Table of Content

25 May 2026, Volume 0 Issue 5

Previous Issue   

Research on the assessment of regional economic resilience and policy response mechanism driven by aerospace information technology

JIN Jianglei, YAN Haowen, ZHANG Xinyue, HU Wencheng

2026, 0(5):  1-6.  doi:10.13474/j.cnki.11-2246.2026.0501

Asbtract ( )   PDF (3079KB) ( )  

References | Related Articles | Metrics

[Purposes] Addressing the challenges of traditional economic resilience assessments,which rely on lagging and coarse-grained statistical data,making it difficult to achieve dynamic measurement and forward looking policy responses,this study explores the potential of space-air information technology in regional economic resilience assessment and policy response mechanisms.It aims to construct an integrated paradigm of “data-assessment-mechanism” to provide scientific support for consolidating achievements and preventing the risk of returning to poverty in poverty-stricken areas.[Methods] Taking 75 poverty-stricken counties in Gansu province as case studies,this research integrates multi-source space-air remote sensing and socio-economic data to build a three-dimensional resilience index system encompassing “exposure-sensitivity-adaptability”.It employs an entropy weight spatial autocorrelation model to assess the spatiotemporal evolution of resilience from 2015 to 2025,utilizes LSTM and multi-scenario simulations to predict resilience trends under different shocks,and designs a closed-loop policy response mechanism of “monitoring-warning-simulation-optimization”,which is validated through eight poverty-stricken counties.[Findings] Compared to traditional methods,the space-air information enhancement model improves sensitivity in drought impact monitoring by 22%,with a correlation coefficient of 0.82 between the change rate of nighttime lights and GDP growth rate.The resilience pattern is “high in the southeast and low in the northwest”,with the northern parts of Linxia Hui Autonomous Prefecture and Dingxi city being “low-low” agglomeration areas.The resilience of the eight counties that have recently been lifted out of poverty is generally low.Vegetation coverage and soil moisture are key resilience indicators.Precise interventions based on space-air information have increased the county resilience index by 28%under extreme drought conditions,with “industry diversification+technology promotion” showing the best effect in drought-affected agricultural counties.[Conclusions] Space-air information technology can effectively compensate for the shortcomings of traditional data,achieving an improvement from post-assessment to pre-warning.This study provides scientific tools for resilience research in poverty-stricken areas and is of great value in improving the theoretical and policy informatics data foundation of economic resilience.

Key technologies and applications of integrated space-air-ground remote sensing monitoring for Shanghai's territorial space

ZHANG Wen, YAO Wenqiang, ZHAO Feng

2026, 0(5):  7-11,16.  doi:10.13474/j.cnki.11-2246.2026.0502

Asbtract ( )   PDF (3231KB) ( )  

References | Related Articles | Metrics

[Purposes]To meet the demand for modernized megacity governance and address traditional territorial space monitoring challenges (wide coverage,rapid changes,cumbersome supervision),this paper elaborates on key technologies and application scenarios of an integrated satellite-air-ground remote sensing monitoring system.[Methods]Based on the framework of “three-dimensional perception-intelligent driving-platform empowerment-scenario innovation”,the paper deeply integrate multi-source remote sensing data (satellite,aerial,UAV fleets).[Findings]Using multi-source heterogeneous data fusion and AI-driven intelligent interpretation,rapid full-domain response to territorial space changes is achieved.Taking Shanghai as a case,technical verifications and practical explorations are conducted in cultivated land protection,ecological security,and immovable cultural relic monitoring.[Conclusions]This method is advanced and practically applicable,providing a reference for refined urban territorial space governance and application scenario innovation nationwide.

A skyline detection algorithm based on improved YOLO11

YANG Gang, WANG Miao, CHEN Si, ZHOU Quan, LI Jiangchuan

2026, 0(5):  12-16.  doi:10.13474/j.cnki.11-2246.2026.0503

Asbtract ( )   PDF (1871KB) ( )  

References | Related Articles | Metrics

[Purposes] Skyline detection plays an important role in geolocalization,flight control,visual navigation,etc.The appearance of the sky and non-sky areas are variable,because of different weather or illumination environment,which brings challenges to skyline detection.[Methods]For these challenges,we proposes the YUNet algorithm,which improves the YOLO11 architecture to segment the sky region and extract the skyline in complicated and variable circumstances.In this research,the YOLO11 architecture is extended as an UNet-like architecture,consisting of an encoder,neck and decoder submodule.The encoder extracts the multi-scale features from the given images.The neck makes fusion of these multi-scale features.The decoder applies the fused features to complete the prediction rebuilding.To validate the proposed approach,the YUNet is tested on Skyfinder,CH1 datasets for segmentation and skyline detection,respectively.[Findings] The test shows that the IoU of YUNet segmentation can reach 0.986,and the average error of YUNet skyline detection is just 1.36 pixels.[Conclusions] YUNet has an excellent performance and speed.And it can complete the sky segment and skyline detection task in the complex environment,which is valuable for engineering applications.

