was conducted using profiles and 3D models along the satellite tracks across Lake Toba. Meanwhile, temporal analysis was performed using temporal data samples on tracks 285, 825, and 1269, which were bounded by the area along each respective track, by selecting data that had the most observation months. Then least square analysis was conducted for temporal modeling Based on the analysis of ICESat-2 data in Tracks 385, 825, and 1269, it was found that the water level in the eastern part of Lake Toba is higher than in the western and central parts, with differences reaching approximately 2.4 meters. A pattern of water level accumulation toward the mainland was also observed in the three-dimensional profiles. Between 2018 and 2024, the lowest water level occurred in 2021. In general, the variation in water levels was influenced by both annual (Sa) and semi-annual (Ssa) cycles, with Ssa being more dominant. These findings provide new insights into the spatial and temporal dynamics of Lake Toba water levels. Keywords: ICESat-2 ATL-13, Lake Toba, Spatial trend, Temporal trend, Water level Share Link | Plain Format | Corresponding Author (Widyani Galih Pangersa)

215	Topic B: Applications of Remote Sensing	ABS-213
DEVELOPMENT OF A NOVEL MULTI-CRITERIA METHOD USING DEEP LEARNING AND OPTIMIZATION FOR IMAGE CLASSIFICATION Tsolmonbayar.Sh1, Bayanjargal.D2*, Batchuluun.Ts1, Tsolmon.R3, Davaajargal.J2, Selenge.M1 and Bayanmunkh.N4,5 1School of Engineering and Applied Science, National University of Mongolia, Ulaanbaatar, Mongolia 2School of Information and Electronic Engineering Applied Matematics National University of Mongolia, Ulaanbaatar, Mongolia (*bayanjargal[at]num.edu.mn) 3School of Art and Sciences Physics department, National University of Mongolia, Ulaanbaatar, Mongolia 4Centre for Policy Research and Analysis of Ulaanbaatar Municipality, Ulaanbaatar, Mongolia 5Mongolian Geospatial Association, Ulaanbaatar 15141, Mongolia Abstract Recent advancements in remote sensing image analysis have increasingly utilized deep learning models, resulting in notable improvements in classification accuracy and computational efficiency. Among these approaches, hybrid methods that combine deep learning with optimization techniques have shown superior performance over conventional single-model algorithms. In this study, we propose a novel classification algorithm called Multi-Criteria Mean Clustering (MCMC). This method integrates deep learning-based feature extraction with a multi-objective optimization framework, enabling it to better capture the diverse characteristics of high-dimensional and heterogeneous remote sensing data. By considering multiple criteria-such as spectral separability, spatial coherence, and class distribution-MCMC enhances clustering robustness and interpretability. The proposed method was applied to a case study in Dornod Province, Mongolia, a region along the Siberian forest boundary known for its complex land cover structure and ecological significance. We used Sentinel-2B multispectral imagery to perform land cover classification. To validate the classification performance, results from MCMC were compared against NDVI-based ground truth data. Correlation analysis revealed a 98% agreement between the MCMC outputs and the NDVI-derived reference map. Additionally, MCMC was benchmarked against two commonly used techniques: Mini-Batch K-Means, known for its scalability, and Random Forest, a widely adopted supervised classification method. Comparative results showed that MCMC either matched or exceeded the performance of these methods, particularly in terms of class boundary delineation and intra-class homogeneity. These findings demonstrate the potential of the MCMC approach in addressing the limitations of existing clustering and classification techniques, especially for complex and heterogeneous remote sensing datasets. Keywords: Multi-Criteria, deep learning, optimization, classification Share Link | Plain Format | Corresponding Author (Bayanmunkh Norovsuren)

216	Topic B: Applications of Remote Sensing	ABS-214
Adaptation of TIN-based Ortho-Mosaicking for DSM Error Mitigation Sunghyeon Kim(a) , Taejung Kim(b*) a) Program in Smart City Engineering, Inha University, 100 Inha-ro, Incheon 22212, Republic of Korea b) Department of Geoinformatic Engineering, Inha University, 100 Inha-ro, Incheon 22212, Republic of Korea *tezid[at]inha.ac.kr Abstract Ortho-mosaicking is a key process required to generate image maps over large areas from UAV imagery. For ortho-mosaicking, multiple UAV images are ortho-rectified by referencing a Digital Surface Model (DSM) and mosaicked to a common image plane. Traditionally, DSMs used for ortho-mosaicking are generated either from LiDAR surveys or dense matching of aerial/UAV imagery. While LiDAR-based DSMs offer high accuracy, they entail high costs and time due to expensive equipment and the very fine spatial resolution of UAV images. In contrast, stereo-based DSMs are more cost- and time-efficient, but frequently suffer from elevation errors in regions such as building boundaries or occlusion zones, leading to geometric inconsistencies in the mosaics. To address these limitations, this study extends the triangulated irregular network (TIN)-based ortho-mosaic approach developed in-house to handle DSM errors. The method divides the DSM space into grid patches, whose sizes and shapes are adaptively adjusted based on local terrain. Three-dimensional points are extracted at the corners of each patch and used to form a TIN for ortho-mosaicking. The TIN vertices are reprojected into the image coordinate system using the original image^s projection model. Image patches are then extracted from the reprojected points and mapped to the mosaicking plane. The method was experimentally validated using UAV datasets over urban areas with significant elevation variation and by incorporating DSMs and DEMs of various resolutions. The experiments evaluated mosaic quality using alignment error, color discontinuity, and mosaic coverage as indicators, under varying grid patch sizes and DSM accuracy levels. Since the proposed method operates on grid patches, it maintains global consistency even under DSM inaccuracies. By adjusting patch size and shape, TIN vertices can omit problematic regions, such as building edges or occlusion zones, thereby reducing distortion and positional errors. Keywords: DSM, UAV Image mosaic, geometric correction, patch size adaptation, urban mapping Share Link | Plain Format | Corresponding Author (Sunghyeon Kim)

217	Topic B: Applications of Remote Sensing	ABS-216
Application of LEI (Landscape Expansion Index) for Monitoring Flood-Prone Areas in Bekasi City in 2005, 2015, and 2025 Agsha Dewantara*, Dinia Izza Rianti, Fauzi Fahmi, Shofura Afanin Nuha, Alfian, Hammam Muhammad Amrullah, Mellinia Regina Heni Prastiwi, Wanda Nelwita Pohan Remote Sensing Graduate Student, Faculty of Geography, Gadjah Mada University, Yogyakarta, Indonesia Abstract Rapid population growth and accelerated urbanization in metropolitan areas, including Bekasi City, indicate uncontrolled land use change. One of its main impacts is the reduction of natural infiltration areas such as rivers, wetlands, and vegetation, which increases surface runoff and exacerbates flood risk. Therefore, it is essential to monitor urban growth spatially and temporally as a foundation for disaster mitigation and land-use management. This study aims to analyze land use changes in Bekasi City in the years 2005, 2015, and 2025, and to identify the directions of urban expansion using the Landscape Expansion Index (LEI) method. Furthermore, the study seeks to map flood-prone areas based on the patterns of built-up area development. The data used includes Landsat 5 imagery (2005), Landsat 7 (2015), and Landsat 9 (2025). Land use classification is performed using the Random Forest machine learning algorithm to distinguish between built-up and non-built-up areas. Subsequently, new built-up zones are analyzed spatially using buffering and overlay techniques to calculate the LEI value for each period. LEI measures the ratio between the edge of new urban patches that adjoin existing built-up areas and the total edge of the new patch. Through this approach, urban expansion patterns such as infill, edge-expansion, or leap-frog can be identified. The LEI analysis reveals the relationship between the direction of urban expansion and increased flood vulnerability. This information is crucial for supporting more adaptive and sustainable spatial planning and serves as a basis for disaster mitigation policy-making in urban areas such as Bekasi City. Keywords: Flood, Landsat Imagery, Landscape Expansion Index (LEI), Land Use Change, Urbanization Share Link | Plain Format | Corresponding Author (Mellinia Regina Heni Prastiwi)

