Table of Content

25 November 2023, Volume 0 Issue 11

Previous Issue    Next Issue

Evaluation of regional landslide susceptibility based on convolutional neural network: a case study of Wanzhou district of Three Gorges Reservoir area

YANG Yanchen, ZHOU Chao, SHI Jiamei

2023, 0(11):  1-6.  doi:10.13474/j.cnki.11-2246.2023.0318

Asbtract ( )  

References | Related Articles | Metrics

Carrying out regional landslide susceptibility assessment is the key to landslide meteorological early warning and risk assessment. Aiming at the fact that many current susceptibility studies do not consider the relationship between the occurrence of landslides and adjacent environments, a regional landslide susceptibility modeling framework based on convolutional neural network (CNN) is proposed. Taking Wanzhou district of the Three Gorges Reservoir area as an example, 12 factors such as slope and aspect are selected to construct an evaluation index system, and the influence of factors on landslide development is analyzed by information method. The local two-dimensional matrix is used to construct the dataset, CNN is used for susceptibility modeling. At the same time, the impact of the size of the local two-dimensional matrix to the accuracy when constructing samples is explored. The results show that landslides are more likely to occur the closer to the reservoir zone, and the water system and human engineering activities have a greater impact on the development of landslides. The accuracy of the CNN model is 0.925, which is significantly higher than that of the machine learning model, and the accuracy can be improved by increasing the local two-dimensional matrix size when constructing the sample. The CNN model has advantages in multidimensional spatial data processing, considering the influence of landslide location and its adjacent environment, and it is an accurate and reliable regional landslide susceptibility evaluation method.

Key technologies and applications of 3D terrain modeling for landslide simulations

LÜ Yijie, YE Jian

2023, 0(11):  7-11,17.  doi:10.13474/j.cnki.11-2246.2023.0319

Asbtract ( )  

References | Related Articles | Metrics

High-resolution landslide terrain data is an important guarantee for the accuracy and visualization effect of landslide simulations. However, rendering all high-resolution landslide terrain data will cause the landslide simulation program to run slowly, managing terrain data hierarchically and loading terrain data blocks into memory in batches for combined rendering will lead to discontinuity and inconsistent resolution of terrain data participating in landslide simulation calculations. To solve the above problems, an improved quadtree LoD terrain modeling method for landslide simulations is proposed in this paper. Based on this method, terrain data is completely read into memory for numerical calculations, and terrain data blocks of different resolutions are dynamically constructed from the terrain data for combined rendering during the visualization stage, which not only provides complete and continuous terrain conditions for landslide simulations, but also ensures 3D visualization effect of landslide terrain, and effectively improves the running speed of simulation programs. More importantly, by using this method for modeling, the results of landslide simulation will not be affected when browsing the simulation scenes during the landslide simulation processes. The experimental results of landslide simulation show that the method proposed in this paper is effective and practical in 3D terrain optimization modeling for landslide simulations.

Monitoring and analysis of landslide deformation based on SBAS-InSAR

YANG Chunyu, WEN Yi, PAN Xing, YUAN Debao

2023, 0(11):  12-17.  doi:10.13474/j.cnki.11-2246.2023.0320

Asbtract ( )  

References | Related Articles | Metrics

Taking the mountain landslide occurred in Zaoling township in Shanxi province on March 15, 2019 as the research object, using Sentinel-1A SAR image data from July 5, 2018 to June 30, 2019 (a total of 30 scenes) before and after the landslide, this papere monitors the deformation data of landslides with the support of SBAS-InSAR technology. The results show that the deformation range of the study area is -52.03 to 33.77 mm/a, and the overall environment is relatively stable. The deformation rate and cumulative deformation variables of the long-term loess plateau are analyzed. The causes of landslides are analyzed based on relevant geological data from the research area. Using the standard deviation ellipse algorithm, it analyzes the spatio-temporal evolution characteristics of surface deformation in the loess plateau area where the landslide is located. The results show that the center of gravity of the standard deviation ellipse shifts to the northwest, the elliptical area decreases slightly, the deformation intensifies in the northwest southeast direction, and the deformation development in the northeast southwest direction is relatively gentle. The azimuth angle rotates counterclockwise, and the displacement is about 17.03 °.

Lunar impact crater detection based on multilevel segmentation

LI Haipeng, DONG Youfu, ZHANG Hao

2023, 0(11):  18-22.  doi:10.13474/j.cnki.11-2246.2023.0321

Asbtract ( )   HTML ( )   PDF (2171KB) ( )  

References | Related Articles | Metrics

The extraction of craters of different sizes on the lunar surface is of great value. Currently, crater detection algorithmsare effective for craters less than one kilometer, but the detection rate of largercraters needs to be improved. We propose an automatic crater detection model with good robustness.Firstly, we generate the terrain parameters based on the who lelunar DEM published by LOLA, then detect craters by object-oriented multilevel sesgmentation combined with machine learning.Three typical regions are selected for experiment and analysis, the recall and accuracy rates for craters between 1~120 km in diameter are 86.5% and 81.2% respectively, with a good detection rate.