Registration-uncertainty-aware instance-level building damage assessment from pre-disaster optical and post-disaster SAR imagery

LI Jiajun, FANG Yilin, DUAN Wenxi

2026, 0(5):  17-21.  doi:10.13474/j.cnki.11-2246.2026.0504

Asbtract ( )   PDF (2039KB) ( )  

References | Related Articles | Metrics

[Purposes]To address the limited availability of high-quality post-disaster optical imagery,as well as the significant modality gap and registration uncertainty between pre-disaster optical imagery and post-disaster SAR imagery,this study proposes a robust modeling method for instance-level building damage assessment to improve the reliability of building-level damage recognition in complex disaster scenarios.[Methods]Building instance priors are first constructed from pre-disaster optical imagery,while a damage evidence field is extracted from post-disaster SAR imagery.On this basis,a candidate alignment modeling and probabilistic evidence aggregation mechanism is introduced to achieve building-level evidence attribution under registration uncertainty.Combined with damage grading and uncertainty calibration,the proposed framework performs instance-level building damage assessment in a unified manner.Main experiments,ablation studies,robustness tests under registration perturbations,and cross-event generalization experiments are conducted on the BRIGHT dataset.[Findings]The results show that the proposed method achieves strong performance in instance-level segmentation accuracy,building-level damage discrimination,and prediction reliability.Specifically,it reaches an mAP of 0.66,a F1 of 0.65,and an ECE of 0.02.Compared with the variant without alignment modeling,the full model improves mAP by about 0.01 and F1 by about 0.05.[Conclusions]This study reformulates the task as a building-level evidence attribution problem under registration uncertainty,rather than a simple cross-modal feature fusion problem.The proposed framework provides a robust and interpretable technical solution for instance-level building damage assessment using pre-disaster optical and post-disaster SAR imagery.

Feasibility study of 3D modeling based on UAV oblique photogrammetry

HAN Tingting, LI Xiaoqiang, FAN Xianchuang, OUYANG Yali

2026, 0(5):  22-26,31.  doi:10.13474/j.cnki.11-2246.2026.0505

Asbtract ( )   PDF (5123KB) ( )  

References | Related Articles | Metrics

[Purposes] To solve the problems of insufficient accuracy and low efficiency in real-scene 3D modeling of complex terrain at water control projects,and to promote the in-depth application of oblique photogrammetry in smart water conservancy construction.[Methods] Taking the Wanjiazhai Water Control Project on the Yellow River as the research object,DJI M300 UAV is used for full-area oblique photography and P1 camera for close-range supplementary photography in key areas to obtain multi-view images with 2 cm ground resolution.Based on DP Modeler software,a refined workflow of “full-area inspection-classification and refinement-monolithic modeling-model fusion” is established to eliminate model defects and construct a surveying and mapping-grade real-scene 3D model.[Findings] The modeling results show that the model has realistic texture and complete geometry,and can accurately capture structural deformation of hydraulic structures.Both horizontal and vertical accuracy meet the requirements of water conservancy engineering surveying and mapping.The operation time is shortened by 60% and modeling cost reduced by more than 30%.[Conclusions] The study indicates that UAV oblique photogrammetry for 3D modeling can effectively solve the problem of high-precision modeling of water control projects in complex canyon terrain,providing technical support and practical reference for digital supervision of water conservancy projects,intelligent construction of hydropower stations,and intelligent upgrading of water conservancy infrastructure.

Integrating multiple factors for automatic planning and service capability assessment of cross-provincial low-altitude visual flight routes in the Sichuan-Chongqing region

ZENG Yixiao, WU Di, ZHOU Tao, CHEN Yu, LI Xiaolong, LI Han

2026, 0(5):  27-31.  doi:10.13474/j.cnki.11-2246.2026.0506

Asbtract ( )   PDF (1712KB) ( )  

References | Related Articles | Metrics

[Purposes] As a core region with high urbanization level and great development potential in western China,the Sichuan-Chongqing region is faced with severe challenges to the safety of inter-provincial low-altitude flight due to complex terrain,changeable meteorological conditions and high-density population distribution.[Methods] To address this problem,this paper proposes an automatic planning method for inter-provincial low-altitude visual flight routes coupling multiple factors such as terrain environment,low-altitude meteorology and airspace control,and constructs an evaluation model of route service capacity based on space syntax theory.[Findings]The results show that the integrated method coupling multi-factor elements can generate inter-provincial low-altitude visual flight routes adapting to complex environments and realize quantitative evaluation of route service capacity.[Conclusions]This method can provide technical support for low-altitude flight safety guarantee and high-quality development of regional low-altitude economy.

Intelligent analysis method for UAV inspection of well-facilited farmland infrastructure maintenance

ZHANG Zhihua, HU Zhaopeng, DING Penghui, LI Zhigang, ZHAO Qian

2026, 0(5):  32-37.  doi:10.13474/j.cnki.11-2246.2026.0507

Asbtract ( )   PDF (2491KB) ( )  

References | Related Articles | Metrics

[Purposes]Well-facilitated farmland infrastructure currently suffers from inadequate post-construction maintenance.While UAV inspection combined with AI analysis offers a viable solution,it is hindered by issues such as insufficient scenario adaptability and suboptimal detection performance.[Methods]This paper proposes a scenario-based,multi-level visual detection framework.For common defects,taking field roads as a case study,we employ the YOLO11 object detection model and introduce the Wise-IoU loss function to enhance detection accuracy.For complex anomalies in specific facilities,taking outlet protection piers as an example,we integrate the precise localization capability of YOLO11 with the semantic understanding of the vision-language model (VLM)Qwen3-VL to construct a few-shot learning detection method.[Findings]Experiments demonstrate that on the field road defect dataset,the proposed method achieves an 8.3 percentage point improvement in mAP₅₀ over the baseline model.On the outlet protection pier dataset,the precision and recall rates reach 86% and 96%,respectively.[Conclusions]This research provides a flexible and reliable analysis method for UAV inspections of well-facilited farmland infrastructure,holding significant promise for engineering applications.

