Table of Content

25 March 2024, Volume 0 Issue 3

Previous Issue    Next Issue

A method of automatic mapping of gullies based on GF-7 satellite image in the black soil region in Northeast China

CHEN Chang, ZHANG Yan, LI Kunheng, YANG Runze, ZHANG Junbin, LIANG Yanrong

2024, 0(3):  1-7.  doi:10.13474/j.cnki.11-2246.2024.0301

Asbtract ( )   PDF (7648KB) ( )  

References | Related Articles | Metrics

In the black soil region of Northeast China, gully erosion is severe and widespread. Currently, gully monitoring in this area relies predominantly on manual interpretation, highlighting the urgent need for a rapid extraction method. This study selects the Mashezi River Basin in Binxian country, Heilongjiang province, a region heavily affected by gully erosion, as study area. Utilizing GF-7 satellite imagery and comparing with manual interpretation results, the accuracy of three automatic gully extraction methods is evaluated: flow-directional detection, machine learning and deep learning. The findings are as follows: ① The flow-directional detection method depends on high-precision topographic data. The vertical accuracy of topographic data generated from GF-7 stereo images is poor, resulting in an overall extraction accuracy of only 6.7%, and this method is unable to automatically extract permanent gullies and ephemeral gullies from GF-7. ②The machine learning approach requires manual setting of segmentation parameters and design of classification features, limiting its degree of automation. It achieves an overall extraction accuracy of 50.7%, with a precision of 83.1% for permanent gullies and only 9.2% for ephemeral gullies. ③The deep learning method adopts an end-to-end approach, without the need to design feature extractors. It offers a high degree of automation with an overall extraction accuracy of 60.8%, achieving 68.1% accuracy in identifying permanent gullies and 69.7% in recognizing ephemeral gullies.

Methodology for estimating Cd content in farmland soil based on GF-1 remote sensing images

ZHANG Longqi, GUO Yunkai, DONG Shengguang, LIU Xinliang

2024, 0(3):  8-12,94.  doi:10.13474/j.cnki.11-2246.2024.0302

Asbtract ( )   PDF (3760KB) ( )  

References | Related Articles | Metrics

This study explores the feasibility of estimating soil cadmium (Cd) content in farmland using GF-1 remote sensing satellite imagery. Correlation and different regression analyses have been separately conducted with sample Cd content and logarithmic, square root, and inverse square root transformation of image spectral image, which is filtered out vegetation information in the pre-processed remote sensing images. The linear regression model, with an accuracy above 0.95, is selected using competitive adaptive reweighted resampling based on the inverse square root transformation. The remote sensing estimates results, however, revealed a significant number of anomalous values in areas such as ponds, flooded rice fields, rooftops, hardened road surfaces and so on. An interpolation is employed with neighboring normal estimates to replace these anomalies, resulting in the final estimated values. Correlation analysis and modeling accuracy assessments suggest that this method is feasible and holds promise for practical applications in soil quality monitoring and land management.

Non-agricultural monitoring and spatio-temporal analysis study of cultivated land based on deep learning method:a case study of Kaiyang county

ZHANG Lanlan, WANG Honglei

2024, 0(3):  13-18.  doi:10.13474/j.cnki.11-2246.2024.0303

Asbtract ( )   PDF (7614KB) ( )  

References | Related Articles | Metrics

How to quickly detect illegal cultivated land non-agriculturalization and understand its spatial distribution and change process is a central issue of fundamentally reducing cultivated land non-agriculturalization.Based on multi-temporal remote sensing image,a monitoring index system for land non-agriculturalization and a sample database with local topographic features are established.A remote sensing change detection model is established using deep learning technology,and applied to the temporal monitoring of land non-agriculturalization in Kaiyang county.On this basis,the spatial and temporal distribution characteristics of regional land non-agriculturalization are discussed using kernel density estimation spatial analysis method.The results show that the combination of satellite remote sensing and deep learning technology can achieve rapid and dynamic monitoring of land non-agriculturalization in a large range.The overall trend of new illegal land non-agriculturalization activities monitored in Kaiyang county from April 2021 to December 2022 is decreasing,but there are local aggregation areas and the number of illegal activities shows relatively obvious seasonal characteristics.