The construction of county ecological network based on MCR and gravity model

GONG Haojie, CHANG Jinsheng, DU Yuyang

2023, 0(11):  23-29,81.  doi:10.13474/j.cnki.11-2246.2023.0322

Asbtract ( )  

References | Related Articles | Metrics

Building an ecological spatial regulation network is of great significance for maintaining the basic foundation of ecological space. This article takes Jingtai county in Gansu province as an example, uses the ecosystem service value assessment model and ecological environment sensitivity assessment model to identify ecological source areas. By using the minimum cumulative resistance model (MCR) and gravity model, resistance surfaces are established, ecological corridors and nodes are extracted, and an ecological spatial regulation network is constructed. The results indicate that the 21 ecological source areas in the study area exhibit a spatial distribution pattern of uniform distribution around the area and scattered sparsity in the middle, with a total of 41 ecological corridors and 43 ecological nodes identified. Based on the different geographical characteristics and land types of key areas, two types of restoration strategies are proposed: natural protection as the main focus, artificial restoration as a supplement, and equal emphasis on both artificial restoration and natural protection. The research results are used to identify ecological restoration areas and construct regional ecological spatial regulation networks.

Mountain shadow removal for high-resolution aerial optical remote sensing images

GUO Wei, LU Wanyun, DENG Tingqi, HUANG Guodong, YOU Wenyu, YANG Rongxin, YU Han, DONG Shouyin, JIANG Rui

2023, 0(11):  30-35,47.  doi:10.13474/j.cnki.11-2246.2023.0323

Asbtract ( )  

References | Related Articles | Metrics

Shadow is a serious interference factor for high-resolution aerial imagery in mountainous areas. Removing mountain shadow can help improve the accuracy and effectiveness of applications, such as real-life 3D construction, forestry investigation, change detection and more. In this paper, the mountain shadow index (MSI) of high-resolution aerial remote sensing image is constructed, and a shadow removal method based on color migration and color equalization is proposed. The experiment is conducted with 0.2 m aerial imagery covering mountainous areas, the results show that MSI and threshold segmentation method can effectively detect the mountain shadow of aerial imagery, and the shadow removal method combining objectified color migration and neighborhood based nonlinear color balance can effectively eliminate mountain shadow, and can also restore texture detail in the shadow area, and make the color more suitable for non-shadow areas, so as to achieve a more balanced effect of the overall color of the image. Statistically analyzing the statistical indicators before and after shadow removal, with the removal of the shadow, the average value and standard deviation of each band increased significantly, indicating that the brightness of the shadow area is improved and the color level is richer.

Accuracy evaluation of consumer UAV tilt photogrammetry for spoil area monitoring

HU Jinru, LAI Linfeng, LU Zhiyuan, ZHANG Xiaofeng, LI Yuan, ZHAO Tingning, WEI Guangkuo

2023, 0(11):  36-41.  doi:10.13474/j.cnki.11-2246.2023.0324

Asbtract ( )   HTML ( )   PDF (3713KB) ( )  

References | Related Articles | Metrics

UAV photogrammetry technology plays an important role in the monitoring of spoil ground.Consumer-grade drones have significant advantages of low threshold and low cost,but there are few studies on the accuracy analysis of photogrammetry using consumer-grade drones.In particular,the latest “specifications for office operation of low-altitude digital aerial photogrammetry” issued in 2021 has higher requirements for the accuracy of measurement.This paper takes the spoil ground of highway construction project as the research object,evaluates the accuracy of tilt photogrammetry using consumer-grade drones according to the image control points laid on the ground,and compares it with the traditional orthophoto grammetry.The test results show that the consumer-grade UAV can improve the measurement accuracy by 25.19%～ 90.68% through oblique photography,which is more significant in elevation.At the same time,the influence of large terrain drop on the measurement accuracy is reduced to centimeter level,which meets the mapping requirements of 1∶500 scale in the specification.The reliability and feasibility of consumer-grade UAV in the monitoring of spoil ground are proved.