Estimation of vegetation biomass in expressway road area: taking Taiyuan-Linxian Expressway in Shanxi province as an example

LIANG Shiqi, LIAN Xugang

2026, 0(5):  38-43.  doi:10.13474/j.cnki.11-2246.2026.0508

Asbtract ( )   PDF (1960KB) ( )  

References | Related Articles | Metrics

[Purposes] Developing a high-precision method for estimating vegetation biomass in expressway road corridors is of great significance for quantifying the ecological carbon sequestration capacity of highway transportation systems.[Methods] Taking the Taiyuan-Linxian Expressway in Shanxi province as the study area,a multi-source remote sensing collaborative approach for vegetation biomass inversion in expressway road areas was proposed.A UAV-scale biomass inversion model was constructed based on UAV LiDAR point clouds and multispectral data to obtain high-precision UAV plot-based biomass reference values.The UAV plots were then standardized to 10 m resolution to generate training samples,which were combined with Sentinel-2 vegetation indices to construct a satellite-scale biomass inversion model.[Findings] The results show that the UAV-scale biomass estimation model achieved high accuracy (R²=0.85).Among the satellite-scale inversion models,the Random Forest model performed best (R²=0.68).[Conclusions] By constructing UAV-based plots and integrating satellite remote sensing data,efficient and accurate estimation of vegetation biomass in expressway road corridors can be achieved,providing a reliable method for expressway ecological benefit assessment and carbon stock research.

Confidence-guided multi-modal Transformer for rice extraction in cloud-prone regions

WANG Junqiang, SUN Zhenhui, MENG Qingyan, ZHANG Linlin

2026, 0(5):  44-49,79.  doi:10.13474/j.cnki.11-2246.2026.0509

Asbtract ( )   PDF (1870KB) ( )  

References | Related Articles | Metrics

[Purposes]Optical remote sensing images of cloudy areas are susceptible to cloud contamination,leading to decreased crop classification accuracy.Synthetic aperture radar (SAR)possesses all-weather imaging capabilities and can complement optical data.This paper proposes a confidence-guided multi-modal Transformer rice extraction method.[Methods]Firstly,based on HLS optical and Sentinel-1 radar time-series data,This method uses cloud masks to estimate the confidence of optical data at each time phase.Then,a Transformer encoder extracts the temporal features of both optical and radar data separately,and a confidence-guided gating fusion mechanism is designed to adaptively fuse the two.Furthermore,a self-supervised temporal reconstruction strategy is introduced,which enhances the model's ability to compensate for SAR information when optical data is missing by performing mask reconstruction on some optical data,thus improving model robustness.[Findings] Experiments show that the proposed method achieves an overall accuracy of 88.00% and a F1 score of 88.25%,outperforming comparative models such as random forest,LSTM,and Transformer.[Conclusions] It effectively improves the rice extraction accuracy in cloudy areas and provides a reference for crop classification under complex climatic conditions.

Identification of 3D urban spatiotemporal evolution patterns and pathways based on human-housing-economy coupling: a case study of Southwest China

SONG Yuxin, WANG Hao, YANG Lan, LIU Caijuan, DU Jun, ZHAO Jun

2026, 0(5):  50-55,71.  doi:10.13474/j.cnki.11-2246.2026.0510

Asbtract ( )   PDF (1719KB) ( )  

References | Related Articles | Metrics

[Purposes] To deepen the understanding of urban evolution processes,this study identifies 3D urban spatio-temporal evolution patterns and pathways in Southwest China from the perspective of human-housing-economy coupling.[Methods] Multi-source data,including building height,building density,population,and GDP,were integrated to construct a multi-period differential indicator system,and multi-temporal K-means clustering together with state sequence analysis was employed for identification.[Findings] From 2005 to 2020,urban evolution in Southwest China experienced a stage-wise process from low-speed steady development,to intensified construction expansion,and then to the coexistence of slowing incremental growth and renewal-oriented quality improvement.Persistent stagnation accounted for the largest share and was mainly distributed in Xizang,Western Sichuan,and Western Yunnan,while steady development and growth attenuation were more common around core cities such as Chengdu and Guiyang.[Conclusions] The integration of multi-source data can reveal the spatio-temporal characteristics of the coexistence of coordination and mismatch among population agglomeration,housing development,and economic growth,providing support for urban evolution monitoring and regional development analysis in complex terrain regions.

Research progress on the concept and key technologies of pan-spatial information system

YIN Xiaoling, BAI Jingwen, YANG Ji, FENG Huihui, LI Yong, DING Xiaohui, HU Hongda, HUANG Wumeng, HOU Zhiwei, WANG Wenpei

2026, 0(5):  56-63.  doi:10.13474/j.cnki.11-2246.2026.0511

Asbtract ( )   PDF (3402KB) ( )  

References | Related Articles | Metrics

[Purposes]In response to the limitations of traditional geographic information systems which primarily focus on Earth's surface and can no longer meet the increasingly complex cross-scale application demands,this study focuses on the pan-spatial information system (PSIS),an important emerging direction in GIS theory.It summarises the background of PSIS development in the context of challenges faced by GIS in the era of big data and analyses its core concepts and modelling theories.[Methods]By systematically reviewing research achievements on PSIS over the past decade,this study identifies and synthesises its main characteristics,key technologies,and typical application cases.[Findings]The findings reveal that PSIS is characterised by diverse spatio-temporal object types,complex and heterogeneous computational environments,highly variable application scenarios,and comprehensive and efficient information processing capabilities.PSIS has already been widely applied in fields such as smart cities and natural resource management.[Conclusions]Future research on PSIS should focus on three main aspects:strengthening theoretical foundations,integrating cutting-edge technologies,and expanding application domains.

