Table of Content

25 August 2025, Volume 0 Issue 8

Previous Issue   

Precision analysis and positioning evaluation of satellite-based precise point positioning service in the Antarctic region

LIU Yang, CHAI Hongzhou, WANG Min, ZHOU Yingdong, SUN Shuang, ZHANG Qiankun

2025, 0(8):  1-6.  doi:10.13474/j.cnki.11-2246.2025.0801

Asbtract ( )   HTML ( )   PDF (3117KB) ( )  

References | Related Articles | Metrics

Satellite-based precise point positioning service offers high-precision positioning in the environment where terrestrial networks are hard to cover,yet there are few applications in polar regions.Based on the measured HAS data of China's 40th Antarctic Expedition and the real-time precision orbit and clock deviation products of MADOCA and CNES broadcast on the network.This paper assesses the availability,accuracy and PPP performance of different products' orbit and clock correction of GPS and Galileo in the Antarctic region.The results demonstrate that GPS and Galileo correction products provided by HAS,MADOCA and CNES are highly accurate and can satisfy the requirement of centimeter-level positioning accuracy,reflecting the applicability of these real-time products in the Antarctic region.Both HAS and MADOCA can provide stable and reliable PPP services,with a horizontal positioning accuracy of less than 0.2 m and a vertical positioning accuracy of less than 0.4 m,conforming to the service standards.This provides a theoretical and practical basis for the application of satellite-based precision single-point positioning technology in polar regions and other high latitudes.

Research on the calibration of aerial cameras in polar environments: taking Antarctica as an example

WANG Weixuan, CUI Yingchun, LI Bingrui, ZHANG Hao, CHEN Shaocong

2025, 0(8):  7-12,31.  doi:10.13474/j.cnki.11-2246.2025.0802

Asbtract ( )   PDF (8830KB) ( )  

References | Related Articles | Metrics

Currently,China primarily utilizes satellite remote sensing and unmanned aerial vehicles (UAVs)to acquire image data in Antarctic regions.While satellite remote sensing offers high efficiency,its image resolution remains relatively low.Conversely,UAV-collected imagery achieves higher precision but suffers from lower operational efficiency and limited accessibility to many areas.These constraints have resulted in low geographic information digitization levels and inadequate high-resolution image coverage in China's Antarctic operations.To enhance both efficiency and accuracy in image data acquisition,during China's 40th Antarctic expedition,an aerial camera was successfully transported and installed on the “Snow Eagle 601”fixed-wing aircraft.The acquisition of high-precision calibration files plays a crucial role in ensuring the accuracy and validity of image data captured by this aerial camera in Antarctica,significantly impacting the quality of final data products.This study focuses on processing calibration data collected by the aerial camera at different flight altitudes from China's Zhongshan Station through distinct methodologies employing control points and precise ephemeris.By conducting accuracy verification and comparative analysis of results obtained from different processing approaches,reliable high-precision calibration files will be established.These outcomes will provide essential accuracy assurance for Antarctic aerial photography missions and serve as critical data support and theoretical foundation for future development of standardized calibration fields.

Analysis and application of BDS-3 new frequency PPP continuity in East Antarctica

MAO Wenbin, YAN Hao, XU Junming, LIU Shiji

2025, 0(8):  13-18.  doi:10.13474/j.cnki.11-2246.2025.0803

Asbtract ( )   PDF (2239KB) ( )  

References | Related Articles | Metrics

Addressing the polar positioning performance of the BDS-3 new frequency,this paper analyzes the B1C/B2a dual-frequency combined precise point position(PPP)accuracy of the BDS-3 system and the coordinate time series variation trend for each tracking station,based on multi-year (2021—2025)measured data from four tracking stations in the East Antarctica region.Experimental results show that BDS-3 offers good satellite availability in East Antarctica,with an average of 9 satellites available and PDOP values below 3.The static PPP performance is excellent,with horizontal positioning accuracy better than 1 cm and vertical positioning accuracy better than 1.5 cm.Each tracking station's coordinates exhibit a significant shift trend,with a maximum annual shift rate reaching the centimeter level,mainly influenced by local ice flow speed and mass changes.Similarly,the study's results can be used to infer ice flow speed and mass changes,providing new references and continuous support for the practical application of BDS-3 new frequency in polar regions.

Ground target localization method combining unmanned aerial vehicle pose information

LUO Qingli, ZHANG Shubin, JIANG Xintao, WEI Jujie, GAN Jun

2025, 0(8):  19-25.  doi:10.13474/j.cnki.11-2246.2025.0804

Asbtract ( )   HTML ( )   PDF (10468KB) ( )  

References | Related Articles | Metrics

Traditional methods for ground target localization using single-frame drone images typically require at least four prior control point information.However,acquiring control points becomes increasingly challenging as their quantity increases,presenting an arithmetic growth problem.This paper proposes a ground target localization method that integrates unmanned aerial vehicle(UAV)pose information.The method combines UAV pose information with three ground control points to establish a geometric optical model.This model enables the mapping relationship between ground target points and UAV image pixels,facilitating the determination of latitude and longitude coordinates for each ground target point.Consequently,precise ground target localization based on a single-frame drone image is achieved,reducing the required number of control points and enhancing localization accuracy.The experimental results indicate that the average positioning accuracy of this method reaches 1.45 pixels,which is 4.61 pixels higher than that of the traditional PnP four-point target positioning method.