Lithology classification of large slope geological outcrop based on UAV multi spectral remote sensing

CHANG Le, HAN Lei, CHEN Zongqiang, SHENG Hui, LUI Shanwei

2023, 0(11):  42-47.  doi:10.13474/j.cnki.11-2246.2023.0325

Asbtract ( )   HTML ( )   PDF (2775KB) ( )  

References | Related Articles | Metrics

Aiming at the problems that satellite images are difficult to obtain geological outcrop data with large slopes, and traditional classification methods cannot effectively use image information leading to geological outcrop section lithology classification accuracy being relatively low, this research obtains high-precision field geological outcrop data with large slope based on UAV remote sensing technology and proposes a multi-scale hybrid feature network model. The results show that the combination of UAV and close photogrammetry is feasible in collecting geological outcrop data. The multi-scale hybrid feature network model can effectively extract the spectral features and spatial features from the multi-spectral images of UAV and realize the high-precision lithology classification of geological outcrops with large slopes. Taking an outcrop in Yuntaishan geopark as an example, the overall classification accuracy of the proposed model can reach 89.91%, and the Kappa coefficient can reach 0.85. The general classification accuracy is nearly 15% higher than traditional machine learning algorithms SVM and MLC, about 10% higher than Inception V3 and ResNet18, and 1.5% higher than Hybrid CNN.

Massive real-time positioning and service algorithms based on satellite navigation reference station network

GU Jianxiang, ZHAO Yipeng, CHEN Yan

2023, 0(11):  48-53,138.  doi:10.13474/j.cnki.11-2246.2023.0326

Asbtract ( )  

References | Related Articles | Metrics

As an important part of modern basic surveying and mapping, its technical means, tools and theoretical methods established and maintained by geodetic datum have changed greatly with the development of society and the progress of science. The Shanghai CORS system for satellite navigation positioning provides high-precision and stable location services for the Shanghai region through the use of grid-based virtual reference station (VRS) technology and the “seven-parameter release” operation mode serves a massive number of users. In this paper, the key technologies used in the construction of the SHCORS system are explored and studied, and its real-time positioning accuracy is tested. The results show that the SHCORS system's accuracy meets practical application requirements.

Multi-source POI location fusion considering address semantics and geospatial features

LI Pengpeng, LIU Jiping, WAGN Yong, LUO An, SANG Yu, YAN Xuefeng

2023, 0(11):  54-60.  doi:10.13474/j.cnki.11-2246.2023.0327

Asbtract ( )   HTML ( )   PDF (1392KB) ( )  

References | Related Articles | Metrics

Multi-source POI location fusion is one of the key technologies for geospatial data matching and fusion. However, due to the difference of location coding and location error between different POI data sources, location fusion becomes more difficult. Multi-source POI location fusion considering address semantics and geospatial feature is proposed. Firstly, semantic features of address attributes are extracted by TextRCNN and graph attention network. Then, Multi-layer perceptron is used to extract geospatial features of location attributes. Finally, multi-source POI location fusion is realized by feature aggregation based on self-attention mechanism. We conduct experimental verification on the POI data of Baidu map, Tencent map and Amap in Chengdu. The results show that this method is significantly superior to the existing methods, and the average location fusion accuracy is better than 12 m.

3D coordinate transformation method of the outlier detection for generalized total least squares

SHEN Zhongshu, XIANG Qianhe, DONG Sixue, WEI Tianshe

2023, 0(11):  61-65.  doi:10.13474/j.cnki.11-2246.2023.0328

Asbtract ( )  

References | Related Articles | Metrics

The parameter estimation will be adversely affected when the observation coordinates are polluted by gross errors. In this study, the 3D coordinate transformation problem is described as a generalized errors-in-variables (EIV) model, and the data snooping algorithm for this model is proposed. Firstly, the generalized total least squares (GTLS) solution of the nonlinear EIV model is derived by using the Euler Lagrange method, and then it is transformed into the classical least squares problem. Two types of test statistics for data snooping are constructed based on the classical least squares theory under the conditions with known and unknown variance component, respectively. The experimental results show that the proposed data detection algorithm can effectively reduce the influence of gross errors and obtain reliable conversion parameters.

Modular extraction method for point cloud of shield tunnel

LIN Lesheng, LIN Song

2023, 0(11):  66-68,106.  doi:10.13474/j.cnki.11-2246.2023.0329

Asbtract ( )  

References | Related Articles | Metrics

This paper proposes an extraction method that combines Hough line detection and image matching for shield tunnel module extraction. Firstly, based on the RANSAC algorithm, the shield tunnel is cylindrical unfolded, and the open tunnel point cloud is binarized. Then, the Hough transform algorithm with constraints is used to extract the shield tunnel ring. Finally, based on the special matching module of the number of bolt holes and bolt hole layout in the middle of the inner sealing module and adjacent module of the tunnel ring, the module point cloud is accurately extracted by combining the ring construction size. The algorithm proposed in this paper is validated by scanning point cloud data obtained from a certain subway, and the results show that the algorithm can effectively achieve modular extraction of shield tunnels.