An improved atrous convolution method for polarimetric synthetic aperture radar data classification

ZHANG Jichao, GAO Zishan, ZHANG Bing

2026, 0(5):  64-71.  doi:10.13474/j.cnki.11-2246.2026.0512

Asbtract ( )   PDF (3142KB) ( )  

References | Related Articles | Metrics

[Purposes] Aiming at the problem that traditional polarization synthetic aperture radar image interpretation methods are difficult to accurately identify ground objects through limited physical models,this paper proposes an improved module based on atrous spatial pyramid pooling convolution.[Methods] The improved module consists of four parallel branches.These four branches process data through different convolution methods,reducing certain computational parameters while ensuring model performance.The improved four branches can effectively extract multi-scale features and complete the learning ability of key information through the convolutional block attention module.[Findings] The experimental results show that the average row accuracy rate of the improved method in water body classification is 91.2%,which is superior to the comparison network.It also performed well in vegetation classification,with an average row accuracy rate of 82%.[Conclusions] This indicates that by increasing the depth of the network,the learning ability of deep learning models for complex data can be enhanced.

Time-series InSAR monitoring and cause analysis of surface subsidence in Suzhou supported by multi-source data

LI Zihao, SUN Chengzhi, BI Lingyu, QIAO Shen

2026, 0(5):  72-79.  doi:10.13474/j.cnki.11-2246.2026.0513

Asbtract ( )   PDF (4205KB) ( )  

References | Related Articles | Metrics

[Purposes]The increasingly frequent occurrence of urban surface subsidence can damage above-ground or underground structures,hinder regional construction and resource development,and effective monitoring of surface deformation is crucial for the management and prevention of urban ground subsidence.[Methods]This study uses 36 Sentinel-1A images covering the main urban area of Suzhou from January 2021 to December 2023.Based on SBAS-InSAR technology,ground subsidence in the main urban area of Suzhou was monitored,obtaining the annual average deformation rate and cumulative subsidence during this period.In addition,combining historical optical images,subway construction data,and precipitation data,the causes of subsidence in the study area were analyzed in detail.[Findings]The results show that the overall deformation rate in the study area ranges from -42.6 to 21.9 mm/a,with a maximum cumulative subsidence of 127 mm,and ground subsidence is influenced by both natural and human factors.[Conclusions]Ground subsidence in Suzhou is affected by multiple factors.Natural factors include the soft soil distribution and rainfall,while human factors include subway construction and operation,increased building loads,and road construction.

Large-scale mining-area subsidence monitoring and analysis using L-SAR data and D-InSAR

ZHENG Meinan, QU Lijia, SHEN Yongtian, ZHANG Xiang, CHENG Zisu

2026, 0(5):  80-88.  doi:10.13474/j.cnki.11-2246.2026.0514

Asbtract ( )   PDF (4025KB) ( )  

References | Related Articles | Metrics

[Purposes]The purpose of this paper is to validate the capability of domestically produced L-band synthetic aperture radar (L-SAR)data for large-scale deformation monitoring in mining areas.[Methods]It takes the Huainan mining district as an example,using six ascending-track and ten descending-track L-SAR scenes,a two-pass differential interferometric synthetic aperture radar (D-InSAR)approach is applied to retrieve large-area surface deformation.[Findings]Results show that the relatively long wavelength of L-SAR helps mitigate decorrelation effects; the mean coherence of the ascending-and descending-track L-SAR interferograms is 0.34 and 0.37,respectively,which ensures the reliability of the D-InSAR measurements.Line-of-sight (LOS)surface deformation derived from ascending and descending tracks is highly consistent in spatial distribution,with maximum cumulative subsidence of -0.65 and -0.69 m,respectively,occurring in the Gubei mining area.Profile-based analysis of cumulative deformation indicates that the deformation magnitude and spatial extent recovered from the two track directions are essentially consistent,but a spatial “shift” between the subsidence bowls detected by ascending and descending data is observed; this shift is mainly at-tributable to horizontal ground motion and geocoding offsets.[Conclusions]The study confirm the ability of L-SAR data to capture large-scale mine subsidence and provide a technical reference for using L-SAR and D-InSAR for mining-area subsidence monitoring.

Spatio-temporal dynamics of lakes in the Yellow River basin from 1986 to 2024

ZOU Yebin, LEI Haonan, WU Yun, CONG Peiyu, PAN Pan

2026, 0(5):  89-96.  doi:10.13474/j.cnki.11-2246.2026.0515

Asbtract ( )   PDF (2313KB) ( )  

References | Related Articles | Metrics

[Purposes]Addressing the problems of insufficient long-term dynamic monitoring of lakes in the Yellow River basin,unclear area-size structure,and poorly understood seasonal fluctuation mechanisms,this study utilized Google Earth Engine and Landsat imagery from 1986 to 2024 to systematically extract lake water bodies using an improved DSWE algorithm.[Methods]The reliability of the method was confirmed through accuracy assessment (overall accuracy 96.95%,Kappa coefficient 0.94).We systematically analyzed the interannual changes,spatial distribution,size-class structure,and seasonal fluctuations in lake number and area,and employed the Mann-Kendall trend test to evaluate change trends,thereby revealing the spatiotemporal evolution patterns of lakes over the past 40 years.[Findings]The results indicate that:①The number of lakes in the basin increased by an average of 46.64 per year,with a significant expansion in total area (Z=7.45,P<0.001),primarily contributed by the upstream region; ②The lake structure exhibits a “scissor-shaped”pattern where a few large lakes (≥50 km²),comprising only 0.12%of the total number,dominate the area (contributing 51.40%),while the vast majority are small lakes;③Small lakes (<1 km²)show significant seasonal fluctuations,whereas large lakes remain relatively stable.[Conclusions]This research provides a scientific basis for water resource management and ecological protection in the Yellow River basin.

