International journal of applied earth observation and geoinformation : ITC journal最新文献

{"title":"Decoding spatial consistency of multi-Source land cover products in China: Insights from heterogeneous landscapes","authors":"Yanglin Cui ,&nbsp;Chunjiang Zhao ,&nbsp;Yuchun Pan ,&nbsp;Kai Ma ,&nbsp;Xiaojun Liu ,&nbsp;Xiaohe Gu","doi":"10.1016/j.jag.2025.104529","DOIUrl":"10.1016/j.jag.2025.104529","url":null,"abstract":"<div><div>High-resolution land cover (LC) data are essential for ecological monitoring and resource management, especially in heterogeneous landscapes containing diverse LC types. With the growing of available LC products, a comprehensive evaluation of their classification accuracy and spatial consistency is important for users’ selection and application. In this study, we compared eight widely used LC products in China, including ESA World Cover (ESA20), ESRI GLC10 (ESRI17, ESRI20), FROM-GLC10 (FROM-GLC17), CLCD (CLCD20), GlobeLand30 (GLB20), GLC_FCS30 (GLC_FCS20), and GLC_FCSD30 (GLC_FCSD20), to examine their performances at both national and regional scales. We employed pixel-wise overlay analysis, visually interpreted validation samples, and classical landscape metrics to assess overall consistency and classification accuracy. The results show that the 30m_combination (CLCD20, GLB20, GLC_FCS20, and GLC_FCSD20) exhibits higher overall consistency at the national scale, with perfect consistency exceeding 60 %. In contrast, the 10m_combination (ESA20, ESRI17, ESRI20, and FROM_GLC17) captures finer regional details but displays greater inconsistencies in central and western regions. ESA20 achieves the highest overall accuracy (OA) at 88.5 % (CI: 88.44 %–88.56 %), while FROM_GLC17 records the lowest at 82.79 % (CI: 82.73 %–82.85 %). Cropland, forest, water, and snow/ice demonstrate higher consistency and classification accuracy (F1-scores &gt; 80 %), whereas wetland, grassland, impervious surfaces, and bare land underperform in fragmented regions. Furthermore, spatial consistency is strongly associated with landscape metrics such as the aggregation index (AI) and contagion (CONTAG), which enhance consistency in large, contiguous patches (e.g., Northeast China Plain). Conversely, edge density (ED) and patch density (PD) show negative associations with consistency, highlighting persistent mapping challenges in fragmented regions (e.g., Yunnan-Guizhou Plateau and Qinghai-Tibet Plateau). These findings offer actionable insights for improving LC mapping in complex terrains and underscore the critical role of landscape metrics in advancing ecological monitoring and resource management.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"139 ","pages":"Article 104529"},"PeriodicalIF":7.6,"publicationDate":"2025-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143816231","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Frequent drought and flood events in the Yellow River Basin, increasing future drought trends in the middle and upper reaches","authors":"Jianming Feng ,&nbsp;Tianling Qin ,&nbsp;Xizhi Lv ,&nbsp;Shanshan Liu ,&nbsp;Jie Wen ,&nbsp;Juan Chen","doi":"10.1016/j.jag.2025.104511","DOIUrl":"10.1016/j.jag.2025.104511","url":null,"abstract":"<div><div>Under global warming, the Yellow River Basin (YRB), serving as an ecological barrier and climate-sensitive region in northern China, faces severe challenges such as frequent extreme droughts and floods, as well as intensifying water resource supply–demand conflicts. To systematically assess the evolution of droughts and floods in the YRB, this study utilizes observational data from 137 meteorological stations and CMIP6 scenario models, employing the dual-index system of the Standardized Precipitation Index (SPI) and Standardized Precipitation Evapotranspiration Index (SPEI) to quantitatively analyze historical drought and flood characteristics, and project future trends under different scenarios. The results indicate that both SPI and SPEI effectively identify drought and flood events, with SPEI demonstrating superior sensitivity to extreme droughts and floods due to its integration of evapotranspiration effects. From 1956 to 2020, the basin’s drought index increased at a rate of 0.003–0.025 per decade, while the flood index changed at a rate of −0.006–0.039 per decade. Droughts were frequent in the middle and upper reaches (30.61 % severe droughts), while flood risks were prominent in the lower reaches (6.25 % extreme floods). Under SSP3-7.0 and SSP5-8.5 scenarios, drought-dominated patterns intensified (severe droughts reaching 62.92 %), and extreme floods showed an increasing trend in the middle and lower reaches. Therefore, the middle and upper reaches should prioritize building drought-resilience systems; the southern and lower reaches should enhance flood-defense infrastructure. Reservoir operations should be optimized using 1–6 months of drought and flood warnings and coupled with groundwater replenishment strategies for 12–24 months of drought cycles.