218	Topic B: Applications of Remote Sensing	ABS-219
A Comparative Study of Machine Learning Classification Algorithms for Benthic Habitat Mapping in West Bali National Park Shofura Afanin Nuha (1*), Dita Amelia Pratiwi (1), Alfian(1), and Pramaditya Wicaksono (2) 1 Departement of Geographic Information Science: Postgraduate Student, Faculty of Geography, Universitas Gadjah Mada, Indonesia *shofuraafaninnuha540682[at]mail.ugm.ac.id 2 Departement of Geographic Information Science: Lecturer, Faculty of Geography, Universitas Gadjah Mada, Indonesia Abstract The West Bali National Park (Taman Nasional Bali Barat/TNBB) encompasses a wide range of ecosystems, including coastal areas that host diverse habitats such as mangroves, seagrass, and coral reefs. This unique conservation area, located in the Jembrana and Buleleng Districts, spans approximately 3,415 hectares and supports the growth and protection of rich biodiversity, including benthic habitats. Benthic habitats are essential components of coastal ecosystems, comprising seagrass, coral reefs, macroalgae, and various types of substrate. This study aims to compare the performance of two machine learning classification algorithms Random Forest (RF) and Support Vector Machine (SVM) in mapping benthic habitat composition in the Teluk Terima area of TNBB, using 3-meter resolution PlanetScope satellite imagery with visible and near-infrared (NIR) spectral bands. The classification focuses on four primary benthic habitat classes: dominant seagrass, dominant coral, dominant macroalgae, and dominant substrate. The RF algorithm produced an overall accuracy of 59.1%, mapping 27.3 ha of dominant seagrass, 35.9 ha of coral, 0.5 ha of macroalgae, and 40.2 ha of substrate. In comparison, the SVM algorithm resulted in a lower overall accuracy of 47.3%, mapping 35.4 ha of seagrass, 43.9 ha of coral, 15.5 ha of macroalgae, and 9.2 ha of substrate. Accuracy comparisons indicate that RF is more stable in identifying seagrass and substrate classes, while SVM performs better in detecting coral and macroalgae, despite imbalances in user and producer accuracy. These findings suggest that the choice of classification algorithm significantly affects benthic habitat mapping outcomes, and Random Forest offers more consistent results in shallow water environments with complex substrate compositions. Keywords: Benthic Habitat, Conservation, Random Forest, Support Vector Machine, Machine Learning Share Link | Plain Format | Corresponding Author (Shofura Afanin Nuha)

219	Topic B: Applications of Remote Sensing	ABS-220
A Cloud-Based System for Monitoring Water Quality in Tuul River from Space Misheel Orgil1*, Bolorchuluun Chogsom1, and Zaya Chinbat2 1Department of Geography, National University of Mongolia, Mongolia 2The united graduate school of Agricultural science, Gifu University, Japan Abstract The Tuul River is an essential water source for the water supply and ecological sustainability of Ulaanbaatar. In this study, using Sentinel-2 satellite imagery data from April 21, 2025, the water quality of the Tuul River was assessed using physical parameters such as turbidity, color, and hydrogen ion concentration in accordance with national standards such as MNS ISO 5667:2001, MNS 0900:2018, MNS 4586:2024, and MNS 3900:1986. For this purpose, the Modified Normalized Difference Water Index (MNDWI), the Normalized Difference Turbidity Index (NDTI), which indicate the state of water quality, were calculated. The results of the field survey conducted on April 20, 2025 were compared and verified with the results of physicochemical measurements and analyses taken within the framework of the Water Quality Study of the Tuul River in the Ulaanbaatar City Area project implemented by the Institute of Chemistry and Chemical Technology in 2021. Using the Google Earth Engine platform, it was demonstrated that large amounts of data can be processed over a wide range of space and time, regardless of computer power, and that this methodology can support and in some cases replace field research. Keywords: Water quality, physical parameters, turbidity, color, hydrogen ion Share Link | Plain Format | Corresponding Author (Misheel Orgil)

220	Topic B: Applications of Remote Sensing	ABS-222
The Capability of Machine Learning Combined with PlanetScope for Detecting Seaweed Using the OBIA Method Arpin Hardiana (a*), Nurjannah Nurdin (b, c) a) Doctor Program in Environmental Science, Faculty of graduate school, Hasanuddin University, Jl. Perintis Kemerdekaan Km. 10, Makassar 90245, South Sulawesi, Indonesia b) Marine Science Department, Faculty of Marine Science and Fisheries, Hasanuddin University, Jl. Perintis Kemerdekaan Km. 10, Makassar 90245, South Sulawesi, Indonesia c) Research Center and Development for Marine, Coastal, and Small Island, Hasanuddin University, Jl. Perintis Kemerdekaan Km. 10, Makassar 90245, South Sulawesi, Indonesia arpinhardi[at]gmail.com Abstract Seaweed has high economic value and plays a crucial role in supporting the well-being of coastal communities and maintaining the balance of marine ecosystems. Accurate detection and mapping of seaweed distribution are crucial for planning sustainable cultivation. This study aims to evaluate the ability of a machine learning algorithm combined with high-resolution PlanetScope satellite imagery to map the dynamics of seaweed cultivation areas over 36 months (2022-2024) using PlanetScope satellite imagery with the Object-Based Image Analysis (OBIA) approach. The OBIA method enables more contextual spatial analysis by segmenting images into homogeneous objects, which are then classified using the Nearest Neighbour machine learning algorithm. The analysis process includes segmentation, extraction of spectral, textural, and spatial features, and object classification based on the training model. The results demonstrate that integrating PlanetScope and OBIA yields high classification accuracy in distinguishing seaweed objects from other types of vegetation and aquatic substrates. The machine learning algorithm has been proven capable of processing complex multivariate data to improve detection accuracy. The study found that peak seaweed cultivation occurs between March and September in most coastal areas. The highest cultivation area varies annually, with peaks in July 2022 (198.51 Ha), April 2023 (264.56 Ha), and March 2024 (205.62 Ha), while the final quarter of the year showed a significant decline, particularly in November 2023 (7.58 Ha) and 2024 (27.98 Ha), with an accuracy of 70.91%. These findings significantly contribute to the use of machine learning-based remote sensing technology to support efficient, accurate, and sustainable coastal resource management. Therefore, this approach has great potential for routine monitoring of seaweed cultivation areas in various coastal areas of Indonesia. Keywords: Seaweed, Machine learning, Planetscope, OBIA Share Link | Plain Format | Corresponding Author (Arpin Hardiana)