Panchromatic and multi-spectral remote sensing image fusion algorithm based on 3D Curvelet transform and structural similarity

WANG Yu, LU Bangyang, SHI Xue, LI Mengmeng

2026, 0(5):  97-102.  doi:10.13474/j.cnki.11-2246.2026.0516

Asbtract ( )   PDF (2175KB) ( )  

References | Related Articles | Metrics

[Purposes]To address the challenge of simultaneously achieving high spatial detail and rich spectral information for traditional remote sensing image fusion methods,this paper proposes a fusion method based on high-frequency detail enhancement and low-frequency feature optimization.[Methods]Firstly,the 3D and 2D Curvelet transforms are used to decompose multi-spectral and panchromatic images to obtain the corresponding high and low frequency coefficients.Then,a component substitution strategy is employed to construct the fusion rule for high-frequency coefficients,and a band-adaptive adjustment coefficient is introduced to optimize the high-frequency fused coefficients.For the low-frequency fusion,a rule is established based on the similarity features between the multi-spectral and panchromatic low-frequency coefficients.Additionally,the Laplacian features of the panchromatic low-frequency coefficients are used to enhance the contour information of the multi-spectral image,further optimizing the low-frequency fused coefficients.Finally,the optimized high-frequency and low-frequency fused coefficients are reconstructed through the inverse 3D Curvelet transformation to obtain the fused multi-spectral image.[Findings]Experimental results on GF-2,QuickBird and WorldView-3 remote sensing images demonstrate that the proposed method achieves significant improvements in both visual perception and quantitative metrics.[Conclusions]It effectively enhances image sharpness and detail representation while maintaining high spectral fidelity.

A semantic segmentation method for urban remote sensing images based on visual state space models

CHEN Chong, YANG Yang

2026, 0(5):  103-109.  doi:10.13474/j.cnki.11-2246.2026.0517

Asbtract ( )   PDF (2182KB) ( )  

References | Related Articles | Metrics

[Purposes] To address the problems of large-scale variation,blurred boundaries,and category confusion in complex urban remote sensing images,a semantic segmentation method based on visual state space models is proposed.[Methods] A dual-branch collaborative encoder is designed to integrate global contextual information and local multi-scale features,and a cross-branch collaboration mechanism is introduced for dynamic feature interaction.A state space-driven progressive decoding strategy is employed to restore high-resolution semantic representations.[Findings] Experiments on typical urban remote sensing images of Changsha show that the proposed method achieves an overall accuracy (OA)of 91.04%,a mean intersection over union (mIoU)of 73.46%,and a mean F1-score (mF1)of 84.18%,outperforming RS3Mamba by 0.93,1.08,and 0.91 percentage points,respectively.More stable performance is observed for structural classes such as roads and buildings.[Conclusions] The results demonstrate that the proposed method effectively improves segmentation accuracy and robustness in complex urban scenes,providing a feasible technical approach for fine interpretation of high-resolution remote sensing images.

Multi-UAV cooperative coverage path planning based on DDQN algorithm

LI Cailing

2026, 0(5):  110-116.  doi:10.13474/j.cnki.11-2246.2026.0518

Asbtract ( )   PDF (6455KB) ( )  

References | Related Articles | Metrics

[Purposes]To optimize the task completion time of multi-UAV collaborative operations and enhance the efficiency of path planning,a multi-UAV collaborative path planning method based on the double deep Q-network(DDQN)algorithm is proposed.[Methods]Firstly,an efficient environmental information map fusion technology is designed,which can quickly and efficiently mark the coverage records of each UAV and detect the positions of obstacles.Secondly,the DDQN algorithm is introduced to minimize the task time,avoid path overlap,area omission and potential collisions.Additionally,a new collaborative learning mechanism is constructed,which can efficiently plan the globally optimal path.Finally,a simulation platform is built to simulate the proposed method and two typical methods are selected for comparative analysis.[Findings]The experimental results show that compared with the other two comparison methods,the proposed method demonstrates superior performance in terms of task completion time and coverage efficiency.[Conclusions]The proposed method can quickly adapt to unknown obstacles and complex environments,achieve efficient and comprehensive coverage of the target area in various task scenarios,and has good response speed and coverage rate.

