Table of Content

25 September 2025, Volume 0 Issue 9

Previous Issue   

Analysis on the relationship between landscape pattern and windbreak and sand fixation services in Xinjiang based on the MGWR model

ZHAO Hui, LIU Qian, ZHANG Min, LI Jiayu

2025, 0(9):  1-7.  doi:10.13474/j.cnki.11-2246.2025.0901

Asbtract ( )   HTML ( )   PDF (1582KB) ( )  

References | Related Articles | Metrics

Xinjiang,a pivotal region in China's ecological security strategy and a key area in the “Three-North” shelter forest program,plays a vital role in windbreak and sand fixation due to its unique geography.This study uses the revised wind erosion equation (RWEQ)to quantify windbreak and sand fixation amounts and combines it with the multi-scale geographically weighted regression (MGWR)model to analyze the spatiotemporal relationship between landscape patterns and windbreak and sand fixation from 2000 to 2020.The results are as follows: ①From 2000 to 2020,the number of patches (NP),average shape index (SHAPE＿MN),and patch density (PD)showed an increasing trend,while the largest patch index (LPI)and cohesion (COHESION)exhibited a decreasing trend.②The average windbreak and sand fixation amount from 2000 to 2020 increased by 70.73 t/km².The windbreak and sand fixation amount remained stable at a high level in the Junggar basin but was relatively low in the desert areas of the Tarim basin.③The MGWR model showed good stability with an Adjusted R² ranging from 0.706 to 0.715.The regression coefficients of NP and COHESION were positive,while those of the other landscape indices were negative.This study provides a basis for refined landscape management and optimized windbreak and sand fixation strategies in Xinjiang,promoting regional ecological sustainability.

Spatio-temporal evolution and coupling coordination of urban resilience and ecosystem services in the middle and lower reaches of the Yellow River

ZHANG Zhaomin, JIN Fengxiang, DOU Fengke, ZHONG Weijie, LIU Wenyi, LIU Yaohui

2025, 0(9):  8-12,33.  doi:10.13474/j.cnki.11-2246.2025.0902

Asbtract ( )   HTML ( )   PDF (1604KB) ( )  

References | Related Articles | Metrics

Urban resilience and ecosystem services are of great significance for sustainable urban development,but the coordination between the two has not been fully explored.In this study,urban resilience and three ecosystem services,including carbon storage,habitat quality,and soil conservation,were assessed from 2012 to 2022 using the entropy weight method and the InVEST model in cities along the middle and lower reaches of the Yellow River.On this basis,the coupling coordination model was applied to explore the coordination relationship between urban resilience and ecosystem services.The results of the study show that from 2012 to 2022,urban resilience increases significantly,with the average index growing from 0.064 to 0.141.The resilience level in the eastern region is generally higher than that in the western region.The coupling coordination between urban resilience and habitat quality showed a decreasing trend.The results of this study are of great significance for promoting ecological protection and high-quality development of cities in the Yellow River Basin,providing scientific basis and reference for achieving ecological civilization and the construction of resilient cities.

Spatio-temporal variation characteristics and causes of eco-environment changes in typical open-pit coal mines from 1986 to 2023

QIAO Linquan, YAN Xingguang, ZHU Fangfang, MA Xiaoliang, DU Hao

2025, 0(9):  13-18,33.  doi:10.13474/j.cnki.11-2246.2025.0903

Asbtract ( )   PDF (1487KB) ( )  

References | Related Articles | Metrics

Mine ecological environment monitoring is the basis for guiding mine ecological restoration and green mine construction,in order to reveal the spatio-temporal change characteristics and causes of typical open pit mining areas,based on Landsat imagery from 1986—2023 and land classification data,and using Theil-Sen median and Mann-Kendall trend tests,we explored the spatial and temporal variation characteristics of vegetation cover,coal dust,soil erosion degree and land classification in the Antaibao and Anjialing mining areas from 1986 to 2023 using Theil-Sen median and Mann-Kendall trend tests.The results show that: The vegetation cover in the mining area shows a trend of decreasing and then increasing; the change trend of coal dust is not significant,and the coal mining face and transportation road are the areas with high incidence of coal and carbon dust; the soil erosion degree in the mining area shows a trend of increasing and then decreasing,and the soil erosion did not show a continuous expansion in the case of production capacity expansion in 2008; and the change of land class trajectory in the mining area is dominated by the transformation of farmland and grassland into the impervious layer,and the change of trajectory in the mining area is dominated by the transformation of farmland and grassland into the impervious layer.This study reveals the ecological evolution of typical surface coal mines in all aspects,and provides scientific references for the sustainable development of green mines and the response to global climate change.