Vehicle speed detection method in single-camera videos in close-range photogrammetry

ZHANG Shaobin, ZHANG Zhihua

2024, 0(3):  19-24.  doi:10.13474/j.cnki.11-2246.2024.0304

Asbtract ( )   HTML ( )   PDF (6568KB) ( )  

References | Related Articles | Metrics

Speed detection plays a crucial role in ensuring the safe operation of vehicles in urban transportation systems, making it essential for maintaining traffic safety. However, existing methods for measuring vehicle speed suffer from high costs, susceptibility to external conditions, and limitations in installation areas. To address these issues, this paper proposes a low-cost and flexible vehicle recognition and speed measurement method based on video imagery. The approach utilizes deep learning techniques to construct the YOLOv4 framework and train it on the COCO dataset for vehicle identification. The recognition method is improved by extracting the pixel coordinates of the midpoint of the lower boundary of the maximum bounding rectangle encompassing the recognized vehicles. Additionally, a close-range photogrammetry method is introduced, and improvements are made to the collinearity equations to enable vehicle recognition in a single-camera setup. The displacement of vehicles is computed within a fixed time interval, and a velocity curve of vehicles within the monitoring area is plotted. Experimental validation is conducted to assess the feasibility and accuracy of the proposed method.

Reconstruction of high-precision real 3D data of offshore islands based on drone tilt photography and laser scanning

ZHANG Shuhang, LI Wangmin, PANG Yinning, CHANG Bingtao, MA Defu, LI Xiulong, ZHANG Wuming

2024, 0(3):  25-30.  doi:10.13474/j.cnki.11-2246.2024.0305

Asbtract ( )   HTML ( )   PDF (5347KB) ( )  

References | Related Articles | Metrics

This paper addresses the needs and technical challenges of real 3D for offshore islands, taking the Miaowan Island in Zhuhai, China as the observation object, using UAV oblique photogrammetry and laser scanning for real 3D data acquisition of offshore islands. Details of the UAV mission planning, ground control and live 3D data processing work, and DEM generation based on parameter-free cloth simulation filtering are presented. The absolute accuracy of the data is verified by checkpoints. The results show that the total mean square error of the acquired ortho-and oblique images are 14.72 and 6.60cm,respectively; the errors of the point cloud, DSM and DEM elevations are 3.10, 6.20 and 7.11cm,respectively, which meet the needs of offshore island live 3D display and survey management.

Spatial-temporal evolution and driving factors analysis of oasis evapotranspiration in the middle reaches of the Keriya River basin

WANG Ranran, Lü Guanghui, HE Xuemin, LI Jinbao

2024, 0(3):  31-36,139.  doi:10.13474/j.cnki.11-2246.2024.0306

Asbtract ( )   PDF (2925KB) ( )  