Research on GNSS+INS navigation and positioning method in high-altitude environments

WANG Yangsheng, FENG Yaxin, LIU Huaying, XUE Jian

2026, 0(5):  117-121,148.  doi:10.13474/j.cnki.11-2246.2026.0519

Asbtract ( )   PDF (2633KB) ( )  

References | Related Articles | Metrics

[Purposes] In densely forested mountainous environment,inter-system bias(ISB)between BDS and GPS significantly increases global navigation satellite system (GNSS)positioning errors.This prevents GNSS from providing high-precision position corrections to the inertial navigation system (INS).Consequently,positioning errors accumulate in GNSS+INS integrated navigation,severely degrading the performance of UAV power line inspection.[Methods] To address this issue,this paper proposes a dynamic adaptive Kalman filtering algorithm that accounts for BDS+GPS ISB.A measurement equation for BDS and GPS ISB in integrated positioning is established based on GNSS+INS dynamic adaptive fusion.A GNSS quality-check strategy is introduced to eliminate observation outliers,thereby significantly improving the accuracy and robustness of state estimation.[Findings]Experimental results demonstrate that the proposed ISB dynamic Kalman filter effectively enhances GNSS+INS position estimation performance.The RMSE are less than 10 cm horizontally and below 8 cm vertically,outperforming both the EKF and the AKF.[Conclusions]These findings provide a valuable design reference for UAV power inspection systems.

AI remote sensing identification model for artificially disturbed patches in Longyan,Fujian province

GU Zhujun, LIU Jia, MAI Xianzhi, WU Jiasheng, YUE Hui, LIN Gengen, HE Yanzi, CAO Zhengjin, LIAO Guanghui

2026, 0(5):  122-127,142.  doi:10.13474/j.cnki.11-2246.2026.0520

Asbtract ( )   PDF (5452KB) ( )  

References | Related Articles | Metrics

[Purposes]In response to the monitoring needs of artificially disturbed land in Longyan,Fujian province,optimize the architecture of the AI remote sensing recognition model for artificially disturbed land and apply it to promote sustainable development of the ecological environment.[Methods]This research constructs a artificially disturbed sample based on 2 m high-resolution imagery from Longyan,Fujian,using six model architectures:DeepLabV3+,PAN,SegFormer,U-Net++,SCSE-UNet,and TransUNet,to develop AI recognition models and conduct remote sensing intelligent recognition and analysis.[Findings]TransUNet has the best overall performance,with an intersection over union of 0.75 and an F1-Score of 0.84,SegFormer follows closely.In the recognition application in Longyan city in 2024,the overall accuracy (OA)of TransUNet is 0.99,the Kappa coefficient is 0.87,the producer's accuracy (Q_pa) is 0.95,and the user's accuracy (Q_ua) is 0.81,achieving superior recognition accuracy and contour detail depiction.[Conclusions]The identification results of TransUNet align with the actual human activity space,with disturbances concentrated in urbanization construction,mining,and along road construction lines,providing efficient and accurate technical support for dynamic monitoring of soil and water conservation in Longyan city.

InSAR monitoring of short-term rapid slip in open-pit mine: a case study of tailings reservoir No.2 and 4,Dexing copper mine

LU Li, XU Xiaobo, QIN Yousen

2026, 0(5):  128-135.  doi:10.13474/j.cnki.11-2246.2026.0521

Asbtract ( )   PDF (5839KB) ( )  

References | Related Articles | Metrics

[Purposes]Tailings reservoir slides and their secondary hazards have become one of the major geological hazards in open-pit mines.Large-scale,short-critical and long-time dynamic monitoring of tailings reservoir landslide hazards is the basis for smart mine construction management and safe exploitation of mineral resources,and is also a powerful guarantee for preventing and warning the occurrence of landslide geological hazards in open-pit mining.[Methods]In this paper,we combine short-critical D-InSAR and long-time sequence PS-InSAR.Based on low-medium resolution Sentinel-1 satellite SAR wide mode images from November 2020 to March 2021,and taking tailings reservoirs No.2 and 4 of Dexing copper mine as the study area,it is used to monitor and analyse the short-time displacement characteristics,long-time displacement trends and influencing factors of the tailings reservoir in terms of slip profiles,PS single point positioning and time-series cumulative slip.Based on the coherence coefficient of SAR image interference,the large-scale slip deformation of tailings reservoir slope is analyzed.[Findings]The monitoring results show that the the tailings reservoir slip is mainly related to the dam elevation gradient and precipitation.Tailings reservoir No.2 slip occurs in the tailings sand reservoir,with a maximum slip rate is -439 mm/a.Tailings reservoir No.4 slip is mainly distributed in the upper half of the dam body,and the slip rate gradually increases from the top of dam to the central drainage channel,with a maximum rate of -619 mm/a.The time series results show that the slip initially starts near the central drainage channel of the dam body,and the maximum cumulative slip is -220 mm.The coherence coefficient method can effectively monitor large-scale slip of tailings reservoir slope.[Conclusions]The results show that the combined InSAR monitoring technique and coherence coefficient method can provide technical support for the large-area,short-period and long time series slip monitoring of the open-pit tailings reservoir.

3D modeling and precision control of plateau mountain photovoltaic field areas based on UAV-SLAM air-ground synergy

YU Shihui, HU Jianliang, MA Xiaobo, HAN Wenqiang, GAO Sha, CHEN Jie, ZHOU Hanwang, ZHANG Ji

2026, 0(5):  136-142.  doi:10.13474/j.cnki.11-2246.2026.0522

Asbtract ( )   PDF (6162KB) ( )  