Downscaling model of SMAP L4 soil moisture product in Inner Mongolia region based on machine learning

BIAN Chaoyang, HUANG Fang, HE Weibing, ZHANG Qiaofeng, LU Tongtong, GUAN Hao

2025, 0(9):  19-25.  doi:10.13474/j.cnki.11-2246.2025.0904

Asbtract ( )   PDF (1429KB) ( )  

References | Related Articles | Metrics

With the intensification of global warming,drought disasters are increasingly frequent in Inner Mongolia region,posing a serious threat to agricultural and animal husbandry production,ecological environment and regional sustainable development.Soil moisture,as the most direct indicator reflecting drought disasters,has a significant impact on agricultural and animal husbandry production and ecological environment in Inner Mongolia region.However,there are difficulties in obtaining soil moisture data with high temporal and spatial resolutions at present,and traditional monitoring methods are difficult to meet the demand.This study utilizes the powerful cloud storage and computing capabilities of Google Earth Engine to obtain long-time series remote sensing data including Sentinel-1 SAR,SMAP L4,Landsat 8,MODIS LST,DEM and other products of Inner Mongolia region,and performs preprocessing,unifying time scales and spatial resolutions.Through correlation analysis,downscaling factors with the greatest correlation with SMAP L4 soil moisture are selected.Random forest,support vector machine and classification and regression tree algorithms are respectively used,combined with the selected downscaling factors,to carry out soil moisture downscaling experiments and obtain soil moisture data with 1 km high spatial resolution and high accuracy in the study area.Finally,the downscaling results are compared and verified with SMAP L4 resampling data and public soil moisture datasets.The results show that the downscaling model based on random forest achieves an average R value of up to 0.84,an average MAE of 0.049 m³/m³,and both RMSE and ubRMSD are significantly smaller than the other two models,demonstrating the best downscaling performance in the study area.Based on innovative data processing methods,fine downscaling factor screening mechanisms,and comparative application of multiple machine learning algorithms,this paper provides an effective solution for obtaining long-term series,high-resolution,and high-precision soil moisture data in Inner Mongolia region,which is of great significance for local agriculture and animal husbandry,drought monitoring,and sustainable development.

Fire detection and ranging method based on binocular vision

ZHU Yuancai, SUN Liying, ZHANG Fan, WU Zhaoli, GAO Xiangdong, JIN Lei, LI Xiaodong, LI Ruoyu

2025, 0(9):  26-33.  doi:10.13474/j.cnki.11-2246.2025.0905

Asbtract ( )   PDF (1647KB) ( )  

References | Related Articles | Metrics

Forest and grassland fires have become a serious public safety issue in today's society,often leading to significant casualties and property losses.This paper proposes a fire target ranging method that combines object detection with stereo matching to achieve the identification and localization of fire targets.The method utilizes a three-layer SGBM algorithm in conjunction with YOLOv5 object detection,applying WLS filtering and optimizing the depth information of targets at different scales to enhance the accuracy and stability of large-scale depth images.Experimental results show the system has high accuracy.When the detected target proportion in testing distance is within 4%,the improved SGBM algorithm achieves high precision,with the relative positioning error kept within 2%.The proposed method exhibits high accuracy and reliability,making it applicable to fire monitoring,early warning,firefighting,and rescue operations in forest and grassland areas.

Spatio-temporal optimization of stock carbon sink value realization under the coupling of county-level land use planning and ecological asset management

JIN Liqiang, PENG Yan

2025, 0(9):  34-38.  doi:10.13474/j.cnki.11-2246.2025.0906

Asbtract ( )   PDF (1232KB) ( )  

References | Related Articles | Metrics

The valuation and effective management of terrestrial carbon sinks have emerged as critical components in global climate change mitigation strategies.While incremental carbon sinks have established trading mechanisms,the valuation and realization of stock carbon sink potential remains a significant challenge,particularly at regional and local scales.This gap in carbon sink management presents a crucial barrier to achieving carbon neutrality objectives and optimizing ecological asset values.This study develops an integrated methodological framework for quantifying and realizing stock carbon sink values at the county level through innovative land use optimization approaches.Employing a multi-dimensional research methodology incorporating spatio-temporal analysis,PLUS＆CA modeling,and dynamic simulation techniques,we examine the complex interactions between land use patterns and carbon sink capacity.Our analysis yields three significant contributions:①development of a comprehensive accounting model that integrates terrestrial stock carbon sink assessment with land use change dynamics; ②establishment of a longitudinal dataset documenting terrestrial ecosystem carbon sink patterns,supplemented by spatial simulation projections; ③construction of a novel value realization framework that transcends conventional carbon market limitations.Our findings demonstrate that land use modifications significantly influence the spatio-temporal distribution of terrestrial stock carbon sinks,while effective value realization mechanisms can optimize land use patterns through feedback mechanisms.The research further reveals that successful implementation requires a systematic approach integrating policy frameworks,market mechanisms,and technological innovations,supported by coordinated stakeholder engagement and adaptive management strategies.

Extracting linear water bodies from water index combined with DEM adaptive search algorithm

XU Wenting, YAN Dongmei, WANG Hu, WU Yarui, WANG Meijing, SHEN Qian

2025, 0(9):  39-44,77.  doi:10.13474/j.cnki.11-2246.2025.0907

Asbtract ( )   HTML ( )   PDF (1569KB) ( )  

References | Related Articles | Metrics

To address the common issue of discontinuities in traditional single water body extraction methods,an adaptive search algorithm combining remote sensing spectral information and DEM is used to extract linear water bodies and assess the algorithm's applicability for extracting linear water bodies from data with different spatial resolutions.The study focuses on the water bodies in Guangzhou,using 30 m Landsat OLI and 16 m GF1 WFV imagery to obtain normalized difference water index information.Then,30 m resolution ASTER GDEM and 12.5 m resolution ALOS elevation data are selected to obtain river network data.By choosing appropriate search windows and elevation thresholds as extraction parameters,river data is extracted.To address the discontinuities in some areas of the water body index extraction results,the river data is used for spatial overlay to obtain the final river information.The results show that compared with the single water body index extraction results,the linear water bodies extracted by the water body index combined with the DEM adaptive search algorithm (NDWI+12.5 m DEM and NDWI+30 m DEM)are continuous and accurate,with overall accuracies of 90.5%and 95%,respectively.Especially the 12.5 m DEM data shows a more obvious advantage in detail capture and has higher precision.