A Wi-Fi indoor positioning method integrating CNN and CapsNet

ZHANG Tianying, SHI Mingquan, CUI Lizhen, QIN Ling

2023, 0(11):  69-74,121.  doi:10.13474/j.cnki.11-2246.2023.0330

Asbtract ( )  

References | Related Articles | Metrics

Aiming at the problem of low positioning accuracy of Wi-Fi indoor positioning method based on location fingerprint, this paper proposes a Wi-Fi indoor positioning algorithm model that integrates convolutional neural network(CNN)and capsule network(CapsNet) which is recorded as CNN-CapsNet. Firstly, the collected RSSI time series information is used to generate the location fingerprint image dataset. Then the CNN primary feature extractor composed of convolution layer and pooling layer is used to complete the conversion from the positioning image to the primary feature map. Finally, the primary feature map is input into CapsNet to obtain the final classification result. The experimental results show that the accuracy of this model is as high as 99.99% and the loss function is as low as 0.009 91 under different vector dimensions and iteration times, which is better than other traditional positioning methods.

Multi-index extraction method of pavement pothole based on binocular structured light technology

WANG Cong, PAN Jun, SUN Shangyu, SONG Weidong

2023, 0(11):  75-81.  doi:10.13474/j.cnki.11-2246.2023.0331

Asbtract ( )  

References | Related Articles | Metrics

Aiming at the problem of multi-dimensional index extraction of pavement potholes in low light and weak texture environment, as well as the application defects of traditional methods and laser technology, the binocular structured light 3D imaging system is used to obtain pavement pit images, and the RANSAC algorithm is used to fit the road plane, segment the pit point cloud, intersect with the fitted plane as the contour point of the top surface of the pit, sort by the two-way nearest point search method, and calculate the area by the contour point method after counter-clockwise connection. Extract the depth of each data point to calculate the pit depth index. The microelement approximation method is used to slice the pit equidistant and calculate the volume of the pit. BPA algorithm is used to reconstruct the surface to provide a more intuitive data perspective for related application scenarios. The maximum relative errors of depth of pit, area, volume in normal light and weak light and weak texture environment compare with manual measurement results are 2.73%, 6.50%, 8.60%, 2.99%, 5.25% and 8.28%,respectively. According to the calculation index, the severity evaluation of the pit meets the consistency standard of pit evaluation compared with manual measurement. It shows the applicability of equipment and detection methods to weak light and weak texture environment, and can provide data support and technical support for pothole evaluation, repair material estimation, maintenance priority determination, pavement maintenance planning and formulation.

Catenary extraction method combined with multi-level index frame and motion vector

LIN Kailun, YANG Yuanwei, GAO Xianjun, TAN Meilin, ZHANG Yue

2023, 0(11):  82-87.  doi:10.13474/j.cnki.11-2246.2023.0332

Asbtract ( )  

References | Related Articles | Metrics

The research on non-contact detection of catenary of electrified railway is of great significance to ensure the safe operation of the railway. The detection work requires a large number of accurate contact point cloud data support. At present, there is a problem that it is difficult to provide accurate contact point cloud data support due to the difficulty of segmentation between catenary components. Aiming at this problem, this paper proposes a catenary extraction method based on the combination of multi-level index and moving vector. Firstly, the railway scene data is simplified by multi-level index frame.Then the center point set at the bottom of the pillar is obtained by constructing the extraction channel through the trajectory line to calculate the moving vector along the rail. Finally, the attitude of the secondary index frame is adjusted to achieve the accurate extraction of the catenary. In this paper, parameter analysis and comparison experiments are designed, and experimental analysis is carried out in the 10 km railway scene. The results show that the precision, recall and F1 of the algorithm in this paper are about 99%, which are better than the reference algorithm, so the algorithm in this paper can adapt to complex scenes.

Dynamic symbol design and multi-scale representation of scientific greening geographical scene

FU Leyi, ZHAO Xiaoyang, WU Kaihua, SUN Ying

2023, 0(11):  88-94.  doi:10.13474/j.cnki.11-2246.2023.0333

Asbtract ( )  

References | Related Articles | Metrics

As an indispensable key element in the high-quality development of urban construction, the new surveying and mapping technology has been adopted for scientific greening. It provides an all-round technical guarantee for national land greening actions. Based on the human cognitive law of spatial graphics and the spatial representation ability of the map, the utilization of map symbols to visualize the key elements of scientific greening can improve the public's overall knowledge of urban greening and make the concept of scientific greening deeply rooted in people's hearts. Little attention has been paid to scientific greening by map symbols. The traditional design is mostly based on simulation, which is difficult to convey thematic characteristics. Under the new concept of “geographical scene” in cartography, the multi-scale scientific greening dynamic scene by the dynamic map symbol design method is constructed in this study. According to the scene characteristics of provincial large-scale,regional medium-scale and plot level small-scale, it integrates multiple symbol design parameters and emphatically highlights the scientific greening attention indicators for each scale based on dynamic characteristics. In addition, a continuous multi-scale scene for switching adaptively for a better reading experience is designed. This is based on the inversely proportional relationship between information details and viewpoint observation distance. In general, the proposed approach may provide support for the representation of scientific greening scenes and assists in the decision-making of greening management departments.