221	Topic B: Applications of Remote Sensing	ABS-223
Analysis of PM10 Air Quality Parameters Based on Satellite Imagery in The Cement Industrial Area Sumarni Hamid Aly 1,2*, A.Azizah Nurul Dinanti 1,2, Nurul Masyiah Rani Harusi 1,2, Ibrahim Djamaluddin 1,2, Poppy Indrayani 3, Hengky Pala^langan 1 1Department of Environmental Engineering, Facutly of Engineering, University of Hasanuddin, St. Poros Malino KM. 6, Bontomarannu, Gowa, South Sulawesi, 92172, Indonesia 2 Transportation and Air Quality Research Group, University of Hasanuddin, St. Perintis Kemerdekaan No. KM. 10, Makassar, South Sulawesi, 90245, Indonesia 3 Department of Environmental Engineering, Faculty of Engineering, Muslim University of Indonesia, Jl. Urip Sumoharjo km.05 Makassar City, 90231, Indonesia Abstract Industry is one of the main causes of global warming and a significant source of emissions. Growth in production rates in a cement industry sector can cause an increase in pollutant emission loads into the air, which has the potential to affect pollutant concentrations in ambient air. This study aims to analyze PM10 concentrations and the Air Pollution Index (API) category based on satellite imagery in the X Cement Industry.This research was conducted with Sentinel-2 and Landsat 8 OLI/TIRS satellite image data using algorithm calculations of PM10 air quality parameters, data used in this study are Sentinel 2 on October 16, 2023, and Landsat 8 on October 16, 2023. The highest PM10 concentration for Sentinel-2 occurred in the operational area (Coal Stock Pile Unit 2/3/4), whereas Landsat 8 detected its highest value in a residential area (in front of Kampung Sela Mosque). The results of the ambient air concentration analysis from satellite imagery were converted into the Air Pollution Standard Index (ISPU) with the Average PM10 ISPU results of 60,10 (Moderate). With an RMSE test of 93,06 for Residential Area and for the Operational 123,47. confirming that PM10 is the dominant pollutant of concern. These findings highlight the significance of particulate matter monitoring in emission prone zones using satellite based approaches. Keywords: Sentinel 2, Landsat 8, Cement Industry, Air Pollution Index, Remote Sensing Share Link | Plain Format | Corresponding Author (Sumarni Hamid Aly)

222	Topic B: Applications of Remote Sensing	ABS-224
Analysis of CO Air Quality Parameters Based on Satellite Imagery in Makassar Metropolitan City Sumarni Hamid Aly 1,2*, Alisyah Ismi Zahra Mardan 1, Nurul Masyiah Rani Harusi 1,2, , Mitani Yasuhiro3 , Khaerul Amru4,2, , Muhammad Rais Abidin5 1Department of Environmental Engineering, Facutly of Engineering, University of Hasanuddin, St. Poros Malino KM. 6, Bontomarannu, Gowa, South Sulawesi, 92172, Indonesia 2 Transportation and Air Quality Research Group, University of Hasanuddin, St. Perintis Kemerdekaan No. KM. 10, Makassar, South Sulawesi, 90245, Indonesia 3 Department of Civil Engineering, Faculty of Engineering, University of Kyushu, Fukuoka, 819-0395, Japan 4 Research Center for Environmental and Clean Technology, National Research, and Innovation Agency (BRIN), Geotech Building 820, Puspuptek Sorong, South Tangerang, Indonesia 5Department of Geography, Universitas Makassar State, St. Mallengkeri Raya, Parang Tambung, Makassar City, South Sulawesi, 90224, Indonesia Abstract Carbon monoxide (CO) is one of the most common hazardous air pollutants in urban areas and requires effective monitoring methods. This study aims to estimate CO concentrations in Makassar Metropolitan City using Sentinel-2 and Landsat 8 satellite imagery and to evaluate the differences between the two datasets. The estimation was conducted using the Somvanshi algorithm, which incorporates green, red, and SWIR reflectance bands, followed by Root Mean Square Error (RMSE) analysis to assess consistency between the datasets. exhibit relatively homogeneous and narrow ranges, reflecting stable spatial distributions of CO, with RMSE values ranging from 0.02 to 0.05, indicating minimal discrepancies. Sentinel-2, with higher spatial resolution, provides greater detail in identifying variations across road types and urban activity centers, whereas Landsat 8 yields more generalized patterns. These findings highlight the potential of multi-satellite approaches in air quality monitoring while emphasizing the need for local calibration to improve accuracy in tropical urban environments. Keywords: Air quality, Carbon monoxide, Sentinel-2, Landsat 8, Remote sensing, Makassar Metropolitan City Share Link | Plain Format | Corresponding Author (Sumarni Hamid Aly)

223	Topic B: Applications of Remote Sensing	ABS-232
Monitoring Flooded Areas in Bekasi City Using Sentinel-1 Backscatter Analysis Hammam Muhammad Amrullah*, Alfian, Mellinia Regina Heni Prastiwi, Wanda Nelwita Damayani, Fauzi Fahmi, Shofura Afanin Nuha, Agsha Dewantara, Dinia Izza Rianti Remote Sensing Graduate Student, Faculty of Geography, Gadjah Mada University, Indonesia Abstract Flooding is one of the most destructive types of hydrometeorological disasters, characterized by fast-moving water, widespread damage, and high frequency of occurrence. The city of Bekasi faces increasingly severe flooding challenges each year, with fluctuating incidence rates. This is triggered by a combination of geographical factors, human activities, and the inadequacy of existing disaster management systems. The objective of this study is to map flood events using Synthetic Aperture Radar (SAR) data in the form of Sentinel-1 imagery processed through the Google Earth Engine platform with the application of backscatter difference analysis and analyze the multi-temporal pattern of flood inundation starting from 2020 to 2025. The results of flood condition mapping research indicate that the Bekasi City area will experience an increase in the extent of affected areas by 2025. Utilizing SAR imagery is one of the methods that can be employed for flood inundation monitoring, particularly by leveraging the spatial and temporal coverage of remote sensing data, which is a key advantage for monitoring floods across extensive study areas over specific time periods. In the context of sustainable development, the use of remote sensing data plays a crucial role, particularly in risk assessment and the development of hazard inventories, as well as in evaluating progress toward Sustainable Development Goals (SDGs), especially in climate change mitigation actions (SDG 13) that may impact communities and in safeguarding terrestrial ecosystems (SDG 15) for the future. Keywords: Flood Monitoring, Sentinel-1, Bekasi City, Google Earth Engine Share Link | Plain Format | Corresponding Author (Mellinia Regina Heni Prastiwi)