Spatio-temporal reflectance fusion for remote sensing images using a double-coupled non-parametric Bayesian approach

CHEN Nan, ZHANG Biao, YANG Nan, LIU Zhouzhou

2025, 0(9):  45-50.  doi:10.13474/j.cnki.11-2246.2025.0908

Asbtract ( )   PDF (1408KB) ( )  

References | Related Articles | Metrics

With the rapid development of remote sensing technology,acquiring remote sensing images that simultaneously possess high spatial and high temporal resolution has become a research hotspot.Traditional single optical sensors are limited by strip width and revisit period,making it difficult to meet both requirements simultaneously.The remote sensing image spatio-temporal reflectance fusion technique effectively addresses this issue by combining images with fine spatial resolution but low acquisition frequency and images with coarse spatial resolution but high acquisition frequency.This paper proposes a method based on a dual-layer spatio-temporal fusion framework,which integrates a cross-resolution attention mechanism and a nonparametric Bayesian dynamic dictionary learning mechanism,aiming to generate fused images with both high spatial and high temporal resolution.Experimental results demonstrate that the proposed method exhibits high fusion accuracy and robustness in regions with phenological changes and abrupt land cover changes,and it can better preserve spectral information and spatial details compared to existing methods.

Remote sensing estimation of aquatic vegetation biomass in Ulansuhai Lake

GONG Liangchen, XU Dongpo, KUANG Zhen, FAN Yingchun, DONG Jiahui

2025, 0(9):  51-58.  doi:10.13474/j.cnki.11-2246.2025.0909

Asbtract ( )   PDF (1505KB) ( )  

References | Related Articles | Metrics

Aquatic vegetation plays an irreplaceable role in the ecological balance and water purification of rivers and lakes.Biomass,as an important biophysicochemical indicator of aquatic vegetation,not only reflects primary productivity but also plays a key role in assessing nitrogen and phosphorus reserves and carbon sequestration in water bodies.To address the challenge of comprehensively surveying aquatic vegetation biomass in the complex lake vegetation environment of Ulansuhai Lake and to account for the spectral response differences between emergent and submerged plants,this study proposes a biomass estimation model combining remote sensing data with field survey data.Using Landsat 8 imagery and data from 32 field sampling points,and based on ENVI and Matlab software,a partial least squares regression model for the retrieval of aquatic vegetation biomass in Ulansuhai Lake is developed and its accuracy assessed.The results showed that the estimation accuracies of the biomass retrieval models for emergent plants and submerged plants reached 92.93%and 79.80%,respectively.This indicates that the partial least squares regression model is suitable for retrieving aquatic vegetation biomass under small sample conditions and can achieve high accuracy in complex lake vegetation environments.The total biomass of aquatic vegetation in Ulansuhai Lake in June and September 2023 was 781.6×10⁴ and 791.9×10⁴ t,respectively,with emergent plants accounting for 78.03%and 76.82%.Compared to June 2023,the area of emergent plants in Ulansuhai Lake decreased in September,the average biomass per unit area increased,and there was no significant change in spatial distribution; The biomass of submerged plants show a notable increase,with the area expanding,approaching the historical high from 1988(102.06 km²).This research provides scientific evidence for subsequent lake ecological management,health assessment,and water quality restoration in Ulansuhai Lake.

Road orthophoto generation based on vehicle-mounted mobile measurement system

ZHANG Panke, MA Hao, ZHANG Weihong, LIU Xianlin

2025, 0(9):  59-63.  doi:10.13474/j.cnki.11-2246.2025.0910

Asbtract ( )   PDF (1303KB) ( )  

References | Related Articles | Metrics

The vehicle-mounted mobile measurement system has the characteristics of fast acquisition of road data.Applying it to the generation of high-precision orthophoto images on the road surface can greatly reduce the measurement blind spots caused by the occlusion of vegetation,dense buildings and 3D structures of ground objects.In this paper,a method of generating orthophoto image of road surface based on vehicle mobile measurement system is proposed.The fisheye camera is innovatively used to solve the problem of limited observation field angle caused by low viewpoint of vehicle.The high-resolution image and high-precision point cloud data of the vehicle are fused to generate high-resolution and high-precision orthophotos.A set of road orthophoto generation schemes for the vehicle-mounted mobile measurement system is summarized,and the reliability and accuracy of the method are verified by experiments.