References | Related Articles | Metrics

Evapotranspiration is a key component in assessing the water-heat cycle of desert oasis areas, and monitoring its spatio-temporal variations and studying its driving forces can provide scientific basis for precise water resources management and ecological environment protection. Taking the middle reaches of the Keriya River basin as the study area, this paper analyzes the spatio-temporal variations of evapotranspiration from 2010 to 2022 based on Landsat remote sensing imagery and the SEBS model. The accuracy of the estimation results is verified using evaporation pan measurements and the Penman-Monteith model, and further investigation is conducted on the influencing factors of evapotranspiration. The results show that:①The correlation coefficient and R² between the SEBS simulated values of evapotranspiration and the evaporation pan observations are 0.93 and 0.87, respectively, with an RMSE of 0.96mm/d. The correlation coefficient and R² between the SEBS simulated values and the Penman formula observations are 0.90 and 0.81, respectively, with an RMSE of 0.64mm/d. ②The actual evapotranspiration from 2010 to 2022 shows a decreasing trend, with a change rate of 14.75mm/a. It decreases in spring, summer, and autumn, while increases in winter. ③The spatial heterogeneity of evapotranspiration is evident, with high values mainly concentrated near the Keriya River and low values distributed in the sandy areas at the edge of the oasis. In the past 13 years, about 70.2% of the study area's pixels show no significant decrease trend, while 10.4% of the pixels exhibite a significant decrease trend. ④Evapotranspiration is significantly correlated with temperature, air pressure, sunshine duration, land surface temperature, and NDVI. It has a weak correlation with wind speed and surface albedo. Overall, the research results can effectively simulate the spatio-temporal heterogeneity of evapotranspiration in the middle reaches of the Keriya River basin. This study contributes to the rational planning and management of water resources in the basin.

Multi-feature and multi-level Sentinel-2 image extraction of lake and reservoir water bodies in Liaoning province

LI Wenkang, ZHAO Quanhua, JIA Shuhan, LI Yu

2024, 0(3):  37-42,106.  doi:10.13474/j.cnki.11-2246.2024.0307

Asbtract ( )   PDF (4738KB) ( )  

References | Related Articles | Metrics

This article takes Liaoning province as the research area, based on the GEE (Google Earth Engine) remote sensing cloud platform, and using Sentinel-2 remote sensing images, proposes a multi-feature and multi-level algorithm for extracting lake and reservoir water bodies. This algorithm selects the automatic water index (AWEI_sh) and the improved normalized water index (MNDWI) to extract water bodies, and uses the normalized vegetation index (NDVI), normalized building index (NDBI), normalized difference red edge index (NDREI), Sentinel-2's B8 and B9 bands, as well as DEM data to multi-level eliminate dark and bright ground noise, and to repair partially missing water bodies in the extraction results that are obscured by clouds and mist. Finally, remove the river and small pixels. This algorithm is used to extract lake and reservoir water bodies in Liaoning province from April, July, and October of each year from 2017 to 2021. Different water body extraction algorithms and water body data products were compared. The experimental results showed that the proposed algorithm had good performance in extracting water bodies under large-scale conditions, with an overall accuracy of over 96%. It can effectively remove dark pixel surfaces such as vegetation and shadows, and ensure the integrity of water body information, It has certain applicability and stability in large-scale water extraction.

Impervious surface extraction and expansion analysis using SVM mixed kernels

JI Jianren, WANG Jingxue, WANG Liqin

2024, 0(3):  43-48.  doi:10.13474/j.cnki.11-2246.2024.0308

Asbtract ( )   PDF (2398KB) ( )  

References | Related Articles | Metrics

Support vector machines use a single kernel function for impervious surface extraction that produces high time complexity and low extraction accuracy.In order to solve the above problems, this paper introduces polynomial kernel functions on the basis of radial basis kernel functions and proposes an impervious surface extraction method with mixed kernel functions. Firstly,since the features with different properties have similar spectral information, this paper combines the spectral information with the image entropy texture information in the process of feature extraction. It is possible to distinguish between the categories of things more clearly. Then,on the basis of the radial basis kernel function, polynomial kernels are introduced to obtain the feature information of images from local and global perspectives respectively, and improves the extraction accuracy of impervious surfaces. Finally,based on the results of impervious surface extraction, the spatio-temporal evolution analysis is carried out.In this paper, Landsat images from 2009 to 2021 in the main urban area of Fuxin city are used for experiments.The experimental results show that the combination of spectrum and entropy texture can improve the feature extraction effect and extraction accuracy of impervious surface.Compared with the single kernel function extraction method, the impervious surface extraction accuracy of the proposed method is improved by 2.5%, indicating the effectiveness of the proposed method.