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"139 ","pages":"Article 104511"},"PeriodicalIF":7.6,"publicationDate":"2025-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143816233","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A SAR wave-enhanced method combining denoising and texture enhancement for bathymetric inversion","authors":"Aijun Cui ,&nbsp;Yi Ma ,&nbsp;Jingyu Zhang ,&nbsp;Ruifu Wang","doi":"10.1016/j.jag.2025.104520","DOIUrl":"10.1016/j.jag.2025.104520","url":null,"abstract":"<div><div>The wave phenomena in SAR images are able to provide water depth information. SAR ocean images are often characterized by unclear wave texture and strong speckle noise, which will hinder the bathymetric inversion. Denoising and texture enhancement are two strategies to improve image quality. However, noise reduction may blur textures, while texture enhancement may amplify noise. To address this, we propose a wave texture enhancement method to balance noise reduction and texture preservation. First, an adaptive total variation bounded Hessian method removes noises from SAR ocean images while preserving texture. Next, an improved Frankle-McCann Retinex method enhances ocean wave features in the denoised image without adding noises. Finally, the resulting image is used for wave-based bathymetric inversion experiments conducted at Car Nicobar Island, Chowra Island, and Dongdao Island, covering depths up to 40 m. The proposed method improved bathymetric accuracy, reducing mean absolute error (MAE) by up to 4.69 m and mean relative error (MRE) by up to 18 %. In addition, the proposed method has a positive effect on the estimation of wavelength and period parameters. Experimental results show that blurred ocean surfaces in SAR images significantly affect bathymetric inversion. Thus, wave enhancement is an important step prior to performing bathymetric inversion.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"139 ","pages":"Article 104520"},"PeriodicalIF":7.6,"publicationDate":"2025-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143815971","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Improving crop condition monitoring using phenologically aligned vegetation index anomalies – A case study in central Iowa","authors":"Haoteng Zhao,&nbsp;Feng Gao,&nbsp;Martha Anderson,&nbsp;Richard Cirone,&nbsp;Geba Jisung Chang","doi":"10.1016/j.jag.2025.104526","DOIUrl":"10.1016/j.jag.2025.104526","url":null,"abstract":"<div><div>Timely monitoring of crop conditions is essential for optimizing and assessing agricultural management. Vegetation indices (VIs) derived from remote sensing data can be useful for assessing crop conditions on a large spatial scale. Traditional crop condition assessments compare a VI in the current year to a baseline VI, averaged over multiple years. However, comparing VIs across years by calendar day may not capture the general crop condition at the same development stage due to interannual variability in planting date and weather. This study proposes a phenological alignment approach for assessing differences in corn and soybean crop condition at commensurate growth stages rather than by day of year. The analysis was conducted in central Iowa, U.S. from 2018 to 2023, which included periods of drought and excess rainfall, providing a high interannual variability in crop phenology and condition. Weekly crop condition and seasonal yield data reported by the USDA National Agricultural Statistics Service (NASS) were compared with aggregated Enhanced Vegetation Index (EVI2) anomalies to evaluate relationships both spatially and temporally. Three EVI2 anomaly time series were computed with temporal alignment based on: day of the year (DOY), days after emergence (DAE), and accumulated growing degree day (AGDD), with a scaled time axis aligned at the emergence date. For the DAE- and AGDD-aligned EVI2 time series, emergence date was determined using a within-season emergence detection approach based on remote sensing. Results showed that EVI2 anomalies perform well in crop condition assessment at 30-m resolution, and correlations improved with DAE and AGDD corrections to the EVI2 time series, reducing the effects of yearly differences in crop phenology. The proposed method has potential for improving within-season crop condition monitoring and yield prediction.