Robot dynamic scene visual SLAM algorithm integrating geometric features and semantic information

HE Tingting, JIANG Xianglong, HE Shengxi

2025, 0(9):  64-69.  doi:10.13474/j.cnki.11-2246.2025.0911

Asbtract ( )   PDF (1285KB) ( )  

References | Related Articles | Metrics

To address the challenges of achieving high-accuracy visual SLAM in dynamic scenes,including low robustness,large localization errors,and feature loss,this paper proposes a visual SLAM algorithm that integrates geometric features and semantic information (Geo-Semantic SLAM).Built upon the ORB-SLAM2 framework,the proposed method deeply integrates semantic segmentation networks with geometric feature extraction techniques,significantly enhancing its adaptability to dynamic environments.Specifically,Geo-Semantic SLAM employs semantic segmentation to eliminate the influence of dynamic objects and introduces a camera pose optimization method based on the semantic information of static objects,effectively compensating for the feature loss caused by dynamic object removal.Experimental validation on the TUM dataset demonstrates that Geo-Semantic SLAM achieves superior localization and mapping accuracy in dynamic environments.Moreover,it consistently outperforms traditional algorithms and semantic SLAM methods across various scenarios.

ADM-YOLOv11:dynamic adaptive multi-scale object detection algorithm for tower video

TANG Zhiqing, ZHANG Tao, WANG Peiyu, XIANG Dao, LIU Haifei, LIU Renfeng, HE Jiangjiang

2025, 0(9):  70-77.  doi:10.13474/j.cnki.11-2246.2025.0912

Asbtract ( )   PDF (1537KB) ( )  

References | Related Articles | Metrics

Aiming at the technical bottlenecks such as the insufficient detection accuracy of multi-scale targets and the low recognition rate of small targets in the large-view dynamic scenes of tower videos,this paper proposes a ADM-YOLOv11 target detection algorithm based on the collaborative optimization of multiple modules.Firstly,the adaptive feature enhancement(AFE)module is embedded in the Backbone network to deeply reconstruct the C3K2 module.Through the spatial context awareness and feature refinement mechanisms,the feature extraction ability of the network for complex scenes is significantly improved.Secondly,the efficient multi-scale attention (EMA)module is integrated into the C3K2 module of the Neck to enhance the detection robustness of the model for multi-scale targets.Thirdly,the ultra-lightweight dynamic upsampler DySample (dynamic upsample)is introduced into the Neck structure to replace the traditional upsampling layer,optimizing the detail expression and semantic fusion efficiency of multi-scale features.Finally,the EMASlideLoss classification loss function is adopted.By using a dynamic weighting strategy,the problem of gradient shift caused by data imbalance is suppressed,effectively improving the generalization performance of the model.The experimental results show that for the model in this study,the mAP_50-95 is increased from 74.8%of the baseline model to 82.6%,and the mAP₅₀ is increased to 96.6%.The ADM-YOLOv11 significantly improves the detection accuracy of multi-scale targets in the dynamic scenes of tower videos.

A lightweight enhanced LiDAR-inertial-visual odometry system

YANG Yanguang, QIAN Jianguo, YU Bin, GUO Jie, JIAO Yang

2025, 0(9):  78-83,104.  doi:10.13474/j.cnki.11-2246.2025.0913

Asbtract ( )   HTML ( )   PDF (1428KB) ( )  

References | Related Articles | Metrics

Lightweight intelligent tracking for enhanced LiDAR-IMU-visual odometry(LIVO)has broad applications in mobile robotics and autonomous driving.This paper proposes LITE-LIVO,a lightweight and enhanced LIVO system built upon FAST-LIVO,integrating LiDAR,IMU and vision sensors for real-time pose estimation and high-precision mapping.To enhance robustness under dynamic lighting,the system employs a deep learning-based feature extraction and sparse optical flow tracking,fusing visual and LiDAR data via Kalman filtering with visual residuals.A tightly coupled visual-IMU odometry (VIO)subsystem filters high-quality visual features from LiDAR point clouds and optimizes visual map management.Experimental results on public datasets and real-world scenarios demonstrate superior performance,especially in complex and degraded environments.This study advances multi-source data fusion techniques,improving localization accuracy and expanding application domains for mobile robots.

Coastline extraction method based on neural network optimized canny operator

WANG Yu, LI Zechen, LIANG Songyuan, SHI Xue

2025, 0(9):  84-90.  doi:10.13474/j.cnki.11-2246.2025.0914

Asbtract ( )   HTML ( )   PDF (1531KB) ( )  

References | Related Articles | Metrics

In order to accurately and efficiently extract coastlines from synthetic aperture radar images,this paper proposes a coastline extraction algorithm based on neural network-optimized Canny operator.The algorithm combines neural networks and statistical regression to adaptively determine the optimal values of the three Canny parameters that are Gaussian filter standard deviation,high threshold,and low threshold,in order to improve the Canny operatorly.Firstly,a neural network model is trained on the training set to obtain the optimal CaPP values for each SAR image.Then,statistical regression and optimization criteria are used to establish the optimal linear combination of CaPP and the SAR image's mean and standard deviation.Finally,the algorithm is experimentally verified using a test set.The experimental results show that the proposed algorithm can adaptively obtain the optimal CaPP values,with the SSIM mean of the coastline extraction results in the test set being 0.912,and the overall accuracy and Kappa coefficient means are 98.55%and 0.966 3,respectively.This demonstrates that the proposed algorithm can accurately extract the coastlines of SAR images by adaptively obtaining the optimal CaPP values.