A non-contact measurement method for cultural heritage and its accuracy analysis

DING Keliang, GAO Qixuan, Lü Ruizhe, ZHANG Xi, ZHENG Chengze

2025, 0(8):  26-31.  doi:10.13474/j.cnki.11-2246.2025.0805

Asbtract ( )   PDF (8247KB) ( )  

References | Related Articles | Metrics

Considering the non-contact characteristics of safety monitoring for cultural heritage buildings,this paper proposes a non-contact continuous monitoring method.The method first establishes a 3D virtual spatial grid according to the specific conditions of the monitoring object,then constructs a rigorous surveying reference network based on the strict spatial geometric relationships between the spatial grid and ground monitoring benchmarks.Continuous monitoring is achieved using the non-contact measurement function of high-precision total stations.The deformation of monitoring points corresponding to spatial grid nodes is determined through the rigorous spatial geometric relationships between spatial grid nodes and ground benchmarks,combined with distance changes from each measurement.Simulations and actual monitoring cases verify the feasibility and effectiveness of this method,which provides a practical monitoring solution for cultural heritage monitoring.This method is also applicable to continuous monitoring of other similar buildings and structures.

Ground deformation monitoring and influence factors analysis of the Gaizi valley near the Karakoram highway based on SBAS-InSAR technology

MO Dandan, HUO Jiuyuan

2025, 0(8):  32-42.  doi:10.13474/j.cnki.11-2246.2025.0806

Asbtract ( )   HTML ( )   PDF (14386KB) ( )  

References | Related Articles | Metrics

The Karakoram highway (KKH) in China and Pakistan has complex geological conditions, peculiar and variable climate, and landslide hazards are frequent along the route, so the investigation and monitoring study of landslide hazards in this region is of great importance for disaster prevention and mitigation. In this study, the small baseline subsets InSAR (SBAS-InSAR) technique is used in combination with optical remote sensing images to monitor the surface deformation and analyze the time-series deformation characteristics of the Gaizi valley section of the China-Pakistan highway.Based on 64 scenes of Sentinel-1 image data covering the study area, the SBAS-InSAR technique is used to obtain deformation distribution maps and time-series deformation features of the study area over the time span. The deformation rate values of the radar line of sight (LOS) in the study area from March 2017 to August 2022 ranged from -33.5 to 11.6 mm/a, with a maximum cumulative deformation of 179.4 mm. On this foundation, the accuracy of the deformation detection results is verified by combining the optical remote sensing images of four typical landslide areas in the region and previous research results, demonstrating that SBAS-InSAR technology is an effective tool for deformation monitoring of landslide hazards. The time series deformation curves are analyzed by using monthly average precipitation, monthly average temperature, monthly maximum temperature, monthly minimum temperature, surface soil moisture data, glacier distribution data, earthquake catalogue, IGBP land cover data and so on. Furthermore, the influence of different factors on the surface deformation of landslides in the study area is explored to provide a scientific basis for early identification and prevention of disasters.

Forest aboveground biomass mapping in the Greater Mekong Subregion using multi-source remote sensing data fusion

YUAN Lili, YANG Xinwei, LI Menghua, CHEN Yuquan, TANG Bohui

2025, 0(8):  43-47.  doi:10.13474/j.cnki.11-2246.2025.0807

Asbtract ( )   PDF (6381KB) ( )  

References | Related Articles | Metrics

Accurate estimation of forest aboveground biomass density is crucial for advancing sustainable forest management.This study focuses on the Greater Mekong Subregion (GMS)and utilizes spaceborne global ecosystem dynamics investigation(GEDI),Sentinel-1,Sentinel-2,and auxiliary datasets to extract 52 feature variables.By applying the LightGBM machine learning model,a 1 km resolution forest aboveground biomass density map of the GMS is generated.The results indicate that the LightGBM model achieved R²=0.65,RMSE=38.11 Mg/hm²,and EA=72.03%.Across the study area,biomass density ranged from 15.16 to 423.87 Mg/hm².The derived biomass product demonstrated strong correlation with the GEDI L4B product (R²=0.52,RMSE=61.91 Mg/hm²).In conclusion,open-access earth observation (EO)data exhibits significant potential for estimating forest aboveground biomass.