References | Related Articles | Metrics

[Purposes]With China's carbon peak target proposed,photovoltaic power generation has gradually become a key pathway for advancing low-carbon transformation.However,conducting three-dimensional modeling and accuracy control studies using traditional surveying methods in high-altitude mountainous photovoltaic plant areas is hindered by complex terrain,making it difficult to efficiently and accurately acquire data.This impedes modeling quality and the implementation of accuracy control research.[Methods]To investigate the applicability of point cloud data collected by UAVs and handheld laser scanners for 3D modeling and accuracy control in high-altitude mountainous PV sites,a pilot study was conducted at the Qubei county high-altitude mountainous PV site in Wenshan Prefecture.The LiDAR360 system was employed for point cloud fusion processing.A comparative analysis using both Context Capture and DJL Terra platforms established a “point-plane-multi-level” accuracy evaluation system.[Findings]Experimental results indicate:Correlation coefficients and maximum deviations across x,y,and z dimensions exhibit weak negative to positive correlations,with maximum deviations consistently within 10 cm.-Modeling efficiency improved by 87%using the DJL Terra platform in the selected test areas,though CC modeling showed significant mosaicking and photovoltaic panel corner point deficiencies.Point distribution was concentrated between 0.059 and 0.113,yielding overall satisfactory results.Finally,parameters,modeling time,and cost considerations for both processing platforms were evaluated.[Conclusions]This study provides quantitative references for establishing digital 3D real-scene modeling and precision control research plans for plateau photovoltaic fields.

A methodology for illegal mining quantification in open-pit areas via integrated application of multi-source techniques

YANG Liping, ZHANG Jiye, CHENG Yu, LI Yin

2026, 0(5):  143-148.  doi:10.13474/j.cnki.11-2246.2026.0523

Asbtract ( )   PDF (3379KB) ( )  

References | Related Articles | Metrics

[Purposes]Aiming at the problems associated with concealed illegal open-pit mining activities,such as strong concealment,difficulties in evidence collection,and high accuracy requirements for mineral resource quantification, this study proposes an integrated identification method combining multiple technologies.These include UAV LiDAR surveying,USV-based underwater topographic mapping,ground-penetrating radar (GPR)detection,GNSS measurement,and geological investigation.[Methods] Taking a suspected illegal mining case in XY county,Jiangsu province as an example,high-precision digital elevation models were acquired using UAV LiDAR,underwater terrain data of water-filled pits were obtained with a USV-mounted single-beam bathymetric system,and the thickness and bottom boundaries of backfilled areas were detected via GPR.Comprehensive analysis was conducted incorporating historical remote sensing images,enabling precise delineation of the extent and depth of illegal excavation.The volume of the suspected illegally mined ore body was estimated using the geological block method,combined with rock physico-mechanical property tests,leading to an accurate determination of the quantity of illegally extracted mineral resources.[Findings]The research demonstrates that this method achieves precise verification of mineral resource volumes across the entire space of illegal open-pit mining areas,integrating land and water surfaces as well as above-ground and subsurface dimensions.[Conclusions]The results are objective,scientific,and reliable.The proposed method can provide robust technical support for law enforcement supervision and judicial identification.

Representation strategies of high-precision map for advanced intelligent driving

DENG Guangran, ZHANG Yongli, FU Xiao, TAO Lan, OUYANG Xinqiu, HAN Jianzi, CHEN Zejia, DONG Guangsheng

2026, 0(5):  149-154.  doi:10.13474/j.cnki.11-2246.2026.0524

Asbtract ( )   PDF (3865KB) ( )  

References | Related Articles | Metrics

[Purposes] High-precision map serves as a core support for advanced intelligent driving,yet its large-scale application faces dual challenges: ensuring driving safety and safeguarding geographic information security.This study systematically examines the core representational requirements of high-precision map and proposes a multi-dimensionally optimized representation strategy.[Methods] The approach incorporates three key technical innovations: expanded map coverage,hierarchical categorization of feature elevations,and decomposition of continuous terrain fluctuations.Simulations,field tests,and case validations are conducted across various typical scenarios.[Findings] The results show that,on the premise of not breaching the bottom line of geospatial information security,this strategy can reduce the safety accident rate of high-level intelligent driving to about 1/10 of the original rate,meeting its safety requirements for misjudgment rate of feature ground objects and ground undulation error.[Conclusions] Based on this strategy,Guangzhou has successfully completed the map review work for five high-precision map projects,with the review efficiency improved by 75%,providing important support for the compliant application of high-precision map nationwide and the large-scale development of the intelligent connected vehicle industry.

Research on the geographic information security guarantee mechanism of spatio-temporal data in intelligent connected vehicles

CHEN Zhuoning, LIU Changchang, HOU Xiaoyu, MA Zhaoting, HE Feng, PAN Zhongkai

2026, 0(5):  155-160.  doi:10.13474/j.cnki.11-2246.2026.0525

Asbtract ( )   PDF (1612KB) ( )  

References | Related Articles | Metrics

With the rapid development of intelligent connected vehicles(ICVs), the spatio-temporal data generated has become a core element supporting the developement of autonomous driving and the vehicle-road-cloud integration. However, it also faces security risks such as geographic information leakage and data tampering.Focusing on the geographic information security issues of ICVs spatio-temporal data, this article summarizes the various types of spatiotemporal data involved in vehicle-road-cloud integration.Targeting the whole-process data processing activities of ICVs spatio-temporal data including collection, transmission, aggregation,fusion, distribution and application, it designs a data security guarantee mechanism that meets the requirements of geographic information security.It provides support for the safe and compliant use of large-scale and high-quality spatio-temporal data in the industry, and offers suggestions for the development direction of spatio-temporal data security in intelligent connected vehicles.