Application experiment and analysis of long-baseline coordinate transfer for BeiDou high-precision position datum

ZHOU Mingduan, BAI Yansong, XU Xiang, CUI Likun, QIN Yuhan, XIE Qianlong, LIN Shiqi

2025, 0(9):  91-97,104.  doi:10.13474/j.cnki.11-2246.2025.0915

Asbtract ( )   PDF (1524KB) ( )  

References | Related Articles | Metrics

Aiming at the current stage of BeiDou system in BDS-3+BDS-2 hybrid service and BDS-3 broadcasts B1C and B2a new frequency signals, a position datum coordinate transfer model based on BeiDou systemis proposed and a BDS position datum long-baseline coordinate transfer network based on B1I+B3I and B1C+B2a data is established for the application testing. Four IGS tracking stations in China and its surrounding areas that simultaneously support BDS-3+BDS-2 and GPS data as well as a fixed station are selected to lay a BDS position datum long-baseline coordinate transfer network.7 sessions of data based on 24 hour continuous observation from 096～102 in 2024 are selected and 4 experimental schemes are designed for data processing, and comparative analysis and evaluation are carried out in terms of the accuracy of phase observation values, post standardized root-mean-square error, quality of baseline vectors,and point results of datum transfer network adjustment.The experimental results show that different from L1+L2 dual-frequency data of GPS, the differences of datum transfer in point direction are both of only millimeter level using B1I+B3I dual-frequency data of BDS-3 and BDS-3+BDS-2, and millimeter to centimeter level using B1C+B2a dual-frequency data of BDS-3, which is effective to be applied to the long-baseline coordinates transfer for BeiDou high-precise position datum.

Elevation compensation method and accuracy assessment for COSMIC-2 tropospheric delay

WU Angdao, TANG Xu

2025, 0(9):  98-104.  doi:10.13474/j.cnki.11-2246.2025.0916

Asbtract ( )   PDF (1443KB) ( )  

References | Related Articles | Metrics

In response to the underestimation of COSMIC-2-retrieved zenith tropospheric delay (ZTD)caused by missing near-surface data,this study firstly employs elevation-based scaling factors derived from the ERA5 dataset,then applies a negative exponential model to correct the COSMIC-2 ZTD,and finally validates the correction accuracy using radiosonde measurements and GNSS ZTD data from the crustal movement observation network of China (CMONOC)across China during 2020—2022.The results indicate that,after applying the correction,the mean RMSE of the COSMIC-2 ZTD decreases by 30.95 and 32.48 cm when compared to the radiosonde and GNSS ZTD references,respectively.Moreover,a month-by-month assessment for 2022 shows that the corrected COSMIC-2 ZTD RMSE stabilizes within a range of 2～8 cm.These findings demonstrate the suitability of the negative exponential elevation model for enhancing COSMIC-2 ZTD data over China,providing reliable tropospheric delay corrections that benefit both GNSS navigation and atmospheric water-vapor monitoring.

A visual SLAM algorithm based on illumination-robust feature extraction and dynamic feature removal

KE Xueliang, XIAO Wei, QU Naizhu, HE Zhijie, HUANG Rui

2025, 0(9):  105-111.  doi:10.13474/j.cnki.11-2246.2025.0917

Asbtract ( )   PDF (1550KB) ( )  

References | Related Articles | Metrics

To address the issue of low localization accuracy in visual SLAM algorithms caused by environmental moving objects and illumination changes,this paper proposes a high-precision visual SLAM algorithm suitable for dynamic and varying illumination environments.This algorithm is based on the VINS Mono architecture,firstly,it performs illumination-robust feature extraction by extracting features from input images through a ResNet network to obtain initial feature point coordinates and illumination-invariant feature maps; then,it performs optical flow estimation on these maps to reduce the impact of illumination on feature tracking.Subsequently,it carries out dynamic feature removal by using pixel-level semantic segmentation with YOLOv8 to mark dynamic objects in the input images as masks.Epipolar geometry constraints are then utilized to remove dynamic features within the mask regions,obtaining stable static feature points for tracking and reducing the impact of dynamic features on the algorithm's localization accuracy.Finally,comparative experiments on the EuRoC,VIODE,and Market datasets show that our method achieves 55.09%lower absolute trajectory error compared to VINS-Mono,demonstrating good localization accuracy in dynamic and varying illumination environments.

An improved semantic segmentation method for bridge cable damage using large-scale segmentation models

DENG Xiaolong, HUANG Zhihai, GUO Bo

2025, 0(9):  112-117.  doi:10.13474/j.cnki.11-2246.2025.0918

Asbtract ( )   PDF (1487KB) ( )  

References | Related Articles | Metrics

Bridge cable damage detection is a critical aspect of bridge safety operation and maintenance. How to quickly and automatically process cable images and accurately detect damaged areas is key to ensuring bridge safety operations.This paper improved the large-scale segmentation model SAM (segment anything model)and applied it to the semantic segmentation of bridge cable damage,providing significant evidence for damage detection.The improvements to SAM consisted of two main points:①Fine-tuning an Adapter on the image encoder and enhancing the model's generalization to bridge cable data through transfer learning methods.②Fine-tuning the prediction head of the mask decoder to enable SAM to perform semantic segmentation without prompts and introduce multi-class capabilities.To verify the advantages of the improved SAM,comparative experiments had been conducted using a dataset of nearly 2200 bridge cable images against DeepLabV3+.The results showed that the improved SAM performed better in scenarios with imbalanced damage category distributions and limited sample sizes.Its semantic segmentation evaluation metrics included a mean intersection over union (mIoU)of 73.41%,an average F1 score of 83.99%,and mean class accuracy of 81.80%.