Comparative analysis of undifferenced and uncombined precise point positioning performance for BDS-3 dual-frequency data

ZHOU Mingduan, CUI Likun, MENG Mingzhi, XIE Qianlong, LI Yueyao, SONG Qiao, YU Runxin

2025, 0(8):  48-54,61.  doi:10.13474/j.cnki.11-2246.2025.0808

Asbtract ( )   PDF (1737KB) ( )  

References | Related Articles | Metrics

In view of the fact that BDS-3 broadcasting B1I/B3I/B1C/B2a signals and the comparative analysis of PPP positioning performance for the current B1I+B3I and B1C+B2a data is still relatively scarce.In this paper,based on the establishment of the BDS-3 undifferenced and uncombined PPP positioning model,an implementation flowchart of integer ambiguity fixed algorithm for BDS-3 undifferenced and uncombined PPP is decuced in detail.The BDS-3 undifferenced and uncombined PPP positioning analysis software (short as UDUC_PPP)using C/C++programming language based on Visual Studio 2022 development platform is designed and developed applied to compare and analyze the positioning performance of BDS-3 undifferenced and uncombined PPP for B1I+B3I and B1C+B2a data.Supported by BDS-3 B1I/B3I/B1C/B2a signals from fourteen representative stations with globally distributed in the MGEX experimental network for 060 d for 24 hours consecutive observation data in 2024 are selected for the positioning performance analysis,the results show that the convergence time and the positioning accuracy of BDS-3 undifferenced and uncombined PPP fixed resolution are better than the float resolution positioning for not only B1I+B3I data but also B1C+B2a data,in which the integer ambiguity resolution success rate is 92.5%above after the positioning convergence for BDS-3 undifferenced and uncombined PPP for both of B1I+B3I and B1C+B2a data,meanwhile the positioning accuracy of BDS-3 undifferenced and uncombined PPP fixed resolution is better than 0.9 cm in the horizontal-RMSE and 2.3 cm in the point-RMSE respectively.For both float resolution and fixed resolution,BDS-3 undifferenced and uncombined PPP positioning in terms of convergence time for B1C+B2a data is better than for B1I+B3I data,which the positioning accuracy is basically comparable.

Spherical target robust fitting under point cloud degradation and its application in power infrastructure measurement

XU Haoxuan, MAO Qingzhou, ZHANG Xu, FENG Zhiqiang, WU Anlei, LI Deyang

2025, 0(8):  55-61.  doi:10.13474/j.cnki.11-2246.2025.0809

Asbtract ( )   PDF (9047KB) ( )  

References | Related Articles | Metrics

Spherical targets are used commonly for point cloud registrations collected by terrestrial laser scanning.To research the applicability of existing methods for robust spherical targets fitting under degradation conditions,four targets with two radii are used to test multiple robust fitting methods at different distances (10～120 m)and integrities (10%～50%).The average radius,spherical fitting accuracy,and distances between fitted center are used as indicators to evaluate the effectiveness of robust fitting methods.The results show that the methods based on angle weighting has the best effect on the larger spherical target,while the method with distance weighting has better effect on the smaller.IGGⅢ method with distance weighting controls the fitting errors under centimeter level in the range of 70 m.Spherical targets with a completeness of 40%～50% can be used for high-precision registration within a range of 70 m,while 50 m for completeness of 30%～40%.The targets with completeness below 20% cannot be used.The registration result of power infrastructure with 4 directions of views further confirms the above conclusion.

Virtual assembly method of bridge components based on point cloud feature extraction

WANG Lei, LI Sheng, DING Xiaoping, LI Ruijie, ZHANG Yuyuan, DENG Wen

2025, 0(8):  62-69.  doi:10.13474/j.cnki.11-2246.2025.0810

Asbtract ( )   PDF (6995KB) ( )  

References | Related Articles | Metrics

Aiming at the current virtual preassembly technology in the application of large-scale girder bridge project,which has the problems of huge data volume,not applicable to bolted structure,and only considering 1+1 splicing,etc,the virtual assembly method of steel girder sections based on point cloud feature extraction is proposed with Shapotou Yellow River highway bridge as the engineering background.The virtual assembling method uses the strategy of local+overall scanning to solve the problem of long data acquisition time and large computing capacity,and the denoising and downsampling of point cloud data is completed by SOR algorithm and spatial sampling algorithm,etc.In order to carry out virtual assembling of bolted components,the planar identification segmentation,spatial point cloud projection downsampling,three-dimensional point cloud circular hole fitting and circular hole center coordinate extraction methods are proposed based on the RANSAC fitting algorithm.Finally,to the problem that only 1+1 splicing is considered in the virtual assembly of components,an overall virtual preassembly method is proposed based on the ICP algorithm from the processing deviation to the splicing deviation.The engineering application shows that the maximum error of scanning recognition and measurement is 0.15 mm,and the qualification rate of virtual assembly of the segment is 97%,which is more comprehensive and accurate than the traditional method; this method has the advantages of high automation,fast simulation efficiency,and is applicable to the whole structure,and it can provide technical support for the virtual assembly of steel girder segments of large bridges.