A study of supraglacial lakes in Baltoro Glacier based on Sentinel-2 images

LIU Xiao, SUN Yongling, SUN Shijin, LI Min

2024, 0(3):  49-53,80.  doi:10.13474/j.cnki.11-2246.2024.0309

Asbtract ( )   PDF (9419KB) ( )  

References | Related Articles | Metrics

Supraglacial lake is an important part of glacier and an indicator of glacier ablation, which not only responds rapidly to global climate change, but also has important significance for the understanding and mastering of regional water resources information. In this paper, the Sentinel-2 images and random forest algorithm are employed to identify and extract the supraglacial lakes of Baltoro Glacier. Furthermore, the spatial distribution characteristics of supraglacial lakes and the relationship between the area/number of supraglacial lakes and glacier elevation are analyzed. The accuracy, completeness and error rates of supraglacial lakes extraction in this paper are 96.07%, 92.18% and 11.59%, respectively. A total of 567 supraglacial lakes are identified in Baltoro Glacier, with an area ranging from 249.46 to 37134m², and most of them are distributed between 3 and 26km from the glacier terminus. Moreover, the number of supraglacial lakes is the largest between 3800 and 4300m with generally large area, and the average area is 1922m². With the increase of elevation, the number and area of supraglacial lakes gradually decrease. In particular, there are only 15 supraglacial lakes with an average area of 356m²in the elevation of above 5300m. The decrease in glacier surface temperature due to the increase in elevation is the main reason for the sharp decline in the number and area of supraglacial lakes.

Research and analysis on land feature coordinate update strategy based on CGCS2000 new framework

CHENG Yingyan, CHANG Chuntao, WANG Hu, XU Yantian, XU Changhui

2024, 0(3):  54-62.  doi:10.13474/j.cnki.11-2246.2024.0310

Asbtract ( )   HTML ( )   PDF (2597KB) ( )  

References | Related Articles | Metrics

This paper analyzes and studies the update strategies and methods of the nation-wide entity coordinate results after the new CGCS2000 frame update, and provides ideas and countermeasures for the future CGCS2000 frame maintenance on aspects of the GPS station layout, data dealing methods and product accuracy possible achieved. In this paper, the data of dense CORS station abroad, which is roughly the same in terrain and area as China, can be downloaded publicly, is used to design and process the GPS observation data set with different time length and the required cm-level and decimeter-level CGCS2000 coordinate data set for this study. The improvement of CGCS2000 coordinates of astronomical geodetic network from decimeter level to centimeter level and the implementation of dynamic updating of CGCS2000 mapping products based on CORS update framework are discussed respectively. The factors that affect the accuracy realization: grid interpolation method, control point density, search range, grid spacing, etc, are fully analyzed by experiments, and it is determined that the maximum control point spacing for improving the precision of CGCS2000 coordinates for astro-geodetic points, is 2.2°, corresponding to 230km, which requires 363 points nationwide. In order to facilitate the maintenance of the regional framework and at least 4 stations guaranteed in each province (city), GPS observations need to be carried out at 2178 astronomical sites nationwide. In order to realize the dynamic update of CGCS2000 mapping products at an interval of 10 years, when 985 CORS stations are deployed, the accuracy of 5cm can be achieved when the control point spacing in the eastern, central and western regions of China is 482, 482 and 107km respectively. 2170 CORS stations are needed to deployed nationwide, which can reach the accuracy of 3cm when the density of control points in the eastern, central and western regions of China reaches 118, 374 and 75km respectively. Because the test area is geographically approximate rather than completely consistent, the conclusions obtained will differ from the reality, and only provide reference for the update mechanism of surveying and mapping products in China.