Generalized application of map generalization methods in the era of pan-map

WEI Zhiwei, YANG Nai, CHEN Yebin, GUO Renzhong

2025, 0(9):  118-125,145.  doi:10.13474/j.cnki.11-2246.2025.0919

Asbtract ( )   PDF (1570KB) ( )  

References | Related Articles | Metrics

Map generalization is a fundamental technology in cartography.As the concept of maps evolves,map generalization also needs to expand its conceptual framework and methodologies to meet the requirements of pan-map presentation and production.This paper discusses the expansion of map generalization in terms of generalized objects,content abstraction,and presentation dimensions within the conceptual framework of the pan-map.It also explores the application of traditional map generalization methods to the creation of various pan-maps,such as flow maps,cartograms,and thematic maps.These applications broaden the scope of existing cartographic generalization methods and provide valuable references for further exploration in the field.

3D reconstruction of offshore islands based on sparse controlled oblique photography

ZENG Yicheng, ZHANG Yihe, DING Jianxun, LI Wangmin, LI Xiulong

2025, 0(9):  126-130.  doi:10.13474/j.cnki.11-2246.2025.0920

Asbtract ( )   PDF (1370KB) ( )  

References | Related Articles | Metrics

In order to solve these problems that the complex terrain and insufficient coordinate reference service of offshore islands,the layout and measurement of image control points in the oblique photogrammetry for offshore islands with high cost,high labor intensity,high risk coefficient and low work efficiency.This paper studies a method for achieving oblique photogrammetry of offshore islands through sparse control.Firstly,lay out a small number of control points according to the area and topographic features of each island and obtain high-precision flight data of position and orientation through the joint solution based on the CORS,GNSS and IMU data of unmanned aerial vehicles.Secondly,the 3D real model is constructed through technologies such as aerial triangulation and image matching.Finally,this paper provides a detailed introduction to the application of this method in island surveying taking Miaowan Island,Niutou Island and Wenweizhou Island in Zhuhai city as experimental cases.The obtained results,after being subjected to precision detection,have been confirmed to meet the accuracy requirements for large scale topographical mapping.This method is practical and feasible,providing an effective approach and reference for the surveying work on offshore islands.

Farmland information extraction and surface change monitoring analysis based on deep learning

JIANG Feng, CHEN Chao

2025, 0(9):  131-134.  doi:10.13474/j.cnki.11-2246.2025.0921

Asbtract ( )   PDF (1527KB) ( )  

References | Related Articles | Metrics

Farmland is an important guarantee for national security,and its spatial distribution is the main basis for farmland protection,national spatial planning,and other management work,while surface change monitoring is a key support for mastering land use and natural resources.With the development of information technology,deep learning based land cover classification and surface change monitoring have been widely studied.This article explores the intelligent information extraction technology of remote sensing based on deep learning.Using two phases of Jilin-1 satellite remote sensing images covering the experimental area as the data source,a large-scale remote sensing model under the deep learning framework is used to extract agricultural patches from a single phase of remote sensing images and monitor surface changes from two phases of images.The accuracy of the extraction results is evaluated by combining remote sensing images with land survey data.The results indicate that the research method can accurately identify farming patches and changing areas,and the boundary shape matches the image,with good potential for application.

Marker region extraction driven by YOLOv7 deep learning algorithm under complex illumination

ZHANG Chen, CHU Yunzhi, WU Zhaofu, XU Lichen, HUANG Jianwei, LI Shuiping

2025, 0(9):  135-139.  doi:10.13474/j.cnki.11-2246.2025.0922

Asbtract ( )   PDF (1395KB) ( )  

References | Related Articles | Metrics

In the fields of computer vision and close-range photogrammetry,marker points are widely used,and their positioning and accurate extraction directly affect the observation accuracy.However,in long-term monitoring,complex lighting conditions can lead to poor recognition and extraction effects of marker points,thereby affecting the monitoring accuracy.For this purpose,this paper proposes a marker point extraction method based on deep learning algorithms.Firstly,a marker point dataset is established by using marker point images in different lighting environments.Then,the marker points are identified under different lighting conditions and accuracy analysis is conducted.Finally,displacement experiments were conducted on the marker point areas extracted by the YOLOv7 algorithm to determine the observation accuracy.The results show that the YOLOv7 deep learning algorithm can quickly and accurately identify the region of interest of the marker points,with a mAP of 95.45%,an F1 value of 94.36%,and a frame rate of only 4.40.Under different lighting conditions,the landmark area can be accurately identified and the observation accuracy is high.The research results can provide effective solutions for the automatic and high-precision extraction of marker point areas in complex environments in long-term dynamic monitoring.