224	Topic B: Applications of Remote Sensing	ABS-233
Analysis of Patch Shapes in Dense Matching for Reducing Disparity Errors in UAV Stereo Images Hongjin Kim (a), Chaeyeon Lee (b), Taejung Kim (b*) a) Program in Smart City Engineering, Inha University 100 Inha-ro, Incheon 22212, Republic of Korea b) Department of Geoinformatic Engineering, Inha University 100 Inha-ro, Incheon 22212, Republic of Korea *tezid[at]inha.ac.kr Abstract Disparity estimation using stereo images acquired from unmanned aerial vehicle (UAV) becomes a core technology for constructing precise 3D spatial information, digital twins, and smart cities. However, dense matching based on fixed-size square-shaped patches often results in mismatches in high-frequency regions such as building boundaries. This is because the traditional square-shaped patches do not sufficiently account for the directional structure or depth discontinuities at object boundaries, and because they treat different regions within the patch equally for computing matching costs. As a result, information near the boundaries becomes averaged, leading to blurred edges and disparity diffusion. This phenomenon degrades the quality of 3D data and causes information loss in object boundary areas. To alleviate this phenomenon, this study analyzes disparity variations within stereo image pairs by applying various types of patches during the matching cost computation stage. The matching cost is calculated using zero-mean normalized cross-correlation (ZNCC). Disparity maps were generated for each patch shape at multiple scales, and the characteristics of the resulting maps such as structural preservation at edges, information averaging, and disparity spreading were compared. Based on the analysis, a final disparity map was generated by combining multi-scale and multi-shaped patches. The effects of this fusion approach were evaluated on reducing mismatches near boundaries, suppressing disparity blurring, and improving the overall resolution and quality of the disparity map. Experimental results showed that the proposed method reduced information loss around object boundaries and improved the overall quality of the disparity map. This suggested that the proposed method not only effectively mitigated the boundary diffusion problem observed in conventional fixed-patch approaches but also contributed to the precision and practicality of UAV based 3D spatial information generation. Keywords: Dense stereo matching, UAV image, Disparity estimation, 3D reconstruction Share Link | Plain Format | Corresponding Author (hongjin KIM)

225	Topic B: Applications of Remote Sensing	ABS-238
Predictive Modelling of Mangrove Above Ground Biomass through the Integration of Spectral Indices and Field-Based Allometric Data Aldea Noor Alina1,2*, Lalu Muhamad Jaelani1, and Noorlaila Hayati1 1Department of Geomatics Engineering, Institut Teknologi Sepuluh Nopember, Surabaya 60111, Indonesia. 2Department of Geography Education, Universitas Negeri Surabaya, Surabaya 60231, Indonesia. Abstract This study aims to develop a predictive model for estimating Above Ground Biomass (AGB) of mangroves by integrating various vegetation spectral indices. The research addresses the challenge of accurately estimating AGB by combining high resolution satellite imagery with field-based allometric data obtained through non-destructive surveys in the mangrove area of East Surabaya Coast. Allometric data sampling was conducted using purposive random sampling with dominantly 30 x 30 meters plot area according to SNI 7724:2019. The data retrieved was circumference or Girth Breast Height (GBH) with standard at 1.3 meters from the ground surface, measurement was conducted for all sapling, poles, and stands of mangrove trees in each plot. The data acquired also height, soil moisture, salinity, and pH level of the plot area The methodology involves spectral indices calculation such as the Normalized Difference Vegetation Index (NDVI), Normalized Difference Moisture Index (NDMI), and Combined Mangrove Recognition Index (CMRI) derived from Worldview-2 Satellite Imagery. These spectral indices are integrated with field derived biomass values to develop multiple predictive models. Allometric equation that incorporate DBH and Height displayed better relation value between Aboveground Biomass with its parameter compared to Aboveground Biomass that only incorporate DBH. The AGB model with a DBH^2H parameter has higher R square value at 0.61 compared to the AGB model with a DBH parameter of 0.39. Stock carbon modeling developed using vegetation indices showed the relationship between field carbon stock data and pixel values in the vegetation index transformation. The results of the correlation test on NDVI and CMRI were positive, meanwhile NDWI showed negative correlation. According to R2 value, NDWI displayed best relation value compared to NDVI and CMRI, even thought the relation were negative with 0.543. Keywords: Above Ground Biomass, Allometric, Mangrove, Spectral Indices, Predictive Model Share Link | Plain Format | Corresponding Author (ALDEA NOOR ALINA)

226	Topic B: Applications of Remote Sensing	ABS-240
Development of a Smart Coastal Environment Management System in Incheon Using Remote Sensing and AI-based Analysis of Marine Debris Distribution Jeon H., Cho H., Jang S.T., Yoon D.H., Choi K.S., Lee S.W., Kim T.H. Ocean Convergence Division, Underwater Survey Technology 21. Inc., Korea Aviation Center, Incheon Technopark, Korea Aviation Department, Incheon Metropolitan City, Korea Abstract Marine Debris, a byproduct of human activities, is often discharged, dumped, or abandoned along coastlines. This not only leads to the destruction of marine habitats but also causes significant economic losses by hindering vessel navigation, reducing fisheries productivity, and degrading coastal landscapes, thereby negatively impacting the tourism industry. To address the issue of marine debris, the international community is actively advancing monitoring systems that utilize advanced technologies such as satellite remote sensing, drones, and AI-based analysis. This technology-driven approach moves beyond conventional methods reliant on visual observation or uncertain estimations, enabling more precise and systematic responses through high-resolution spatial data and temporal change analysis. Korea is also actively promoting the adoption of monitoring systems utilizing advanced technologies. Through this approach, the country aims to establish strategic debris collection measures-such as the designation of priority management areas and the determination of optimal collection periods-based on scientific evidence. Located in the mid-northern part of the Yellow Sea, Incheon is a semi-enclosed coastal area where sea water exchange is limited, creating conditions that allow external inputs such as marine debris to accumulate easily. These environmental characteristics make it well-suited for applying image-based detection technologies and analyzing spatiotemporal variation patterns. This suitability makes Incheon an ideal location for demonstrating national marine debris policies and introducing advanced monitoring technologies, which is why it has been selected as a target site for various pilot projects. In this study, high-resolution SkySat satellite imagery (spatial resolution: ~0.5 m) was used to develop a machine learning model based on the spectral reflectance characteristics of two target materials-white styrofoam and orange buoys. The model was applied to detect and classify beach litter in island regions of Ongjin County, Incheon, on a monthly basis, with the aim of analyzing the spatiotemporal changes in litter distribution over time. In addition, simulated beach litter larger than 1 meter in size was placed along the shoreline, and spectral data were acquired using a hyperspectral camera mounted on an unmanned aerial vehicle (UAV). This field-based assessment aimed to explore more precise detection and analysis of beach litter compared to high-resolution satellite imagery. The detection results obtained using remote sensing technology and machine learning models are provided through the Marine Keeper platform(http://mk.helios.pe.kr) as time-series graphs of beach litter by type, along with map-based location information. The accumulated monthly detection results are used to generate spatiotemporal distribution forecasts through AI-based prediction technology. Such AI-driven spatiotemporal analysis and prediction are expected to support data-driven decision-making and serve as a foundation for establishing monitoring standards, ultimately contributing to the improved efficiency of coastal environmental management strategies. Keywords: Marine Debris- Remote Sensing- High-resolution Satellite Image- Incheon Share Link | Plain Format | Corresponding Author (TAEHO KIM)