Application of improved PointNet++ model in extracting road rods

SUN Duanzheng, GAO Fei, YE Zhourun, WU Yanan, ZHANG Shufeng, XIE Ronghui

2023, 0(11):  95-99.  doi:10.13474/j.cnki.11-2246.2023.0334

Asbtract ( )  

References | Related Articles | Metrics

Aiming at the problems of manually designed features for data types, poor universality, and low automation in the extraction of existing road rods, a road rod semantic segmentation method based on an improved PointNet++ deep learning network is proposed in this paper, which realizes the segmentation of road rods. First, the parameters of the original network model such as receptive field and block size are adjusted to make the model more suitable for road point cloud data.And then,aiming at the problem of unbalanced point cloud data, the focus loss function is used as the loss function of the model, so that the categories that occupy a relatively small proportion can be fully trained.At last, to address the problem of the PointNet++ network not considering the relationship between the features of each point in the neighborhood when extracting features, a neighborhood feature aggregation module is used to fuse neighborhood information and improve the learning ability of the network model for point cloud features. To verify the effectiveness of the proposed method, an improved network model was used to conduct experiments on a self-built dataset composed of road point clouds. Compared with the classic PointNet++ network, the segmentation accuracy of rod-shaped objects was significantly improved. The intersection over union (IoU) on simple and complex roads increased by 8.44% and 15.25%, respectively, reaching 98.88% and 92.50%.

Evaluation of Zhengzhou city 3D geological environment carrying capacity based on entropy weight TOPSIS analysis

LI Zhengqian, LI Hongwei, ZHAO Shan

2023, 0(11):  100-106.  doi:10.13474/j.cnki.11-2246.2023.0335

Asbtract ( )  

References | Related Articles | Metrics

The integrated development of urban areas above and below ground is becoming an inevitable trend in contemporary urban development. The carrying capacity of the 3D geological environment of the city is an indispensable component of urban master planning. This article uses Zhengzhou city as an example and utilizes data such as engineering geology, hydrogeology, the current status of underground space development, and surface ecological and transportation environments to create a 3D model of Zhengzhou city's geological environment. Subsequently, relevant indicators are selected, and the entropy-weighted TOPSIS method is used to evaluate Zhengzhou city's geological environment carrying capacity. The results indicate that the carrying capacity of the developed areas in Zhengzhou city is the weakest, the carrying capacity in the northern riverside areas is relatively weak, the central urban areas have a moderate carrying capacity, and the southern areas have relatively strong carrying capacity, providing a reference for the development of underground spaces in Zhengzhou city.

Applicability analysis of monitoring terrestrial water storage changes in the Songliao Basin based on SWARM

SUN Weicheng, WEI Dehong, LUO Zhujian, XUAN Jianhao, ZHANG Xingfu

2023, 0(11):  107-111.  doi:10.13474/j.cnki.11-2246.2023.0336

Asbtract ( )   HTML ( )   PDF (1746KB) ( )  

References | Related Articles | Metrics

In view of the gap between GRACE and GRACE-FO satellites and the damage of the satellite accelerator, SWARM satellites monitoring can be used as an effective supplementary means. This paper selects the SWARM time-variable gravities from ASU, IGG and COST-G to monitor the terrestrial water storage changes in the Songliao Basin, and compares them with the GRACE and GRACEE-FO time-variable gravity. The results show that: ①The accuracy of the first 10-degree coefficients of each SWARM time-variable gravities are close to that of the GRACE gravity, and the signal-to-noise ratio of the IGG-SWARM gravity after 1200 km Gaussian filtering is 62.47% and 55.99% higher than that of the ASU and COST-G gravities. ②The IGG-SWARM time-variable gravity can detect the spatiotemporal characteristics of the large-scale terrestrial water storage in the Songliao Basin, and identify the significant drought and flood events. Both the IGG-SWARM and GRACE and GRACE-FO time-variable gravities indicate that the terrestrial water storage shows an overall upward trend from July 2015 to December 2020 in the Songliao Basin, and the correlation coefficients between them can reach more than 0.6 in the data overlap periods. Therefore, SWARM time-variable gravity can be applied to monitor terrestrial water storage changes in the Songliao Basin.