Installation error calibration method for shipborne 3D LiDAR in unstructured scenes

CHEN Menhao, LI Zhenbo, ZHU Yabing, HUANG Yi, BU Xianhai, YANG Fanlin

2025, 0(9):  140-145.  doi:10.13474/j.cnki.11-2246.2025.0923

Asbtract ( )   PDF (1389KB) ( )  

References | Related Articles | Metrics

The installation error of shipborne 3D laser scanning system is composed of installation angle error and offset error,among which the impact caused by installation angle error is the most serious.The traditional positioning error calibration method is greatly limited by the geometric structure of the calibration site.Therefore,this paper proposes a shipborne 3D laser installation error calibration method with multiple feature constraints in unstructured scenes.Firstly,the ICP algorithm is used to pre register the point clouds in the overlapping area of the round-trip survey line.The Euclidean distance threshold is set to retrieve the nearest point between the two point clouds after registration as the undetermined feature point.Then,the final feature point pair is obtained by using the normal vector and curvature as constraints.Finally,retrieve the original observation values corresponding to the feature points and substitute them into the calibration model to calculate the placement angle.The measured data show that the proposed method can effectively achieve precise calibration of the placement angle error of shipborne 3D laser scanners in unstructured scenes.After calibration,the internal consistency accuracy is 0.086 m,and the standard deviation of point offset is 0.014 m.The calibration results are relatively stable and the internal consistency accuracy is high.

Application of ground penetrating radar interference noise location method in transmission line investigation

YIN Xiaomin, RAN Yiding, LOU Fengqiang, LI Xiaodong, TANG Xiaoxian, GUO Mingrui, GE Junkai, WANG Xiaolong, SUN Huaifeng

2025, 0(9):  146-151,156.  doi:10.13474/j.cnki.11-2246.2025.0924

Asbtract ( )   PDF (1730KB) ( )  

References | Related Articles | Metrics

Ground penetrating radar (GPR)uses the propagation and reflection of electromagnetic waves to acquire subsurface information.RTK,tape measure,odometer and other methods are commonly used to locate underground anomalies in the measurement of transmission tower lines by ground penetrating radar.However,RTK is complex and expensive,while other methods may be inaccurate or not usable on rough terrain.This paper presents a method to accurately locate underground anomalies of transmission lines without additional positioning.The interference noise of transmission towers in GPR images can be used to locate underground defects.After the localization is completed,the curvilinear wave transform and Stolt velocity migration method are combined to remove the interference noise,and the noise-free B-scan image is obtained.

Application of airborne thermal infrared remote sensing in geothermal exploration of Zigui county

WU Jing, HE Hongwei, XIAO Qian, WANG Ziwei, LIU Qi

2025, 0(9):  152-156.  doi:10.13474/j.cnki.11-2246.2025.0925

Asbtract ( )   PDF (1400KB) ( )  

References | Related Articles | Metrics

Traditional geothermal exploration methods feature long-term cycles and high costs.In Zigui county,located in the mountainous area of western Hubei province,complex topography further inflates exploration expenses.This study applies airborne thermal infrared remote sensing (TIRS)technology to overcome these challenges.A 1.05 km² rectangular area centered on Wulong hot spring in Zigui county is selected.A DJI Mavic 3 unmanned aerial vehicle(UAV)equipped with a thermal infrared camera conducted winter daytime and nighttime aerial surveys,identifying five geothermal anomaly zones.After interpretation and screening,two geothermal target areas are determined.Field investigations show that one area matched the Wulong hot spring outcrop,while the other is a newly discovered geothermal site.The results prove that TIRS technology is reliable and applicable in geothermal exploration,effectively shortening the exploration cycle and reducing costs.

High-precision intelligent monitoring of deformation and inclination of outer tank of LNG storage tank

LI Zhen, LU Te, ZHANG Yunwei, LEI Jiangkai, WU Feng'an

2025, 0(9):  157-162.  doi:10.13474/j.cnki.11-2246.2025.0926

Asbtract ( )   PDF (2136KB) ( )  

References | Related Articles | Metrics

This study enables intelligent LNG(liquefied natural gas)tank monitoring by combining ground LiDAR,UAV radar,static level gauges,and vibrating wire sensors to analyze outer tank deformation in Guangxi's Beihai LNG project.Findings reveal, 0～5 m tank zone exhibits convex deformation (max 28.7 mm)due to gravity-load-induced concrete creep; longitudinal deformation uniformity exceeds transverse,attributed to more uniform longitudinal prestress; surface flatness deviation (21.3 mm)exceeds standard,suggesting uneven settlement; relative settlement range (-1.26 to 0.82 mm)meets requirements; concrete circumferential stress inversely correlates with temperature as foundation constraints generate additional stress during thermal variation.

BIM model individualization decomposition and adaptive tile generation technology

KE Shuisong, DONG Xuehui, XIAO Haibo, LI Fan, PENG Yingyu

2025, 0(9):  163-167.  doi:10.13474/j.cnki.11-2246.2025.0927

Asbtract ( )   PDF (1240KB) ( )  

References | Related Articles | Metrics

With the rapid development of digital twin city technologies,building information modeling (BIM)has become increasingly critical in urban planning,design,construction,and operational management.However,the widespread application of BIM model in digital twin cities often faces challenges posed by massive data volumes,which directly impact real-time visualization of models as well as the real-time responsiveness,interactivity,and scalability of digital twin cities.This paper proposes a method integrating BIM model disaggregation techniques with an adaptive tile generation strategy.This approach adheres to the 3D Tiles 2.0 standard and incorporates a composite compression algorithm combining MeshQuan vectorized compression and edge collapse methods,aiming to optimize the storage,transmission,and rendering performance of BIM model in web environments.Additionally,experiments are designed to evaluate loading efficiency,simplification effectiveness,and rendering performance,with comparative analysis against original BIM model.Experimental results demonstrate that the optimized BIM model achieve significant improvements in loading efficiency,substantial reductions in file size and triangle face counts,while the algorithm's execution time is validated.These findings confirm the high efficiency of this technology in enhancing BIM model loading and rendering performance on web platforms.