227	Topic B: Applications of Remote Sensing	ABS-242
Spatiotemporal Analysis of Surface Urban Heat Islands using Otsu Threshold and Gaussian Approach for Local Climate Change Detection Atika Izzaty (a)*, Hone-Jay Chu (b) and Mohammad Adil Aman (b) a) Geodetic Engineering, Universitas Hasanuddin, Makassar 90245, Indonesia *atikaizzaty[at]unhas.ac.id b) Department of Geomatics, National Cheng Kung University, Tainan 701, Taiwan Abstract Both visible and subtle forms of natural events are common on Earth. In light of these continuous changes, it is essential to monitor and comprehend environmental conditions. Because urbanized areas are centers of industrial development and population mobility, climate change is especially noticeable there. This study uses the Otsu Threshold and Gaussian Mixture Model (GMM) techniques to examine environmental changes, with a focus on fluctuations in air and land surface temperatures. The ability to identify different environmental changes has been greatly improved by remote sensing technology, especially when using multispectral sensors on Landsat satellite imagery. Trends for air temperature have increased noticeably in recent years, indicating notable environmental changes in the Taipei Area as compared to Makassar City between 2018 and 2023. The climate patterns in Taipei have become more unpredictable throughout the course of these five years. Likewise, Makassar has experienced erratic and sometimes sudden wet seasons. The outcomes of the Otsu and GMM models, which successfully apply thresholding techniques to precisely identify urban heat zones, both exhibit these patterns. Interestingly, there was an increase in surface temperatures in 2022. Pixels with temperatures below 19.07 degree Celsius were categorized as low-temperature areas in the GMM model, while pixels with temperatures exceeding 25.70 degree Celsius were categorized as high-temperature zones. In contrast, the Otsu model divided temperatures into four thresholds: 19.07, 21.69, 23.32 then 25.70 in degree Celsius. By using thresholding and cluster analysis approaches, both models showed excellent skills in efficiently processing the data. The GMM model clustered the data according to the thermal conditions identified in each satellite image, whereas the thresholding strategy provided a unique temperature threshold to each image. Keywords: SUHI, Otsu Threshold, GMM, Landsat-8, Multispectral Share Link | Plain Format | Corresponding Author (Atika Izzaty)

228	Topic B: Applications of Remote Sensing	ABS-244
Evaluating the Sensitivity of Spectral Indices for Detecting Waste Dumping Sites Using Satellite Imagery Abdullah Harun Incekara, Dursun Zafer Seker Tokat Gaziosmanpasa University, Faculty of Engineering and Architecture, Department of Geomatics Engineering, 60150 Turkiye Istanbul Technical University, Civil Engineering Faculty, Department of Geomatics Engineering, 34469 Turkiye Abstract Waste management is a crucial issue for sustainability. Regular and unregulated waste storage sites need to be monitored for their environmental impacts and mapped for risk assessment. The primary requirement for these processes is the identification of waste storage areas, such as open dump sites. Remote sensing techniques using satellite imagery offer an effective means for such identification. In this study, Normalized Differerence Vegetation Index (NDVI), Normalized Differerence Water Index (NDWI), Normalized Difference Built-up Index (NDBI), Soil Adjusted Vegetation Index (SAVI), and Dump Detection Index (DDI) were applied to Sentinel imagery to detect various dumping sites (both regular and unregulated) across Turkiye. The sensitivity of these indices for detection was examined based on the similarities and differences in reflectance of landfills with different land use and land cover classes. In the application conducted for landfills with known locations, the Dump Detection Index emerged as the more useful index, while other indices were found to be insufficient for initial detection. Indices other than DDI were found to help make potential inferences. Keywords: Remote sensing, Spectral indices, Waste dumping, Dump Detection Index (DDI) Share Link | Plain Format | Corresponding Author (Dursun Zafer Seker)

229	Topic B: Applications of Remote Sensing	ABS-247
Spatial Analysis of Accessibility, Vegetation Density, and Equity of Urban Green Spaces: A Case Study of Esenyurt, Istanbul Tarik Emre Yorulmaz, Ugur Alganci, Dursun Zafer Seker Istanbul Technical University, Graduate School, Geomatics Engineering Program, 34469 Istanbul, Turkiye Istanbul Technical University, Civil Engineering Faculty, Geomatics Engineering Department, 34469 Istanbul, Turkiye Abstract Urban green spaces (UGSs) are essential for sustainable development and improving residents quality of life. However, urban green space is not always equally accessible to all city residents. This inequality is linked to a wider range of injustices, including disparities in health and well-being, and is a significant environmental justice issue. Accordingly, in this study, the spatial distribution of accessibility, vegetation density, and equity in Esenyurt, Istanbul, which is the most populous district of the city and the country, was analyzed. Using GIS-based network analysis, we found that approximately 813.27 hectares of residential areas are served by green spaces within a 500-meter walking distance, while 650.26 hectares remain unserved. The normalized vegetation density (NDVI) analysis showed moderate overall vegetation density, with significant variations across neighborhoods. Equity analysis revealed that green space per capita is only 1.01 m2, significantly below the 10 m2 threshold established by Turkiye official regulations. Based on the findings of this research, it is obvious that urban planners and decision-makers should prioritize the development of new urban green spaces to address accessibility and equity gaps. Keywords: Urban green spaces, GIS-based network analysis, Green space accessibility, Vegetation Density, Spatial equity, Urban sustainability Share Link | Plain Format | Corresponding Author (Dursun Zafer Seker)

230	Topic B: Applications of Remote Sensing	ABS-249
Drought Monitoring and Analysis Using Multi-Source Data and Machine Learning Abdullah Sukkar, Ozan Ozturk, Dursun Zafer Seker ITU, Istanbul Technical University Abstract Drought is a major environmental challenge with widespread implications for ecosystems, economies, and societies. Despite its critical importance, drought remains one of the most challenging natural disasters to monitor due to its gradual onset and complex interactions with climatic and environmental factors. In a conflict-affected region such as northeast Syria, understanding the drought patterns and trends is a crucial step in post-conflict rehabilitation, particularly because this area relies heavily on agriculture. To enhance our knowledge of the drought phenomenon in northeast Syria, a comprehensive big dataset was created, including a variety of meteorological, vegetation, and soil parameters. Then, a machine learning model based on the XGBoost algorithm was employed to assess the most important features affecting the drought. The Standardized Precipitation Evapotranspiration Index, Vegetation Health Index, and Soil Moisture anomaly were selected as targets to represent the meteorological, vegetation, and soil data. For more reliable results, when selecting a target, the data of this target was left out of the training process, which enables the detection of how other parameters affect that target. The results showed that the most important parameter affecting the drought is the temperature. Keywords: Drought monitoring- Multi-Source data integration- XGBoost- Drought indices Share Link | Plain Format | Corresponding Author (Dursun Zafer Seker)