An odometer-aided GNSS/INS integrated navigation algorithm under the framework of FGO

TANG Weiming, QI Kepei, DENG Chenlong, ZOU Xuan, LI Yangyang, HU Zeqi

2024, 0(3):  63-68.  doi:10.13474/j.cnki.11-2246.2024.0311

Asbtract ( )   PDF (1759KB) ( )  

References | Related Articles | Metrics

In a complex observation environment, the GNSS signal of the GNSS/INS integrated navigation system is susceptible to be disturbed and result in a rapid decline in the accuracy of INS independent navigation. Aiming at the above problems, this paper studies the odometer-assisted GNSS/INS integrated navigation algorithm based on the factor graph, uses the odometer observation information combined with the non-holonomic constraint to construct the heading speed constraint equation, and adopts factor graph optimization for parameter estimation which conducts multiple linearization calculations and multiple iterations at the same time. The results of real vehicle experiments show that when the GNSS signal is good, the factor graph-based method has a faster convergence time than the filtering method, and the convergence speed is increased by about 10 times; when the GNSS signal is interrupted, the positioning accuracy of odometer-assisted integrated navigation system in the E and U direction has increased by 83% and 89% respectively. And compared with the conventional Kalman filter method, the positioning accuracy in the E and N can respectively be improved by using factor graph optimization in this paper. There are 63% and 70% improvements.

Robust regression soil moisture retrieval method for GNSS-IR multi-star fusion

WANG Shitai, YANG Kexin, YIN Min, MA Yue, JIANG Wei, LIU Xu, WEI Jialin

2024, 0(3):  69-74.  doi:10.13474/j.cnki.11-2246.2024.0312

Asbtract ( )   PDF (1718KB) ( )  

References | Related Articles | Metrics

Global navigation satellite system interferometry (GNSS-IR) can extract effective information such as soil moisture and sea surface height by analyzing interference information between direct and reflected satellite signals. In order to reduce the weight of outliers by using robust regression method, this paper proposes to reduce or offset the influence of anomalous observation data on soil moisture inversion. In order to verify the application range of the model, multi-star fusion experiments are carried out on the basis of robust regression, which effectively improve the accuracy of soil moisture inversion. The results show that compare with the traditional linear regression method, the RMSE and MAE of the proposed method are reduced by 8.38% and 8.91% on average for a single satellite, and by 15.18% and 16.42% on average for two satellites,by 21.00% and 22.97% on average for three satellites, by 26.25% and 28.71% on average for four satellites.

Research and application of multi-source data acquisition method for inland underwater and nearshore terrain

JIA Su, XIE Jiafen, DING Shijun, ZHANG Jinying, WANG Ju

2024, 0(3):  75-80.  doi:10.13474/j.cnki.11-2246.2024.0313

Asbtract ( )   PDF (4845KB) ( )  

References | Related Articles | Metrics

The acquisition of inland underwater and nearshore topographic data is helpful for comprehensive development and refined management of inland waters, and has important social, economic and ecological significance. Aiming at the problem that a single data acquisition is difficult to meet the complex terrain measurement, this paper proposes a multi-source data acquisition method, which adopts aerial photography, airborne LiDAR, multi-beam and single-beam bathymetry system, artificial measurement and other methods to acquire multi-source data. Data processing obtains terrain data that comprehensively covers underwater, near-shore, and land-water junctions. This paper takes Nishan reservoir as an example for experimental application. The results show that the method in this paper not only realizes the acquisition of integrated topographic data on water and underwater, but also ensures the reliability of the accuracy of the measurement results. If can be widely used in large and medium-sized reservoir survey project to provide technical support for inland underwater and nearshore topographic survey.