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"139 ","pages":"Article 104526"},"PeriodicalIF":7.6,"publicationDate":"2025-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143816226","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Estimating water levels in reservoirs using Sentinel-2 derived time series of surface water areas: A case study of 20 reservoirs in Burkina Faso","authors":"Audrey Kantz Dossou Codjia ,&nbsp;Komlavi Akpoti ,&nbsp;Moctar Dembélé ,&nbsp;Roland Yonaba ,&nbsp;Tazen Fowe ,&nbsp;Soumahila Sankande ,&nbsp;Modeste G. Déo-Gratias Koissi ,&nbsp;Sander J. Zwart","doi":"10.1016/j.jag.2025.104523","DOIUrl":"10.1016/j.jag.2025.104523","url":null,"abstract":"<div><div>Reservoirs play a significant role in the mobilization of water resources in Burkina Faso, contributing to the management and availability of water for various purposes. Operational management of reservoirs requires accurate and timely water level information, which remote sensing can provide cost-effectively and with limited resources. In this study, the surface area of 20 reservoirs is first determined using a Random Forest classifier and Sentinel-2 images acquired between 2015 and 2022. The accuracy of the classified surface water areas is evaluated by calculating 5 accuracy assessment metrics. The classifications were validated using manually digitized water areas from high-resolution Google Earth images and compared to the Dynamic World (DW) land cover dataset. Afterward, the spatial variation in the areal extent of the reservoirs is analyzed over time. A linear relationship is established between the estimated surface area and the corresponding observed water level of the reservoirs. The results indicate that reservoir surface areas were accurately classified with Sentinel-2 images (Kappa above 90.35%) for all dates. Moreover, validation with high-resolution images provided an <span><math><msup><mrow><mi>R</mi></mrow><mn>2</mn></msup></math></span> of 0.99 and a Normalized Root Mean Square Error (NRMSE) of 3.53%. Smaller reservoirs exhibit significant variations in surface areas over time as compared to larger ones, which are more stable. The relationship between surface area and water level is satisfactory (<span><math><msup><mrow><mi>R</mi></mrow><mn>2</mn></msup></math></span> ranging from 0.76 to 0.97) for 14 of the 20 analyzed reservoirs. The remaining six reservoirs are affected by aquatic plant intrusion which leads to an underestimation of the surface area. The high accuracy and operational feasibility of the proposed approach demonstrate that Sentinel-2 imagery and machine learning techniques can be recommended for reservoir mapping within the framework of water level monitoring in Burkina Faso.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"139 ","pages":"Article 104523"},"PeriodicalIF":7.6,"publicationDate":"2025-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143816230","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Accurate sub-seasonal root-zone soil moisture prediction using attention-based autoregressive transfer learning and SMAP data","authors":"Lei Xu ,&nbsp;Xihao Zhang ,&nbsp;Xi Zhang ,&nbsp;Tingtao Wu ,&nbsp;Hongchu Yu ,&nbsp;Wenying Du ,&nbsp;Zeqiang Chen ,&nbsp;Nengcheng Chen","doi":"10.1016/j.jag.2025.104532","DOIUrl":"10.1016/j.jag.2025.104532","url":null,"abstract":"<div><div>Root zone soil moisture (RZSM) is an important hydrological variable for agricultural planning and water resources management. The Soil Moisture Active Passive Level 4 (SMAP L4) data demonstrates great value in RZSM estimation. Accurate sub-seasonal RZSM prediction based on SMAP L4 holds great significance for agricultural management and drought assessment. Current deep learning-based RZSM prediction models tend to accumulate error in long-term forecasting and the limited SMAP RZSM samples may result in insufficient model generalization. To address these issues, this study proposes a multi-head self-attention-based autoregressive transfer learning model based on long short-term memory (MAATL) model for sub-seasonal RZSM prediction. The proposed MAATL model is evaluated over the Continental United States (CONUS) for 1- to 60-day RZSM prediction and compared with some ablation and long short-term memory (LSTM) models. The results showed that compared with LSTM, the skills of the MAATL model were significantly improved, with an average correlation coefficient increase of 18.26% and a root mean square error (RMSE) reduction of 42.55%. Furthermore, 118 in-situ soil moisture stations are used for predictive validation and the proposed MAATL model demonstrates higher accuracy compared to the Global Forecast System (GFS) and the LSTM model, with an average correlation skill improvement of 16.02% and 15.08% for MAATL over GFS and LSTM, respectively. These findings indicate superior performance for the proposed MAATL model in sub-seasonal RZSM prediction, which has great potential for agricultural drought preparations.