231	Topic B: Applications of Remote Sensing	ABS-250
Marine Heatwaves Study Based on Copernicus OSTIA L4 Satellite Derived Sea Surface Temperature and Geospatial Analysis in the Arafura Sea (1982-2024) Attaqwa R., Maulana A., Saputri A.D., Harini S.H.N., Ramadhan D.A., and Helmi M. Department of Oceanography, Faculty of Fisheries and Marine Science, Universitas Diponegoro, Indonesia Abstract Atmosphere and ocean interactions have driven significant rises in sea surface temperature (SST) across the Arafura Sea, Indonesia, exacerbated by extensive marine heatwaves (MHWs). Understanding these events is essential due to the region ecological and economic importance, especially for fisheries and coral ecosystems. This study aims to analyze the temporal and spatial characteristics of MHWs in the Arafura Sea between 1982 and 2024. This research is among the first to comprehensively examine MHWs in this region using a long-term SST dataset. We employed Hobday et al. hierarchical detection algorithm and utilized OSTIA L4 Copernicus Global Ocean Physics Reanalysis SST data (1982-2021) along with Near Real Time SST data (2022-2024). We identified 65 distinct MHWs events, accounting for 2,325 MHW days over the 42-year period. The most intense event occurred from 7 September to 29 November 2022 (84 days), with a peak anomaly of 2.26 degrees Celsius and an average intensity of 1.43 degrees Celsius. Spatially, MHWs frequency and duration peaked in the central and northern Arafura Sea, with coastal maximum intensities reaching 2.0&#8239-degrees Celsius. These results confirm an intensifying MHWs trend in the region. The findings are vital for advancing oceanographic knowledge, supporting marine resource management, and guiding coastal climate resilience strategies. Keywords: Marine Heatwaves, Arafura Sea, Remote Sensing, Sea Surface Temperature, Geospatial Analysis Share Link | Plain Format | Corresponding Author (Rizal Attaqwa)

232	Topic B: Applications of Remote Sensing	ABS-251
Modeling the Relative Risk of Air Pollution on Respiratory Health Using Remote Sensing Techniques in Chonburi Province, Thailand Pramet Kaewmesri, Chanika Sukawattanavijit, Kulapach Lhapawong, Phukrit Sriwilas Geo-Informatics and Space Technology Development Agency (Public Organization), Bangkok, Thailand Abstract Air pollution remains a major environmental and public health concern, particularly in rapidly urbanizing and industrializing regions such as Southeast Asia. This study investigates the relationship between air pollutants and respiratory health outcomes in Chonburi province, Thailand-a densely populated and industrialized economic hub. The objective is to quantify the relative risk (RR) of respiratory diseases associated with various air pollutants, using a combination of ground-based and satellite-derived data. Daily data from 2017 to 2020 were collected, including concentrations of fine and coarse particulate matter (PM1, PM2.5, PM10), gaseous pollutants (SO2, NO2, CO, and O3), and meteorological parameters such as temperature, humidity, and rainfall. Health outcome data were obtained from hospital records related to respiratory illnesses. Poisson regression and machine learning models were employed to assess the health impact under multiple exposure scenarios. To enhance spatial and temporal coverage, satellite data from the Sentinel-5P mission were integrated, particularly for gases such as NO2, SO2, CO, and O3. This fusion of remote sensing with ground-level observations allowed for a more comprehensive and high-resolution estimation of pollution exposure. The findings reveal that PM1 has the strongest association with respiratory disease incidence, with an RR of 1.12 (95% CI: 1.08-1.17), surpassing the effects of PM2.5 (RR = 1.08) and PM10 (RR = 1.05). Notable health impacts were also observed for NO2 (RR = 1.06) and SO2 (RR = 1.03), while O3 exhibited a neutral or slightly inverse effect. This study highlights the importance of including PM1 in routine air quality monitoring and demonstrates the potential of integrated ground and satellite data to inform a scalable, data-driven public health alert system. The methodological framework can support national-scale health impact assessments and policy making in Thailand. Keywords: Air Pollution- Respiratory Disease- Relative Risk (RR)- PM1 and Fine Particulates- Machine Learning in Public Health Share Link | Plain Format | Corresponding Author (Pramet Kaewmesri)

233	Topic B: Applications of Remote Sensing	ABS-252
Study of Internal Solitary Waves Characteristics Using Synthetic Aperture Radar Imagery and the Korteweg-de Vries Numerical Model in the South Andaman Sea Alfiza N., Gaol M.G.L., Putri K.M.A., Mahendra S.A.H., Zhafran M.L. and Helmi M. Departement of Oceanography, Faculty of Fisheries and Marine Science, Universitas Diponegoro, Indonesia Departement of Marine Science, Faculty of Fisheries and Marine Science, Universitas Diponegoro, Indonesia Abstract Internal Solitary Waves (ISWs) are subsurface phenomena generated by the transformation of internal tides as they interact with complex seafloor topography, leading to intense vertical mixing and influencing subsurface transport processes. The South Andaman Sea, characterised by strong density stratification and intricate bathymetry, presents favourable conditions for ISW generation. This study aims to identify and characterise ISWs in this region using Sentinel-1 Synthetic Aperture Radar (SAR) imagery. Wave parameters were derived through the Korteweg-de Vries (KdV) model and validated with in situ oceanographic variables including temperature, salinity, and eastward and northward current velocities at propagation points. One dominant propagation point was identified, exhibiting a maximum wave amplitude of 558.92 m, a phase speed ranging from 3.29 to 6.54 m/s, horizontal current velocities up to 42.12 m/s, and vertical velocities reaching 10.32 m/s. The ISWs at this location displayed strong and coherent solitonic behaviour, supported by high kinetic and potential energy values and stable nonlinearity and dispersion parameters. The most energetic wave showed a kinetic-to-potential energy ratio of 539.77. These results underscore the utility of Sentinel-1 SAR imagery in detecting ISWs and demonstrate the value of integrating remote sensing with numerical oceanographic modelling to enhance our understanding of subsurface wave dynamics in tropical oceanic environments, particularly for risk mitigation and the enhancement of navigational safety, both at the surface and underwater, as well as for the security of offshore structures. Keywords: Internal Solitary Waves, SAR, Korteweg-de Vries, Andaman Sea Share Link | Plain Format | Corresponding Author (Neisha Alfiza)