Application of multi-source data fusion in 3D reconstruction of Guangfu ancient city

ZHANG Xinhang, ZHANG Pei, QI Liang, CHEN Yongli

2023, 0(11):  112-115.  doi:10.13474/j.cnki.11-2246.2023.0337

Asbtract ( )  

References | Related Articles | Metrics

Due to the limitations of objective factors such as object occlusion and aerial photography blind area in the data acquisition process of traditional tilt photography, which cause problems such as 3D model distortion, holes, texture rasping, etc., the method of laser point cloud and tilt photography multi-source data fusion is selected for fine modeling of ancient buildings, which can improve the visual effect and accuracy of 3D models and expand the application field of achievements. In this paper, taking the historical block of guangfu ancient city as an example, through the fusion of multi-source data such as tilt photography, airborne LiDAR, ground hand-held radar and UAV re-shooting, the 3D real scene fine modeling is carried out, and good experimental results are obtained.

Research and application of deep learning-based framework for monitoring and detecting new illegal construction

KANG Tingjun, CHEN Ruixin, SUN Ying, XIA Yixiong, WANG Bin

2023, 0(11):  116-121.  doi:10.13474/j.cnki.11-2246.2023.0338

Asbtract ( )  

References | Related Articles | Metrics

Scientific and intelligent control of new illegal constructions is an inevitable requirement for high-quality urban development. Aiming at the characteristics of new illegal constructions with diverse types, strong concealment, administrative intersection and difficult disposal, this paper adopts a building boundary extraction algorithm based on high-resolution remote sensing images and management data of functional departments, coupled with FPN and Mask-RCNN multi-task fusion. It constructs a whole chain management framework of new illegal constructions from “Construction Behavior Monitoring-Multi-source Data Fusion Analysis-Collaborative Assignment Governance”, which provides technical support for dynamic monitoring and precise management under the high-altitude perspective of the city.

Subpixel water extraction method for Sentinel-2 image

XIONG Longhai, HE Yingqing, LIU Momo, LI Jun

2023, 0(11):  122-127.  doi:10.13474/j.cnki.11-2246.2023.0339

Asbtract ( )  

References | Related Articles | Metrics

Combining Sentinel-2 image and other high-resolution satellite data, it is of great significance to monitor water resources factors such as water surface rate, water storage and ecological flow in long series, high frequency and large range. In order to improve the accuracy of water extraction and solve the problem of spatial scale effect when extracting water from multi-source moderate-and high-resolution satellite data, a subpixel water extraction method for Sentinel-2 image(SWES) is proposed in this paper. First, RWI is used to extract pure water pixel, then the expansion algorithm is used to extract the mixed water pixel. Finally, in order to solve the problem of intra-class spectral changes of surface objects, MESMA considering spatial information is used to solve the water abundance in the mixed water pixel. The results of the three experimental areas all showed that SWES achieved a good effect, the average RMSE was 0.147, and the water extraction effect is better than automatic subpixel water mapping method(ASWM), especially in the pond aquaculture area with more mixed pixels of water and land. The water area obtained by SWES in experimental area also has high accuracy, with the average relative error of 8.03%, which is lower than the 20.23% of ASWM. The results show that SWES can effectively improve the accuracy of water area extraction.

Research on 3D modeling method of city with multi-source data fusion

HU Xiaojing, LIU Yu, CHEN Yanbo, LI Shudan, LI Yao

2023, 0(11):  128-131.  doi:10.13474/j.cnki.11-2246.2023.0340

Asbtract ( )  

References | Related Articles | Metrics

In order to make full use of the advantages of multi-source data and avoid the defects of single data, we use multi-source data to explore the key technologies of multi-source data fusion. Subsequently, a multi-source data fusion modeling method was proposed, which is used to realize the 3D reconstruction of urban entity of a compound in Shaanxi province,and the LOD1—LOD3 level city models are constructed respectively. Finally, the accuracy of the modeling results and the precision of the model were evaluated by using the measured control points reference. The results show that this method can realize multi-source data fusion modeling efficiently. The results of fusion data modeling have improved the precision and precision of the model. The accuracy of the city model with different levels of detail constructed by multi-source fusion data meets the requirements of CityGML for urban entities.

Human-machine collaborative intelligent extraction method of production and construction projectdisturbed patches based on remote sensing image

WANG Songlun, MA Xiaonan, PAN Zixuan

2023, 0(11):  132-138.  doi:10.13474/j.cnki.11-2246.2023.0341

Asbtract ( )  

References | Related Articles | Metrics

Disturbed patch interpretation of production and construction projects in soil and water conservation is mostly realized by artificial visual interpretation of remote sensing images. In the actual work process, there are some problems such as low efficiency, high cost and strong subjectivity.This paper proposes a human-machine collaborative intelligent extraction framework for production and construction project disturbance patches, which combines intelligent extraction model with remote sensing supervision cooperation platform. The change detection dataset is constructed by elements annotation, data enhancement and other means. And then, the improved U-Net++ model is used to carry out the intelligent extraction of production and construction project disturbance patches. The results show that the average accuracy of the model is 79.59%, and the area recall rate of the model is 80.90%.In addition, aiming at the problems that the model easily extracts the pseudo-variation or cloud obscured region incorrectly, patch fragmentation, or irregular contour boundary, a distributed parallel collaborative interpretation platform is built on the basis of automatic extraction results. The platform can realize the functions of adding, deleting, creating, quality inspection, and so on.The final results are fed back to the model as new samples to further improve the performance of the model. Thus form a virtuous cycle between the sample and the model, and improve the actual work efficiency.