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"139 ","pages":"Article 104532"},"PeriodicalIF":7.6,"publicationDate":"2025-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143816232","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Analyzing the impact of area of interest (AOI) size and endmember selection on evapotranspiration (ET) estimation through a contextual model (SEBAL)","authors":"Hamza Barguache ,&nbsp;Jamal Ezzahar ,&nbsp;Jamal Elfarkh ,&nbsp;Said Khabba ,&nbsp;Salah Er-Raki ,&nbsp;Valerie Le Dantec ,&nbsp;Mohamed Hakim Kharrou ,&nbsp;Ghizlane Aouade ,&nbsp;Abdelghani Chehbouni","doi":"10.1016/j.jag.2025.104514","DOIUrl":"10.1016/j.jag.2025.104514","url":null,"abstract":"<div><div>Accurate estimation of evapotranspiration (ET) is essential for effective water resource management, particularly in arid and semi-arid areas. Advancements in remote sensing technology have made ET models indispensable, offering high-resolution spatial and temporal assessments. Contextual models such as the Surface Energy Balance Algorithm for Land (SEBAL) are particularly valuable for ET estimation. However, one major challenge for these models is the identification of endmembers representing the wet and dry extremes within the AOI. Furthermore, the influence of AOI size on endmember selection raises important considerations for model performance. This work examines how the size of the AOI and endmember selection impact heat flux estimation using the SEBAL model. The research was conducted in an olive orchard at the Agdal site in Marrakech, from May 2022 to April 2023, and at a rainfed wheat field at the Sidi Rehal site from August 2017 to March 2019, using Landsat imagery (L8 and L9) and ERA5 land reanalysis data. For that, SEBAL was applied to six different AOI, ranging from small and homogeneous areas to the full extent of the Landsat imagery. Based on comparisons of SEBAL estimates with eddy covariance data collected from the Agdal site, the analysis shows that difficulties in accurately identifying endmembers are influenced by the size of the AOI. For homogeneous areas, the model struggles to capture the full range of heat fluxes, leading to poor regression relationships. Conversely, applying a shapefile that covers the entire Landsat imagery led to a more uniform distribution of latent heat flux, especially in winter/spring (when the climatic demand is low), which reduced the model’s ability to capture spatial variability. The AOI, which includes a mix of agricultural areas, bare soil, water bodies, and small towns, and whose boundary is relatively close to the measurement station, yielded the best results. It achieved R2 values of 0.95 for H and 0.88 for LE, with RMSE values of 51.24 and 52.41 W/m² for H and LE, respectively. At the regional scale, the larger AOI size produced the lowest results with greater dispersion at the rainfed wheat site, with RMSE values of 104.99 and 93.30 W/m² for H and LE, respectively. In contrast, segmenting the region into optimal size of AOI produced more accurate results, achieving R2 values of 0.96 for H and 0.92 for LE, with corresponding RMSE values of 56.9 and 35.88 W/m², respectively. These findings emphasize the critical role of AOI size and endmember identification in improving SEBAL model accuracy and enhancing ET estimation for the sustainable management of water resources at both local and regional levels.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"139 ","pages":"Article 104514"},"PeriodicalIF":7.6,"publicationDate":"2025-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143816227","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Assessing change point detection methods to enable robust detection of early stage Artisanal and Small-Scale mining (ASM) in the tropics using Sentinel-1 time series data","authors":"Mensah Isaac Obour ,&nbsp;Barrett Brian ,&nbsp;Cahalane Conor","doi":"10.1016/j.jag.2025.104525","DOIUrl":"10.1016/j.jag.2025.104525","url":null,"abstract":"<div><div>Artisanal and Small-Scale mining (ASM) provides essential livelihoods for many in developing countries but often lacks regulation, leading to environmental issues such as water pollution and deforestation. Timely and accurate mapping of ASM activities is vital for responsible mining that benefits the environment and local communities. Synthetic Aperture Radar (SAR) is crucial for regular ASM monitoring in cloudy regions due to its ability to penetrate clouds. However, atmospheric effects can limit its effectiveness, particularly with shorter wavelengths in wet tropical areas during the rainy season. This study utilised a time series smoothing technique to improve Sentinel-1 (S-1) SAR time series data, reducing SAR noise and atmospheric effects from heavy rainfall for early ASM activity detection. We tested three change point detection (CPD) methods, including cumulative sum (CuSuM), pruned exact linear time (PELT), and binary segmentation (BinSeg) in the Western and Ashanti wet regions in southern Ghana using the smoothed S-1 data for early ASM detection. We observed a relatively fast response of ASM activity tracking when utilising smoothed S-1 data at both sites for VV and VH polarizations during the rainy seasons. However, VH polarization is more effective than VV polarization during rainy seasons. While all CPD algorithms showed similar performance, CuSuM had the shortest lag time of up to 9 days, compared to 11 days for PELT and BinSeg. This method significantly reduces ambiguity caused by heavy rainfall when identifying change points due to ASM activity, making it a viable option for near real-time monitoring in wet tropical regions.