Remote sensing evaluation of spatiotemporal characteristics of ecological restoration for river basins

WANG Yan, XIAO Haiping, ZHANG Xinjie

2025, 0(8):  70-75,82.  doi:10.13474/j.cnki.11-2246.2025.0811

Asbtract ( )   PDF (2936KB) ( )  

References | Related Articles | Metrics

In recent years,the upper reaches of the Gan River basin have faced serious ecological and environmental problems such as soil erosion,water pollution,and soil erosion.Since 2016,Ganzhou has continuously applied for and implemented ecological protection and restoration work for mountains,rivers,forests,fields,and lakes.Therefore,conducting remote sensing assessments of ecological quality before and after restoration is of great significance.This article is based on Landsat images from 2012 to 2022,and combines the modified normalized forel ule index(MNFUI)and kernel normalized difference vegetation index (kNDVI)to construct the modified remote sensing environmental index (MRSEI).The Mann-Kendall (M-K)test is used to analyze the spatiotemporal trends of ecological environment quality in the Ganjiang River Basin.The results showed that: ①Compared with RSEI,MRSEI contains richer information on the water color status of the monitoring area and alleviates the problem of vegetation index oversaturation,making it more suitable for the upper reaches of the Gan River basin with a large water area. ②The mean MRSEI values of the study area from 2012 to 2022 were 0.79,0.71,0.72,0.74,0.75,and 0.77,respectively.The overall trend showed a sharp decline followed by a gentle rise,which was highly coordinated with the watershed ecological restoration launched in 2016 and verified the effectiveness of ecological restoration in improving environmental quality.③The range of local autocorrelation “high-high” and “low-low” agglomeration areas in the upper reaches of the Ganjiang River basin showed significant changes over time,with the “high-high” agglomeration areas mainly distributed at the source of the basin,while the “low low” agglomeration areas highly overlapped with the built-up areas of cities in southern Jiangxi.The research results can provide data reference for ecosystem restoration work in the Ganjiang River basin.

Monitoring of TLS point cloud deformation in large underground caverns

WANG Haofan, LI Biao, LI Tao, XIAO Peiwei, QIAN Hongjian, XU Nuwen

2025, 0(8):  76-82.  doi:10.13474/j.cnki.11-2246.2025.0812

Asbtract ( )   PDF (5218KB) ( )  

References | Related Articles | Metrics

Deformation control in large underground caverns may pose a serious threat to personnel safety and engineering progress.Deformation monitoring of underground cavern is of great significance to prevent engineering disasters.To solve the problem of low deformation monitoring efficiency and incomplete information in large-scale underground cavern engineering,a deformation observation technology based on TLS point cloud is proposed in this paper.This technology includes semi-automatic point cloud noise reduction combining RANSAC Shape Detection algorithm and surface variation,and calculation of caverns surface deformation based on M3C2 algorithm,which can realize comprehensive and efficient monitoring of large-scale underground caverns deformation.This technique is applied to monitor the support deformation in a typical area of the main building of Xulong Power station,and it is found that there are obvious deformation bands in the downstream side arch of Yc0+140 to Yc0+170 during the frequent construction stage,and the results are consistent with the traditional deformation monitoring results on site.The observation results provide more comprehensive three-dimensional deformation information for deformation control of large underground caverns and improve the efficiency of deformation monitoring.

Genetic algorithm for stable updating of point feature annotation in geographic animations

WEI Zhiwei, YANG Nai, DING Su, CHEN Yebin, GUO Renzhong

2025, 0(8):  83-88,94.  doi:10.13474/j.cnki.11-2246.2025.0813

Asbtract ( )   PDF (3429KB) ( )  

References | Related Articles | Metrics

This study focuses on the issue of point feature label updates in geographic animations and proposes an optimized label placement method based on a genetic algorithm.The method aims to improve the temporal stability of label placements in geographic animations,preventing frequent changes and conflicts in label positions between frames.By analyzing the constraints of label placement,this paper takes into account various factors such as label overlap,positional priority,association,and temporal stability,and introduces an adaptive genetic algorithm to optimize the label placement in geographic animations.Additionally,to validate the proposed method,a prototype tool for geographic animation production was developed.Experimental results demonstrate that the proposed method effectively reduces the variation in label positions between frames in geographic animation,ensuring visual consistency in label placement,though it slightly increases the algorithm's execution time.

Target vehicle trajectory deduction method based on weighted Bayesian network

BIAN Yuxia, ZHU Zijie, ZHOU Ye, LI Xinyi

2025, 0(8):  89-94.  doi:10.13474/j.cnki.11-2246.2025.0814

Asbtract ( )   PDF (1476KB) ( )  

References | Related Articles | Metrics

In view of the increasingly complex structure of the road network and the continuous increase in traffic flow in the modern urban traffic environment,the trajectory of vehicles in a large transportation network has become extremely complex.The existing traffic data collection and analysis methods face challenges in deducing the complete driving trajectory of vehicles.In order to highly restore the real driving trajectory of vehicles and improve the degree of coordination of various traffic data,this study proposes a target vehicle trajectory deduction method based on weighted Bayesian network.Specifically,the topology of the road traffic network is directly mapped to the Bayesian network architecture,the influencing factors affecting driving decisions are extracted and quantified,and the weight analysis method is used to determine the weight coefficients of the influencing factors,so as to construct a weighted Bayesian network model for vehicle trajectory deduction.Taking the campus road network as the sample area for experiments,the proposed method can more accurately deduce and reproduce the complete driving trajectory of the target vehicle in a wide range of traffic network,which provides strong theoretical support for solving the problems of intelligent transportation construction,route planning and vehicle tracking.