Deformation monitoring from InSAR with multi-time baseline sets: a case study of Pingzhai Reservoir

HUANG Youju, QIN Yiting, WU Hui, WEI Qiang

2023, 0(11):  139-144.  doi:10.13474/j.cnki.11-2246.2023.0342

Asbtract ( )  

References | Related Articles | Metrics

Time baseline plays an important role in temporal InSAR deformation extraction. Aiming at the problem of difficult extraction of deformation information in mountainous areas with high vegetation cover and water content and large topographic relief, the historical image data of Sentinel-1A radar satellite from January 2019 to December 2020 are acquired with Pingzhai Reservoir near Guiyang as the study area, and different time baseline sets are constructed to compare and analyse the extracted surface deformation information. The baseline conditions that are most suitable for the extraction of deformation in the study area are selected by the principle of shortest time baseline sum, and the subsidence rate of the study area for two years is obtained based on the SBAS-InSAR technique. The results show that the most accurate deformation information is extracted from the 60 d time baseline set in the study area, and there are fewer misjudgments and omissions than other time baseline sets. The relationship between the temporal variation of dam deformation and rainfall is also analysed in the context of the average rainfall data from 2019 to 2020 in Guiyang and geometric level monitoring data. It is concluded that there is a strong correlation between rainfall and dam deformation, and it is also concluded that the deformation information extracted based on the SBAS-InSAR technique is convergent with the level monitoring results.

Evaluation of the optimization effect of “ecological-production-living” spatial pattern for comprehensive land management in plain agricultural area

JIANG Ling, ZHANG Dapeng, HUANG Danni, HUANG Xiaoli, CHEN Xi, LI Peng, CHEN Wanli, WANG Yongfeng, WANG Dong

2023, 0(11):  145-151.  doi:10.13474/j.cnki.11-2246.2023.0343

Asbtract ( )  

References | Related Articles | Metrics

In this study, a method for evaluating the optimization effect of the “ecological-production-living” spatial pattern about comprehensive land management in the whole region based on three dimensions (spatial quantity, quality and structure) is constructed. The Fandian Community (provincial pilot project of comprehensive land management), Feidong county is acted as an example to analyze the optimization effect of spatial pattern. The results show that:①The three dimensions (spatial quantity, quality and structure) of spatial pattern in Fandian community will be optimized by comprehensive land management through the reconstruction of spatial organization, elements and structure.②Because the optimization effect of the “ecological-production-living” spatial pattern is greatly affected by the optimization effect of the production spatial pattern.More attention should be focused on optimizing the production space when in the practice of comprehensive land consolidation.③The problems of low value and low efficiency in production space, disordered waste in living space, disordered pollution in ecological space from the un-optimization of spatial pattern will be solved through comprehensive land management.

Entity model and its application for thematic analysis of construction land

JIANG Fengchai, LIAN Xu, XU Ran, LEI Lizhen, LIU Xuejun

2023, 0(11):  152-157.  doi:10.13474/j.cnki.11-2246.2023.0344

Asbtract ( )  

References | Related Articles | Metrics

The traditional construction land data organization and management mode based on GIS segmentation and layering is difficult to meet the needs of land life cycle data association application in the process of natural resources digital transformation. How to release the value of natural resources and spatial geographic data to meet the diversified and rapid data thematic analysis needs,in the process of natural resources digital transformation still has a large exploration space at this stage. Based on the analysis of the policies and regulations and the technical guidance documents of the Ministry of Natural Resources,this paper puts forward the idea of building the physical model of construction land,selects the pilot area to build the thematic analysis data environment of construction land,and explores the implementation path of one-click rapid thematic analysis around five typical thematic application scenarios,including batch but not supplied quantity accounting,construction land backtracking,idle land analysis,temporary land monitoring and supervision,and suspected illegal land identification.It provides exploration experience for the construction and application of physical data in the process of digital transformation of natural resources.