Intelligent production method of basic geographic entity data based on multi-agent collaboration

PENG Song, WU Hao, YU Youqi, LI Zongze, LIU Yujun

2026, 0(5):  161-168.  doi:10.13474/j.cnki.11-2246.2026.0526

Asbtract ( )   PDF (2723KB) ( )  

References | Related Articles | Metrics

[Purposes]To address bottlenecks such as fragmented workflows,high dependence on manual labor,and quality control difficulties in basic geographic entity production,this study explores a new production model to satisfy the large-scale and high-timeliness update requirements of new-type fundamental surveying and mapping.[Methods]A LLM-driven multi-agent collaborative production method is proposed.A collaborative system comprising task orchestration,entity recognition,attribute relationship,and quality inspection agents is established to achieve the adaptive deconstruction of production goals and the execution of intelligent action chains.[Findings]Application in an experimental area shows that the system can generate diverse categories of geographic entities and realize an intelligent closed-loop covering the entire process from task orchestration to quality inspection.An agile quality control mode of “simultaneous production,inspection,and correction” has been successfully implemented.[Conclusions]This approach provides a feasible technical path and reference paradigm for the intelligent transformation of fundamental surveying and mapping,effectively improving production efficiency and data consistency.

Application of AI-assisted high-resolution satellite remote sensing in natural resource regulation enforcement in Guiyang

PEI Zhigang, AO Chenghuan

2026, 0(5):  169-173.  doi:10.13474/j.cnki.11-2246.2026.0527

Asbtract ( )   PDF (2266KB) ( )  

References | Related Articles | Metrics

[Purposes]To evaluate the efficacy of AI-augmented high-resolution satellite remote sensing in enhancing the efficiency and accuracy of natural resource regulation enforcement in Guiyang.[Methods]A human-machine collaborative technical framework was established by integrating commercial high-temporal-resolution imagery with deep learning and GIS technologies.The system operationalized a “change identification-clue filtration-enforcement verification” pipeline,wherein AI automatically extracted potential change indicators followed by manual verification and legality assessment to detect four violation categories: construction,paved surfaces,water bodies,and earthwork operations.AI-driven change detection demonstrated 1155-fold efficiency gain over manual interpretation.Detection accuracy descended in the following order: water bodies>buildings>paved surfaces>earthworks.Violations were identified within 20 days post-occurrence.The empirical data facilitated over 200 enforcement cases,resulting in the reclamation of 1555 acres of cultivated land,431 acres of permanent prime farmland,and 328 acres of ecological red zones.[Findings]AI-derived change indicators showed 100% spatial concordance with national supervisory data.[Conclusions]AI-enhanced remote sensing significantly improves natural resource enforcement efficacy,with human-machine collaboration representing the optimal paradigm for monitoring complex mountainous terrains.

Research and exploration on the dual-blended practice teaching model for surveying and mapping engineering in higher education

ZHANG Min

2026, 0(5):  174-178.  doi:10.13474/j.cnki.11-2246.2026.0528

Asbtract ( )   PDF (1441KB) ( )  

References | Related Articles | Metrics

[Purposes]Addressing the “three highs and three difficulties”(high cost,high difficulty,high risk; difficult implementation,difficult observation,difficult reproduction)in practical teaching for surveying and mapping majors,as well as the inadequacy of traditional blended learning in meeting students' online skill acquisition needs,this study explores a novel teaching model tailored to cultivating high-quality technical talents through the deep integration of virtual simulation technology with education.[Methods]Firstly,the core connotation of the “dual-blended”practical teaching model is explicitly proposed,namely the integration of virtual and real environments and the blending of online and offline approaches.Secondly,the content framework of this model is designed from three dimensions—resources,activities,and evaluation—while the teaching process is restructured into three stages: “pre-learning,mid-term reinforcement,and post-transfer.”Finally,the model's feasibility and effectiveness are validated through teaching practice.[Findings]The results demonstrate that this model effectively overcomes the bottleneck of online practical teaching,establishes a learning environment supporting students' repeated skill training,facilitates the systematic integration of their knowledge and skill systems,and achieves the organic unity of three-dimensional teaching objectives: knowledge acquisition,skill application,and professional competence.[Conclusions] As both an inheritance and innovation of traditional blended learning,this model provides an operable theoretical framework and practical pathway for cultivating technical talents in higher education institutions.

Indoor automatic 3D modeling technology integrating handheld SLAM LiDAR and image data

ZHANG Shijie, MIAO Peipei, XU Linjie, GAO Yunlong, PENG Fangyuan, LI Chunnan, SUN Xinhao

2026, 0(5):  179-184.  doi:10.13474/j.cnki.11-2246.2026.0529

Asbtract ( )   PDF (4626KB) ( )  

References | Related Articles | Metrics

[Purposes] To address the dual requirements of high-precision geometry and realistic textures for indoor 3D modeling,as well as the limitations of single data sources and the inefficiency of traditional registration methods,this paper proposes an efficient,robust,and highly automated indoor 3D modeling technology.[Methods] A handheld laser SLAM scanner integrated with a camera is used to simultaneously collect high-precision point clouds and texture images.Initial POS data are derived from factory calibration parameters.An improved point cloud-image registration algorithm combining geometric constraints and feature matching is proposed to achieve high-precision data fusion.Subsequently,3D reconstruction is accomplished through point-cloud-driven geometric modeling and optimized texture mapping.[Findings] Experimental results indicate that the geometric error of the 3D model is less than 2 cm,with realistic textures.The entire process takes 58.5 min,significantly outperforming traditional manual modeling methods.[Conclusions] This technology leverages the complementary strengths of LiDAR point clouds and image data,reduces sensitivity to lighting conditions,improves modeling efficiency and automation,and provides a reliable technical solution for indoor realistic 3D reconstruction.