Construction of multimodal population spatialization model via IVYA-SIAM joint optimization and its driving effect analysis

WANG Lizhi, XIAO Dongsheng

2025, 0(8):  95-99,106.  doi:10.13474/j.cnki.11-2246.2025.0815

Asbtract ( )   PDF (9550KB) ( )  

References | Related Articles | Metrics

Aiming at the precision bottleneck and insufficient spatial heterogeneity analysis in existing models due to single-algorithm dependence,this study proposes a three-tier “multimodal ensemble-parameter adaptation-feature enhancement” optimization framework.First,multi-source data (such as nighttime lighting,building outlines)are integrated to construct a secondary model (N-MLP)via stacking random forest,XGBoost,and MLP.Then,the IVY algorithm (IVYA)is introduced for dynamic hyperparameter optimization,and a spatial interaction-augmented attention mechanism (SIAM)is designed to enhance geographical spatial dependence analysis through parallel attention architectures.Finally,a dual-scale validation system (400 m grid and township/street levels)is established in Chengdu,and the driving effect of population distribution on drone logistics demand is analyzed via a low-altitude economy demand elasticity model.Results show that the optimized SIAM-IVYA-N-MLP model achieves an R² of 0.947 9 at the grid scale,with MAE and RMSE reduced by 14.67%and 3.38%,respectively.At the township/street scale,the R² reaches 0.971 6.A 1%increase in main urban population density drives a 1.19%growth in drone logistics demand.This study provides an operational technical pathway for high-precision population spatialization and low-altitude economic infrastructure planning.

An efficient image dataset selection method based on the greedy algorithm and spatial relationship self-optimization

DONG Siyuan, YANG Yuanwei, GAO Xianjun, TAN Meilin, DU Bin, QU Weijun, CHEN Ningsheng

2025, 0(8):  100-106.  doi:10.13474/j.cnki.11-2246.2025.0816

Asbtract ( )   PDF (3015KB) ( )  

References | Related Articles | Metrics

The massive accumulation of remote sensing data makes it particularly important to efficiently and accurately retrieve high-quality image collections that meet the application requirements of the region of interest.To address the large number of retrieved images with high overlap,which makes it difficult to meet application requirements directly,this paper proposes an efficient image dataset selection method based on a greedy algorithm and self-optimization of spatial relationships.Firstly,the image dataset is pre-filtered based on image quality and intersection relationships.Then,a block-based selection is performed using a greedy algorithm.Finally,the quality of the selection results is enhanced through an optimization strategy that integrates the greedy selection results with the topological relationships of the region of interest.The results show that the selection results of the proposed method have significantly lower redundancy and higher utilization than other methods.This method can efficiently select a high-quality image dataset that fully covers the region of interest,featuring recent imaging times,low cloud cover,high resolution,and fewer images.

A method for assessing the similarity of linear features oriented towards coastal currentness analysis

MA Mengkai, DONG Jian, JI Ran, XIE Tian, WANG Dong

2025, 0(8):  107-111,117.  doi:10.13474/j.cnki.11-2246.2025.0817

Asbtract ( )   PDF (1258KB) ( )  

References | Related Articles | Metrics

This paper takes the updating of nautical charts as an example in the context of geospatial data updates,discussing the requirements and methods for currentness analysis during the process of geospatial data updates.It emphasizes the critical role of quantitatively assessing the currentness of coastlines in nautical chart updates and analyzes the limitations of existing currentness assessment procedures and traditional similarity assessment methods.In response to these situations,this paper proposes a method for assessing the similarity of linear features oriented towards coastal currentness analysis.This method mainly consists of three steps: Firstly,a resampling method considering the accuracy of linear features is designed for preprocessing the update data. Secondly,a workflow for extracting and transforming shape features of linear features is established to ensure consistency of feature extraction results across multiple scenarios such as mirroring,rotation,scaling,and translation. Finally,a criterion for judging the similarity of linear features based on shape feature extraction results is proposed,achieving quantitative comparison between updated and original data.Experimental results show that this method has strong robustness and the ability to match local line segments overall.By extracting the shape features of linear features,it realizes the quantitative evaluation of linear feature similarity,providing effective technical support for nautical chart updates.

UAV-based object recognition dataset for coastal sewage outfalls

YIN Junjie, GUAN Daiwanjing, LI Hao, ZHANG Xiaoyang, MA Yujie, XING Hanfa

2025, 0(8):  112-117.  doi:10.13474/j.cnki.11-2246.2025.0818

Asbtract ( )   HTML ( )   PDF (3975KB) ( )  

References | Related Articles | Metrics

The identification of coastal sewage outfalls is a crucial aspect of marine supervision,providing essential safeguards for the ecological and resource security of marine areas.Addressing the current challenges of insufficient specialized datasets and the lack of precision in target recognition algorithms for coastal sewage outfall detection using unmanned aerial vehicle (UAV)imagery,this study constructs a high-quality dataset of coastal sewage outfalls and proposes an enhanced detection method based on the improved YOLOv8n model.Initially,focusing on the coastal region of Yangjiang city,Guangdong province,the study employs UAVs to capture images at various altitudes,establishing a comprehensive dataset that encompasses diverse characteristics of sewage outfalls.Subsequently,the YOLOv8n model is augmented with the SimAM parameter-free attention mechanism to refine feature extraction and fusion,alongside the integration of NWD and CIoU loss functions to address issues of boundary ambiguity and target overlap.Experimental results demonstrate that the enhanced model surpasses the original in terms of precision,recall rate,and mAP,achieving an mAP of 98.27%.This research offers an intelligent solution for monitoring coastal sewage outfalls,contributing technological support for marine supervision and pollution control.