High-precision 3D real scene construction method and practice for natural resources survey and monitoring system in Inner Mongolia Autonomous Region

LI Dezhong, YAO Na, XUE Yongjun, XIONG Wenhao, WANG Longxiu

2023, 0(11):  158-162.  doi:10.13474/j.cnki.11-2246.2023.0345

Asbtract ( )  

References | Related Articles | Metrics

The establishment of a natural resource survey and monitoring system requires a unified 3D spatial positioning framework and data analysis foundation. This paper takes into account the topographic features, regional economic pattern, and high-quality land and space system goals of Inner Mongolia Autonomous Region, and studies the overall layout and construction method of building a high-precision 3D real scene natural resource survey and monitoring data base by layering, grading, and different precisions, focusing on the technical route, technical indicators and practical results of the 3D real scene construction of the terrain level and city level. The accuracy of the relevant practical results is better than the designed technical indicators. The results have effectively served Inner Mongolia's natural resource management, urban land space monitoring, 3D spatio-temporal database construction and other purposes, and promoted the construction process of 3D real scene Inner Mongolia.

Improvement and practice of geographic information science professional training program in the era of big data intelligence

CHEN Jie, DENG Min, LIU Qiliang, SHI Yan, LIU Huimin

2023, 0(11):  163-167.  doi:10.13474/j.cnki.11-2246.2023.0346

Asbtract ( )  

References | Related Articles | Metrics

Geospatial and temporal information has empowered all walks of life. According to different professional positioning, each university has determined the characteristic, direction and focus of professional construction based on its own conditions. The professional cultivation of geographic information science in the author' school has the typical engineering advantages and geoscience application service characteristics, adhering to the training idea of “thick theory-heavy technology-strong service”, and exploring the key measures for high-quality talent training under the background of big data and artificial intelligence. By continuously improving the quality talent training program, improving the curriculum system, optimizing course content, and innovating the teaching mode, the healthy development of the geographic information science major in the author' school.

DSRT teaching mode driven by practice and innovation:teaching innovation in GIS Spatial Analysis course

LI Shaoying, ZHANG Xinchang, WU Zhifeng, CHEN Chengjing

2023, 0(11):  168-172.  doi:10.13474/j.cnki.11-2246.2023.0347

Asbtract ( )  

References | Related Articles | Metrics

Based on the cross-disciplinary characteristics and past teaching pains of GIS Spatial Analysis, our team explores the teaching mode of GIS for the international academic frontier and the latest needs of industries in line with the concept of emerging engineering education. A diversified 'situation-role-task' mode (DSRT) that balances practical and innovation abilities has been proposed. Through content reconstruction, comprehensive situational activities design and four-dimensional assessment system construction, interests of students would be stimulated. These measures comprehensively cultivates student's geoscience literacy, engineering practice, innovation and complex problems solving abilities, which can provide references for the cultivation of composite geographic information science talents in universities.

Method for extracting indoor parking lot ground marking elements based on handheld laser scanning point cloud

LIU Yuntong, HUANG Jinting, WANG Jiayao

2023, 0(11):  173-176.  doi:10.13474/j.cnki.11-2246.2023.0348

Asbtract ( )  

References | Related Articles | Metrics

In order to meet the demand for rapid mapping of indoor parking lots, a method is proposed for extracting ground marking features in indoor parking lots based on handheld LiDAR point clouds. Firstly, to reduce the memory space requirements for feature extraction, the entire point cloud is divided into regular grids. Then, the RANSAC plane fitting method is used to extract the ground point cloud within each grid. In order to extract ground marking features, a ground image is generated based on the ground point cloud. On this basis, the BiSeNet network is employed for semantic segmentation of different marking features, such as lane lines, parking lines, and directional arrow markings, to obtain the corresponding pixels. For lane lines and parking lines, a line extraction method based on Hough transform is used, while for ground arrow markings, a template matching method is applied for extraction. Experimental results demonstrate that the proposed method can quickly extract both structural elements and marking features from scanned data, significantly reducing manual mapping workload and improving the efficiency of indoor parking lot mapping.

Railway line informatization based on multi-source point cloud data

YANG Zhijian

2023, 0(11):  177-181.  doi:10.13474/j.cnki.11-2246.2023.0349

Asbtract ( )  

References | Related Articles | Metrics

Compared with traditional measurement methods, mobile scanning technology can improve the efficiency of measurement. However, point cloud data has the characteristics of large volume, high density and redundant data. During the measurement process, the scanner coverage is large and a large number of invalid point clouds are obtained. To solve this problem, a method of extracting railway line information based on multi-source point cloud data is proposed. According to the reflection intensity and geometric features of the rail point cloud, the center line of the line is extracted by the idea of differential. The dimension reduction of the rail point cloud model is processed, and the rail model is established according to the allowable error of the rail head width of the high-speed railway road rail as the convergence condition, and the plane line and longitudinal and transverse sections of the line are extracted. Based on the reflection intensity, the rail is quickly segmented, and the Dynamo programming language is used for secondary development of Revit to quickly model, which plays an important role in promoting the intelligent and visual level of railway design.