234	Topic B: Applications of Remote Sensing	ABS-253
A Deep Learning Approach for Remote Sensing Based Estimation of PM2.5 in Urban Areas: A case study of Greater Taipei Area Wiso V.C. (1)*, Hsuan Ren.(2), Lin T.H.(3) Centre for Space & Remote Sensing Research (CSRSR), National Central University 300 Zhongda Road, Zhongli District, Taoyuan City 320317, Taiwan (ROC) Abstract Air pollution remains a threat, with particulate matter being one of the several key indicators of pollution. Numerous studies show that particulate matter of diameter 2.5 micrometer or smaller impacts public health, hence the need for more accurate models. Traditional numerical methods fail to capture pollutants^ spatiotemporal patterns- however, neural networks can learn from environmental data inputs to obtain correlations. In this study, a Temporal Convolution Network Bi Long Short-Term Memory hybrid model is proposed. TCN BiLSTM is for Particulate matter monthly estimation in 2023 using Taiwan^s Ministry of Environment air quality historical data from July 2015 to December 2022, including weather parameters and land use data, to produce a 1km map via k nearest neighbors inverse distance weighting. The former model captures causal patterns while the latter handles long range temporal dependencies. The preliminary results compare the proposed model against classical models like Convolution Neural Network and Long Short-Term Memory. The proposed model performs better compared to the others with a metric evaluation of R2 of 0.784 and root mean square error of 2.225 microgram per cubic metre in the GroupKFold cross validation. This study aims to contribute to the bulk of air quality monitoring research and hopefully help mitigate the missing data issue, mainly due to cloud cover on satellite observations during aerosol optical depth retrieval. Future work will evaluate how the model predicts in the 4 common seasons: winter, spring, summer and autumn Keywords: Particulate Matter (PM), Spatiotemporal, Temporal Convolutional Network, Long Short-Term Memory Share Link | Plain Format | Corresponding Author (Vincent Cletus Wiso)

235	Topic B: Applications of Remote Sensing	ABS-255
Mapping Coral Reef Habitat Using High-Resolution Satellite Imagery and Temporally Disparate In Situ Data: A Multi-Depth Analysis Batrisyia Che Azani1, Nurul Ain Mohd Zaki2,3*, Mohd Zainee Zainal4, Rozaimi Che Hasan4 and Tajul Rosli Razak5,6 1Research assistant/ Postgraduate Students, Faculty of Built Environment, Universiti Teknologi MARA, Perlis Branch, Arau Campus, 02600 Arau, Perlis, Malaysia 2Senior lecturer, Faculty of Built Environment, Universiti Teknologi MARA, Perlis Branch, Arau Campus, 02600 Arau, Perlis, Malaysia 3Associate Fellow, Institute for Biodiversity and Sustainable Development, Universiti Teknologi MARA, Shah Alam, 40450, Malaysia 4Senior lecturer, School of Engineering and Technology, Jalan Sultan Yahya Petra, Universiti Teknologi Malaysia, 54100 Kuala Lumpur, Wilayah Persekutuan Kuala Lumpur, Malaysia 5Senior lecturer, College of Computing, Informatics and Mathematics, Universiti Teknologi MARA, Shah Alam, Malaysia 6Postdoctoral Researcher, Department of Architecture and Built Environment, University of Nottingham, United Kingdom Abstract Over the past few years, coral reefs have faced multiple threats from environmental and anthropogenic stressors. It is beyond doubt that monitoring coral reefs is crucial particularly for habitat changes detection. Despite that, matching the date for both satellite images with field data oftentimes can be challenging. Therefore, this study used Pleiades high-multi resolution satellite image of Pinang Island, Terengganu, Malaysia for coral reef habitat classification and in situ data from CPCe photo quadrats and underwater photographs collected in 2024 which are temporally delayed. The satellite image, limited to surface reflectance, was processed to extract spectral and texture indices. In situ data, including 376 georeferenced quadrats with depth and substrate composition from 40 locations, were integrated and stratified by depth (<5m and >5m) to assess classification performance across varying optical conditions. Furthermore, random forest algorithm classification will be applied as machine learning classifiers with underwater photographs as visual reference to address the four-year temporal gap. This approach aims to demonstrate how older satellite data can support coral reef mapping when recent or up-to-date imagery is unavailable through the integration with more recent field observations, from several years later. Keywords: coral reef mapping, temporal mismatch, Pleiades satellite, machine learning classification Share Link | Plain Format | Corresponding Author (Batrisyia Che Azani)

236	Topic C: Emerging Technologies in Remote Sensing	ABS-260
Automated Plot-Level Paddy Land Mapping in Smallholder Agriculture Using Deep Learning and Multi-Temporal Satellite Imagery - A Sri Lankan Case Study Rajanayake R.M.A.B.1*, Fernando W.J.G.S.T.V.2, and Dharmawansha. B.A.D.K.H.3 GeoEDGE (Pvt) Ltd Abstract ABSTRACT Accurate and efficient mapping of paddy lands in Sri Lanka remains a critical challenge due to the heterogeneous and homogeneous nature of plot-level cultivation, coupled with the limitations of manual surveys and conventional remote sensing techniques. Existing methods struggle with spectral similarities between paddy fields and other vegetation, particularly in smallholder-dominated landscapes where field boundaries are irregular and fragmented. This study develops an automated deep learning model to precisely identify and measure paddy land extent using multi-temporal Sentinel-2 satellite imagery at 10-metre resolution. The proposed approach combines a U-Net convolutional neural network with spectral-temporal feature extraction, integrating vegetation indices (NDVI, NDWI) and phenological characteristics to improve discrimination of paddy fields from other land covers. Model training and validation use a ground-truth dataset of 8,000 plot boundaries from key paddy-growing areas in Sri Lanka. Comparative evaluation demonstrates the model^s superior performance, achieving 91.4% classification accuracy and an F1-score of 0.88 for paddy identification, significantly outperforming conventional machine learning approaches such as random forest (78%) and support vector machines (73%) in plot-level delineation. The system successfully addresses key challenges in smallholder paddy mapping, including mixed cropping patterns and seasonal variations in field conditions. This AI framework replaces labour-intensive surveys with accurate paddy mapping, yield estimation, and decision support, and can be adapted for similar regions to support sustainable land and yield management. Keywords: Deep learning, convolutional neural networks, remote sensing, paddy land mapping Share Link | Plain Format | Corresponding Author (Anuradha Bandara Rajanayake)

237	Topic C: Emerging Technologies in Remote Sensing	ABS-264
Adaptive Real-Time Anomaly Detection and Data Imputation in Multi-Device IoT Streams via Nonlinear Dimensionality Reduction and Dynamic Sliding-Window Statistics Rahmat Faisal ESRI Indonesia Abstract The rapid growth of diverse IoT deployments has made real-time monitoring of multi-sensor streams essential for ensuring system reliability and data integrity. In this work, we propose a novel framework that combines nonlinear dimensionality reduction with adaptive sliding-window statistical analysis to enable scalable, online anomaly detection and missing data imputation. Sensor data are first compressed via a lightweight autoencoder that captures complex nonlinear correlations across sensors in a compact latent representation. A dynamic sliding-window statistical module then adaptively adjusts the window size and computes key features such as mean, variance, entropy, and reconstruction error from recent latent embeddings. Anomalies are detected using context-sensitive thresholds derived from continuously updated statistical baselines, while missing or corrupted values are imputed using both latent autoencoder outputs and interpolation guided by recent temporal trends. This architecture is designed for deployment in edge-cloud hybrid environments optimized for resource-constrained IoT devices. The framework supports accurate, low-latency anomaly detection and data repair across heterogeneous IoT streams, adapting to evolving sensor behaviors without relying on predefined thresholds or offline training. This approach provides a robust and scalable solution for maintaining high data quality in complex IoT ecosystems. Keywords: Anomaly Detection, IoT, Real-Time Anomaly Detection Share Link | Plain Format | Corresponding Author (Rahmat Faisal)