Hyperspectral and multi-spectral data fusion method combined with deep spatio-spectral-temporal features

PAN Chen, WANG Xiaochu, WANG Zhiwei

2025, 0(8):  118-122.  doi:10.13474/j.cnki.11-2246.2025.0819

Asbtract ( )   PDF (5480KB) ( )  

References | Related Articles | Metrics

To address the limited spatial resolution of hyperspectral satellite imagery,this study proposes a data fusion method driven by deep spatio-spectral-temporal features.The method integrates the rich spectral information of hyperspectral images with the fine spatial details of multi-spectral data,aiming to generate fused imagery with both high spectral and spatial resolution.The fusion network is built upon a generative adversarial network (GAN)architecture,with an optimized feature fusion strategy that significantly enhances the network's capability in handling multi-resolution data.Experiments conducted on a comprehensive dataset,comprising both spaceborne and airborne hyperspectral imagery,demonstrate that the proposed method notably improves image quality,outperforming conventional approaches in terms of spatial detail preservation and spectral consistency.Quantitative evaluation using multiple metrics further confirms the robustness and effectiveness of the method.This study provides essential technical support for the enhancement and application of hyperspectral remote sensing imagery,offering important theoretical and practical value.

Optimal planning algorithm for the verification route of discrete patches in cultivated land intercropped with trees based on “space-air-ground-spectrum”

SHI Guigang, DU Xiaoxue, PAN Yanxi, ZOU Tao, GAO Yang, LIU Hang

2025, 0(8):  123-127,136.  doi:10.13474/j.cnki.11-2246.2025.0820

Asbtract ( )   PDF (1232KB) ( )  

References | Related Articles | Metrics

This paper delves into the critical issue of optimal planning algorithms for discrete patch verification paths in intelligent assessment of the status of farmland intercropped with trees based on the “sky-space-earth-spectrum” technology.By integrating multi-source data such as satellite remote sensing,drone aerial photography,artificial intelligence knowledge graphs and GIS,this paper proposes three intelligent path planning methods and analyzes their advantages,the methods aim to enhance the efficiency and accuracy of verifying the status of farmland intercropped with trees,providing scientific basis and technical support for natural resource investigation and monitoring,as well as farmland protection.

Underwater trash detection algorithm based on image enhancement and improved RT-DETR

LI Chao, LIU Qingyi, ZHANG Jiawei, SHI Yong, YANG Min

2025, 0(8):  128-136.  doi:10.13474/j.cnki.11-2246.2025.0821

Asbtract ( )   HTML ( )   PDF (8721KB) ( )  

References | Related Articles | Metrics

With the increasingly prominent problem of marine environmental pollution,rapid detection and cleaning of underwater garbage are particularly urgent.A new underwater garbage detection algorithm based on improved RT-DETR is proposed to address the issues of poor image quality,severe exposure to light,overlapping and varying shapes that lead to poor detection performance.Aiming at the problems of color cast and low contrast in images,an enhancement algorithm combining contrast enhancement and adaptive color compensation is designed for image preprocessing.In response to the demand for lightweight mobile device models,the FasterNet Block module is introduced to improve the backbone network and reduce the number of model parameters.To address the issue of weak lighting in underwater environments,the HS-FPN advanced filtering feature fusion pyramid fusion strategy is adopted to solve the problems of severe feature loss and low discrimination.For small targets in images,a GELAN generalized efficient layer aggregation network is adopted to improve the representation ability of the model.To address the issue of large differences in garbage size caused by spatial location,an Inner-ShapeIoU loss function combining Inner-IoU and ShapeIoU is introduced to improve the robustness of object detection.The experimental results show that the proposed method effectively solves the problems of image color cast and low contrast.Compared with the original model,the detection accuracy has been improved by 3.9 percent,and the number of parameters has been reduced by 26.3 percent.The underwater garbage detection performance is superior.