Binocular vision SLAM algorithm with LightGlue

WANG Zhigang, FENG Kai, LIU Yichen, LIU Hui, ZHANG Wenjun

2025, 0(9):  168-172.  doi:10.13474/j.cnki.11-2246.2025.0928

Asbtract ( )   PDF (1310KB) ( )  

References | Related Articles | Metrics

Aiming to address the issue of poor localization accuracy of ORB-SLAM2 in texture-poor and low-light environments,this paper proposes a binocular visual SLAM algorithm integrated with LightGlue.By leveraging self-attention and cross-attention mechanisms,LightGlue significantly enhances the accuracy and speed of feature matching,demonstrating superior performance in challenging conditions such as texture-poor and low-light environments.Additionally,an improved RANSAC sampling strategy is introduced,which employs weighted sampling to mitigate the impact of mismatched point pairs,thereby further enhancing the robustness and efficiency of the algorithm.Experiments conducted on the EuRoC dataset and real-world scenarios show that the proposed algorithm achieves substantial improvements over ORB-SLAM2,with a maximum error reduction of 48.2%and an average error reduction of 26.2%.The localization error is reduced to the centimeter level,indicating higher accuracy in complex scenes.

Integration and optimization of shipborne integrated measurement system for hydrological parameter

ZHANG Xiaohao, WANG Xingwen, SUN Zhenyong, MA Yaochang, YE Fei

2025, 0(9):  173-176,184.  doi:10.13474/j.cnki.11-2246.2025.0929

Asbtract ( )   PDF (1391KB) ( )  

References | Related Articles | Metrics

To address the limitations of traditional observation methods in terms of observation efficiency and data accuracy for hydrological sediment monitoring in mountainous reservoir environments,this study proposes an integrated land-water measurement technology based on manned vessel platforms.By integrating laser scanners,inertial navigation systems (INS),GNSS,single/multi-beam echosounders,and water level gauges on dedicated hydrological survey vessels,we establish a comprehensive reservoir-borne measurement platform.This system enables simultaneous monitoring of multiple hydrological elements including littoral zone topography,underwater terrain,water level,and surface flow velocity/direction.Application in the Baihetan reservoir area and accuracy analysis demonstrate that the shipborne integrated measurement system improves field observation efficiency by 4.5 times,with relative errors of topographic measurement accuracy better than 2%and water level accuracy errors below 1%.The proposed vessel-based integrated hydrological measurement technology proves feasible for rapidly acquiring comprehensive hydrological data including complex land-water terrain in mountainous reservoir regions.

The virtual installation of UAV oblique photography and 3D laser scanning technology in mountainous photovoltaic installations application exploration in design

WANG Yapeng, LI Na

2025, 0(9):  177-180.  doi:10.13474/j.cnki.11-2246.2025.0930

Asbtract ( )   PDF (1232KB) ( )  

References | Related Articles | Metrics

At present,the mountain has become one of the important site selection areas for the construction of photovoltaic power stations,but due to the limitations of the terrain environment,the transportation and installation of related devices are difficult,and the cost of installation trial and error of photovoltaic power stations is high.In this paper,the 3D terrain model of the mountain is obtained based on the oblique photography technology.Firstly,it provides an accurate terrain foundation for the layout design of the photovoltaic array.Secondly,the 3D laser scanning reverse modeling technology is applied to establish an important BIM model of photovoltaic equipment to realize the integration of design information.Finally,Cesium is applied for information fusion and light and shadow simulation analysis.It is designed to optimize the layout,orientation and inclination of photovoltaic arrays,improve power generation efficiency and reduce construction costs.The results show that the proposed method can significantly improve the design efficiency and quality,and provide strong technical support for the sustainable development of mountain photovoltaic power stations.

The cultivation mode of innovative and entrepreneurial skills in surveying and mapping laboratories in universities from the perspective of “craftsmanship spirit”

MA Chunyan, GUO Min, QI Xiudong, WU Xifang, REN Xiaofang, WANG Yu

2025, 0(9):  181-184.  doi:10.13474/j.cnki.11-2246.2025.0931

Asbtract ( )   PDF (1173KB) ( )  

References | Related Articles | Metrics

Under the background of the “Emerging Engineering Education” initiative,the surveying and mapping discipline in higher education institutions needs to be guided by the “craftsmanship spirit” to explore a training model that deeply integrates laboratory-based innovation and entrepreneurship capabilities.This paper analyzes current issues in surveying laboratory teaching,such as insufficient practical skills and weak innovative awareness,and proposes a “trinity” training system for laboratory innovation and entrepreneurship skills centered on the “craftsmanship spirit.” By optimizing curriculum design,strengthening industry-university collaboration,and deepening project-driven approaches,the system aims to enhance students' professional skills,innovative capabilities,and vocational literacy,cultivating interdisciplinary professionals with meticulous excellence for the surveying engineering field.