238	Topic C: Emerging Technologies in Remote Sensing	ABS-266
AeroVision: Onboard AI for Crisis Monitoring JAYARAMA PRADEEP(a), RITHESHA S(b) (a)Department of Electrical and Electronics Engineering, St. Joseph^s College of Engineering, Chennai, India (b)Department of Electrical and Electronics Engineering, St. Joseph^s College of Engineering, Chennai, India Abstract In this age of increasing number of natural disasters, the need for quick situational awareness and real-time environmental monitoring has never been more urgent. This research involves creating an autonomous drone-based remote sensing platform that combines deep learning methods with edge computing to provide fast and accurate assessments of disaster areas. The system uses the NVIDIA Jetson Orin Nano as its onboard AI processor, allowing for real-time analysis of aerial images captured with a mix of RGB and depth sensors, including an Intel RealSense Camera. A Pixhawk 4 flight controller ensures stable navigation and autonomous route planning. Additionally, thermal cameras improve the drone^s ability to find heat signatures, which is crucial for locating survivors in areas affected by disasters. Unlike traditional remote sensing methods that depend on cloud processing after data collection, this drone system processes information at the edge. This reduces delays and allows for immediate action. The platform applies AI algorithms to detect features such as debris, water bodies and people, which are essential for emergency management and relief efforts. This research connects remote sensing, edge AI and UAV systems, providing a low-cost, scalable, and effective solution for real-time disaster monitoring. The proposed system advances smart remote sensing and shows how emerging technologies can help tackle important global issues. Keywords: Edge AI- Autonomous drones- Disaster response- Semantic segmentation- NVIDIA Jetson- UAV Share Link | Plain Format | Corresponding Author (JAYARAMA PRADEEP)

239	Topic C: Emerging Technologies in Remote Sensing	ABS-268
Global High Resolution (0.05 degree) Mapping of Soil Freeze-Thaw Dynamics via Optimized Microwave-Optical Fusion and DFA Algorithm Tianjie Zhao, Defeng Feng, Pei Yu, Ziqian Zhang State Key Laboratory of Remote Sensing and Digital Earth, Aerospace Information Research Institute, Chinese Academy of Sciences Abstract Near-surface soil freeze-thaw (F/T) cycles critically regulate global hydrological, ecological, and climatic processes. While passive microwave remote sensing enables all-weather F/T monitoring, its coarse spatial resolution limits fine-scale applications, and performance disparities among microwave indicators remain unquantified. To address these gaps, this study first systematically evaluated eight key microwave indicators-including liquid water indices (NPR, MPR, QE, NFDI) and surface temperature indices (TbV6.9-36.5)-and their 16 combinations across six soil networks on the Tibetan and Inner Mongolian Plateaus. Results identified the quasi-emissivity (QE) and TbV36.5 as optimal universal indicators, with their synergy achieving robust performance. The discriminant function algorithm (DFA) significantly outperformed threshold-based methods, particularly under snow/vegetation interference. Building on this optimized detection framework, we developed a novel downscaling approach integrating passive microwave and optical data to generate the first long-term (2002-2023), high-resolution (0.05 degree), daily seamless global F/T dataset. Validation against in situ networks confirmed an overall accuracy of 83.78%-matching coarse-resolution fidelity while enhancing spatial detail. This dataset revealed new global dynamics: regions north of 45 degreeN exhibit 187.8 +- 12.7 mean annual frost days, with 14.35% showing significant declines in frozen persistence- freeze onset occurs on day 240.3 +- 7.2 annually, delayed across 9.10% of areas. By elucidating microwave indicator synergies, advancing algorithm robustness (DFA), and delivering a high-resolution global F/T product, this research enables refined quantification of land-atmosphere interactions critical for hydrological, erosion, and climate models. The dataset is openly available at https://doi.org/10.11888/Cryos.tpdc.301551. Keywords: Passive microwave- Downscaling- Soil freeze-thaw cycle Share Link | Plain Format | Corresponding Author (Jingyao Zheng)

240	Topic C: Emerging Technologies in Remote Sensing	ABS-22
Automated Monitoring of Floating Roof Oil Tanks Using High-Resolution SAR Imagery and Computer Vision Techniques Yi-Keng Chen (a*), Zheng-Shin Su (b), Shu-Wei Chang (c), Jen-Yu Han (d) a) Department of Civil Engineering, National Taiwan University, Taiwan b) Department of Civil Engineering, National Taiwan University, Taiwan c) Department of Civil Engineering, National Taiwan University, Taiwan d) Department of Civil Engineering, National Taiwan University, Taiwan Abstract Oil tanks are critical national energy infrastructures, and monitoring their fuel capacity is essential, especially in remote or inaccessible regions. This study presents an automated methodology to estimate and monitor floating roof oil tank volumes using high-resolution Synthetic Aperture Radar (SAR) imagery. Unlike fixed-roof tanks, floating roof tanks exhibit observable geometric changes in SAR images due to variable oil levels, making them suitable for remote sensing-based fuel estimation. Leveraging the advantages of SAR-such as all-weather, day-and-night imaging capabilities-we developed a computer vision-based framework incorporating image preprocessing, noise reduction, and geometric normalization to address the inherent speckle noise and imaging distortions of SAR. The methodology includes automated extraction of oil tank features and Hough Transform-based fitting of elliptical or circular arcs to detect tank boundaries, accounting for deformations due to floating roof positions. To enhance accuracy and robustness, we introduce a size normalization procedure and apply constraints to remove secondary reflections that cause false arcs. A U-Net neural network model is further employed for semantic segmentation and feature extraction, serving as a cross-verification mechanism for the automated measurements. Compared to previous approaches that require optical imagery or rely on multi-temporal datasets, our method minimizes data requirements and improves generalizability across different geometric conditions. Experimental results demonstrate the framework^s effectiveness in accurately estimating tank dimensions and fuel levels, even under challenging imaging conditions. This study offers a scalable and practical solution for strategic fuel monitoring in inaccessible regions, with applications in national energy planning, disaster response, and security analysis. Keywords: SAR, Floating Roof Oil Tank, Computer Vision, Hough Transform, U-Net Share Link | Plain Format | Corresponding Author (Yi-Keng Chen)

Page 8 (data 211 to 240 of 351) | Displayed ini 30 data/page
<< PREV 1 2 3 4 5 6 7 8 9 10 11 12 NEXT >>

Powered By Konfrenzi Ultimate 1.832M-Build8 © 2007-2025 All Rights Reserved