Leakage detection method for metro shield tunnels by fusing sparse convolution

SUN Zexin, ZHANG Anyin, DUAN Juju, JIANG Jundi, SHEN Yueqian, WANG Yibo

2025, 0(8):  137-141.  doi:10.13474/j.cnki.11-2246.2025.0822

Asbtract ( )   PDF (2784KB) ( )  

References | Related Articles | Metrics

Traditional convolutional neural networks face challenges in accurately detecting leakage in shield tunnels,particularly due to feature distortion and inefficiencies in handling sparse point cloud data.To address this,we propose a Sparse U-Net based leakage detection method,leveraging three-dimensional LiDAR point clouds.This method incorporates voxelization,hash table and rule-based sparse convolution operations to efficiently capture linear leakage features.An encoder-decoder architecture is employed for precise leakage segmentation,and Focal Loss is introduced to address class imbalance.Experimental results demonstrate the proposed method significantly improves both accuracy and computational efficiency,achieving increases of 5.52% in IoU and 3.41% in accuracy compared to traditional methods,providing an efficient and reliable solution for leakage detection in shield tunnels.

Intelligent collaborative DOM production technology based on remote sensing image production platform

WANG Yingmou, LI Lei

2025, 0(8):  142-148.  doi:10.13474/j.cnki.11-2246.2025.0823

Asbtract ( )   PDF (9843KB) ( )  

References | Related Articles | Metrics

This paper investigates the intelligent collaborative DOM production technology based on a remote sensing image production platform.This technology effectively addresses the issues of low efficiency,insufficient accuracy,and cumbersome processes associated with traditional DOM production by constructing a control point database and a multi-software intelligent collaborative platform.The establishment of the control point database enables the integration and efficient utilization of historical control point data,significantly reducing fieldwork and production costs.The multi-software intelligent collaborative platform combines the functional advantages of software such as INPHO,Tian Gong 2D integrated software,and DPGrid,achieving full-process optimization from aerial triangulation to DOM production.This effectively improves the geometric accuracy and visual effects of DOM while significantly increasing production efficiency.The experimental results show that this technology can effectively improve the efficiency and quality of DOM production,reduce fieldwork and costs,which has considerable value for promotion and application.

Real-scene 3D data acquisition and fusion technologies inside and outside caves:take Yixing Shanjuan Cave for an example

GUO Zhendong, WU Hao, GU Zhengdong, HUANG Liang

2025, 0(8):  149-152.  doi:10.13474/j.cnki.11-2246.2025.0824

Asbtract ( )   PDF (3762KB) ( )  

References | Related Articles | Metrics

Aiming at the insufficient research on modeling complex indoor and underground spaces in real-scene 3D construction,this paper proposes a 3D modeling method that integrates 3D point clouds and video imagery.Firstly,high-precision laser point cloud data inside the cave is acquired using SLAM technology,while multi-angle video imagery is collected via close-range photogrammetry.Then,the SFM algorithm is employed to generate dense matching point clouds,and the ICP algorithm is applied to achieve precise registration of heterogeneous data,constructing a 3D cave model with both structural features and texture information.Finally,the indoor model is fused with an outdoor terrain-level real-scene 3D model obtained from UAV oblique photogrammetry,forming a unified digital twin platform.The results demonstrate that this method achieves high-precision reconstruction and virtual-real integration of indoor and outdoor scenes,providing a reliable technical reference for modeling complex underground spaces.

GNSS-assisted UAV aerial photography technology for obtaining data for quality inspection of new surveying and mapping products

ZHANG Yongli, ZHU Wenchao, FAN Yewen, QU Zhi

2025, 0(8):  153-158.  doi:10.13474/j.cnki.11-2246.2025.0825

Asbtract ( )   PDF (2165KB) ( )  

References | Related Articles | Metrics

The development of new basic surveying and mapping has driven the innovation of quality inspection technologies.This paper aims to study the technology of acquiring quality inspection data for new surveying and mapping products without ground control points based on a new-type surveying and mapping equipment-the unmanned aerial vehicle (UAV)aerial photography system.Based on the relevant research achievements of the UAV aerial photography system and the standards related to aerial photography,this paper discusses the quantitative relationship between aerial photography flight parameters and the data accuracy of surveying and mapping products,and presents the corresponding mathematical model.At the same time,it analyzes three modes of GNSS-aided UAV aerial photography,and designs a ground control point-free UAV aerial photogrammetry scheme for acquiring data used in quality inspection of large-scale surveying and mapping products.The feasibility of the scheme combining the flight parameters designed by the mathematical model with GNSS-aided UAV aerial photography has been verified through the experiment in Luoding,Guangdong,which can obtain high-precision quality inspection data corresponding to the scale.

Application of BIM reverse modeling based on multi-source data fusion in the preservation of historic and cultural blocks

XING Wang, FANG Zheng, XU Yi, ZHANG Canghao, SUN Lianzeng, WANG Zhaoze

2025, 0(8):  159-163,178.  doi:10.13474/j.cnki.11-2246.2025.0826

Asbtract ( )   PDF (3507KB) ( )  

References | Related Articles | Metrics

Addressing the issues of low model completeness and unstable data accuracy in traditional modeling of existing buildings in historical and cultural districts,a technical system of “comprehensive acquisition-data fusion-intelligent reconstruction” is proposed.By utilizing 3D laser scanning and multi-view photogrammetry technologies,millimeter-level geometric frameworks and high-resolution texture information of buildings are obtained respectively.An improved SICP algorithm is employed to achieve precise fusion of multi-source point clouds.Finally,using BIM reverse modeling to construct a realistic 3D model containing building information.The results indicate that this method achieves millimeter-level geometric accuracy and over 98%completeness of the model,supporting multi-dimensional spatio-temporal information overlay analysis.It provides a full lifecycle solution for preventive conservation,virtual restoration,and revitalization of the district.