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"139 ","pages":"Article 104525"},"PeriodicalIF":7.6,"publicationDate":"2025-04-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143799634","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"ESMII-Net: An edge-synergy and multidimensional information interaction network for remote sensing change detection","authors":"Yixin Chen ,&nbsp;Xiaogang Ning ,&nbsp;Ruiqian Zhang ,&nbsp;Hanchao Zhang ,&nbsp;Xiao Huang ,&nbsp;You He","doi":"10.1016/j.jag.2025.104507","DOIUrl":"10.1016/j.jag.2025.104507","url":null,"abstract":"<div><div>In recent advancements, deep learning-based methods for change detection have demonstrated rapid capabilities to identify alterations across extensive regions, underscoring significant research and application potential in remote sensing change detection. Nonetheless, these methods currently encounter limitations in feature extraction, often leading to blurred edges and challenges in identifying small-scale changes. To overcome these challenges, we introduce the Edge-Synergy and Multidimensional Information Interaction Network (ESMII-Net) specifically designed for remote sensing change detection. We achieve feature enhancement through the Multidimensional Information Interaction Fusion Module (MIIFM) and, by integrating the edge aware decoder and the Edge-Synergy Module (ESM), guide the model to acquire effective edge information, thereby improving change detection performance. Furthermore, during the loss function formulation, we have incorporated a Small Object Enhancement Factor (SOEF) to prioritize small object detection. An edge-awareness map is also utilized within the model to accurately delineate change edges and assess their influence on adjacent changed pixels. The efficacy of our model and its innovative components has been validated through experimental results on two public datasets, showcasing improved capabilities in detecting edges and small objects.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"139 ","pages":"Article 104507"},"PeriodicalIF":7.6,"publicationDate":"2025-04-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143791089","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"DMP-PUNet: A novel network for two-dimensional InSAR phase unwrapping under severe noise and complex fringes conditions","authors":"Yu Chen ,&nbsp;Shuai Wang ,&nbsp;Yandong Gao ,&nbsp;Yanjian Sun ,&nbsp;Jinqi Zhao ,&nbsp;Kun Tan ,&nbsp;Peijun Du","doi":"10.1016/j.jag.2025.104519","DOIUrl":"10.1016/j.jag.2025.104519","url":null,"abstract":"<div><div>In the processing of Interferometric synthetic aperture radar (InSAR) data, two-dimensional (2-D) phase unwrapping (PU) is crucial for ensuring the quality of InSAR data inversion. Traditional methods, based on the assumption of phase continuity, often struggle with abrupt terrain changes and the influence of severe noise, leading to poor performance or failure. To address these challenges, this paper presents a dilated multi-path phase unwrapping network (DMP-PUNet) for 2-D PU under conditions of severe noise and complex fringes. To train this model, we developed a multi-effect interferometric phase simulation (ME-IPS) strategy that aims to simulate interferometric phases that closely resemble real-world conditions by comprehensively considering various factors, including terrain and digital elevation model (DEM) errors, atmospheric turbulence, vegetation effects, baseline geometry, multiple scattering, and noise. This simulation, combined with quasi-real interferometric phase data obtained from DEM inversion algorithms, forms the comprehensive training dataset. Finally, experiments on simulated data, quasi-real data, the InSAR-DLPU dataset, and InSAR data demonstrate that DMP-PUNet outperforms existing methods. For simulated data, DMP-PUNet achieved an overall average mean absolute error (MAE) in residuals of 0.221 rad, improving accuracy by 54.75 % with an average processing time of 0.81 s. For quasi-real data, the average MAE was 0.320 rad, a 119.06 % increase in accuracy, with an average processing time of 0.82 s. For the InSAR-DLPU dataset, overall, the MAE of DMP-PUNet was 20.34 % to 64.96 % lower than that of the best-performing baseline method (DLPU), with an average processing time of 1.90 s. For InSAR data, DMP-PUNet performed stably, with lower noise levels, smooth phase transitions, and deformation spatial patterns and profile shapes that conform to the laws of mining subsidence, averaging a processing time of 1.71 s, outperforming existing methods.</div></div>","PeriodicalId":73423,"journal":{"name":"International journal of applied earth observation and geoinformation : ITC journal","volume":"139 ","pages":"Article 104519"},"PeriodicalIF":7.6,"publicationDate":"2025-04-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143791085","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}