Identification of survey and monitoring land parcel changes based on semantic of geographic information

GUO Xiuli, WANG Guoliang, SHI Jing, SHANG Yongfu, DOU Xiaonan

2025, 0(8):  164-168.  doi:10.13474/j.cnki.11-2246.2025.0827

Asbtract ( )   PDF (4384KB) ( )  

References | Related Articles | Metrics

In natural resource monitoring work,there is an increasing demand for the frequency and timeliness of dynamic monitoring.In order to improve the timeliness of monitoring,this paper takes the semantic recognition of natural resources geographic information as the starting point,and establishes a set of land use relationship maps and semantic knowledge base.Through monitoring the relevant keywords on the Internet,the key information of changes in land parcel entities is obtained by comparing the knowledge maps,so as to realize automatic recognition of changes in monitored land parcels based on semantics.Combining the positioning function of basic geographic information,a set of rapid meaning discrimination mechanisms has been formed,which effectively supports the coordinated management of natural resources and ecological environment restoration work.

Accuracy verification methods for road high definition navigation electronic map

LIU Miao, LI Liang, DING Keliang, YUAN Lin

2025, 0(8):  169-173.  doi:10.13474/j.cnki.11-2246.2025.0828

Asbtract ( )   PDF (4192KB) ( )  

References | Related Articles | Metrics

High definition(HD)maps have entered a phase of rapid development,where the accuracy of map data directly determines the safety and reliability of autonomous driving systems.The accuracy verification of HD maps is critical for ensuring the quality of map products.This paper systematically elaborates on accuracy verification methods for road-level HD electronic navigation maps.Taking HD maps generated by mobile surveying as an example,we propose a comprehensive framework covering validation processes,scenario selection,truth point collection and matching,and precision analysis.Validation experiments are conducted on expressways in Hefei,Liu'an and Chuzhou (Anhui province)demonstrate that this method effectively controls map data quality,achieving absolute accuracy of 50 cm (2σ)and relative accuracy of 20 cm (3σ),which meets the requirements for autonomous driving and intelligent transportation systems.The proposed methodology provides reusable technical references for the industry.

Urban multi-dimensional sensing infrastructure: architectural innovations and monitoring applications

HU Yaofeng, CHENG Xiangbing, CHEN Jiaqi

2025, 0(8):  174-178.  doi:10.13474/j.cnki.11-2246.2025.0829

Asbtract ( )   PDF (2252KB) ( )  

References | Related Articles | Metrics

Historical and cultural blocks are core areas where urban historical and cultural heritage is centrally preserved,serving as a critical component and essential focus in the protection of historical and cultural cities.This paper addresses persistent issues in China's historical and cultural blocks,including lagging supervision and management,recurring damage to traditional architectural features,and outdated dynamic monitoring methods.We propose the concept of establishing an urban multi-dimensional sensing network,investigate its key technologies,and validate the significant effectiveness of the sensing network implementation through a pilot application in Guangzhou's historical and cultural blocks.

GAMIT multi-system GNSS data processing to determine Guangzhou CORS coordinate datum

CHEN Jiaqi, CHENG Xiangbing, HU Yaofeng, LI Wenyi

2025, 0(8):  179-184.  doi:10.13474/j.cnki.11-2246.2025.0830

Asbtract ( )   PDF (5221KB) ( )  

References | Related Articles | Metrics

To unit of Guangzhou CORS reference station network and BeiDou ground-based monitoring station network coordinate datum,and explore the feasibility of using the BeiDou system to maintain the surveying datum of regional CORS station network,we carry out GAMIT high-precision baseline solution and 3D constrained adjustment for 6 sets of dual-frequency data from 27 stations in the Greater Bay Area.The results show that normalized root mean square error(NRMSE) of the baseline solutions is better than 0.25,and the ambiguity resolution success rate of GPS,GLONASS,and Galileo baseline solutions exceeds 90%.The plane direction of baseline repeatability is better than the elevation direction,the fixed error of linear fitting is in the order of millimeter,and the scale error is better than the GNSS Class B 10^-8 accuracy requirements.The baseline adjustment accuracy of the four systems is the highest,and average RMSE of the single system adjustment coordinate difference is 1.1 mm in the north direction and 1.5 mm in the east direction.There is inconsistency in the elevation direction,and the maximum difference in the elevation of Galileo and GLONASS adjustment coordinates is more than 20 mm.This paper verifies the feasibility of GAMIT processing BDS data to independently maintain regional CORS coordinate datum.On the whole,the accuracy of BDS new signal frequency B1C/B2a and B1C/B3I is equivalent to that of GPS L1/L2 dual-frequency combination,and superior to BDS B1I/B3I and Galileo E1/E5a dual-frequency combination.The accuracy of various indicators of the GLONASS G1/G2 dual-frequency combination is relatively low.