BIM-oriented indoor topology-grid hierarchical path planning method

GUO Ruirong, LI Chaokui, LI Hao, CHEN Jun

2024, 0(3):  81-87.  doi:10.13474/j.cnki.11-2246.2024.0314

Asbtract ( )   PDF (6681KB) ( )  

References | Related Articles | Metrics

A BIM-based indoor topology-grid hierarchical path planning method is introduced to solve the problems of large search space, low efficiency, and poor obstacle avoidance safety in traditional methods in complex indoor environments. Firstly, a BIM model with a complex structure is established to extract the obstacles in the model, the semantic and geometric information of the channel, and the correlation information between channels, etc. The basic navigation map is obtained by the grid mapping method, and the map is layered by using the hierarchical map idea to obtain the topology-grid layered map. Secondly, the offline prior road network between each sub-area in the topology layer is generated through a thinning algorithm, and the optimal path is selected using the Dijkstra algorithm. The optimal path of the raster layer is quickly and efficiently searched using the improved A* algorithm. Construct a complete global optimal path by splicing the local optimal path of the topology layer and the grid layer. Finally, Comparing this method with the standard A* algorithm and ant colony algorithm, while ensuring calculation efficiency, it not only reduces the path search space, but also ensures the safety of the optimal path, comprehensively verifying the superiority of the proposed path planning method.

Green visible index extraction and analysis of street view image using DeepLabv3+model:taking within the Third Ring Road in Beijing as an example

WANG Hongyan, CHE Xianghong, XU Xinchao, XU Shenghua, LI Hongsheng

2024, 0(3):  88-94.  doi:10.13474/j.cnki.11-2246.2024.0315

Asbtract ( )   PDF (9892KB) ( )  

References | Related Articles | Metrics

The extraction of green vision index based on semantic segmentation model lacks applicability. Based on DeepLabv3+semantic segmentation pre-training model and manually-labeled samples, this study adopts transfer learning strategy to build a semantic segmentation model for street view image, and evaluates model performance. Then, based on the semantic segmentation model of street view image, GVI within the Third Ring Road in Beijing is extracted and calculated, and the spatial distribution characteristics of GVI at point and line scales are analyzed. The results show that:①Compared with DeepLabv3+green vision segmentation pre-trained model, F1 value and mIoU value of the transferred model are increased by 7% and 3%, respectively.②The GVI in the study area at the point scale has a clustering pattern which is high in the northwest and low in the southeast. There are 58.1% of street view sampling points with the GVI values ranging from 0 to 0.15 which indicates a relatively low green vision perception degree within the Third Ring Road in Beijing.③The GVI values at the linear scale are low on the ring road and high between the rings and witnesses a center diverges outward. There are 59.8% of roads with the GVI values ranging from 0 to 0.15. This study can provide an significant reference for improving the perception degree of urban street greening and urban spatial planning.

On the collaborative technical framework of land spatial governance driven by high-quality development

LI Yaxing, LI Xiaoming, LIAO Chuangchang, JIANG Lin, HONG Wuyang, WANG Weixi, GUO Renzhong

2024, 0(3):  95-100,144.  doi:10.13474/j.cnki.11-2246.2024.0316

Asbtract ( )   HTML ( )   PDF (2565KB) ( )  

References | Related Articles | Metrics

The support technology for modernization of land and space governance faces a series of serious challenges. In order to respond to the new height strategy of high-quality development, this paper compares the current situation, problems of the main technologies (perception technology, digital intelligence technology, planning technology and early warning technology) under the support technology system for modernization of land and space governance and the needs for the new requirements in the new era, builds a new support technology oriented to perception stereoscopic, data valorization. Finally, this paper designs and constructs a modernization platform system of self-awareness, self-learning, self-adaptive and self-governance for the concept of high-quality development.

An experimental eye-tracking study of map cognitive load differences for urban structural patterns

LI Xue'er, ZHENG Shulei

2024, 0(3):  101-106.  doi:10.13474/j.cnki.11-2246.2024.0317

Asbtract ( )   HTML ( )   PDF (3825KB) ( )  

References | Related Articles | Metrics

In the field of spatial cognition, with the advancement of cartographic technology, 3D maps, image maps, virtual maps and so on are appeared, which brings about the problem of unclear applicability of map cognition, and lack of quantitative research on differences in cognitive load. This paper is oriented to the problem of urban structural form cognition, taking 2D maps, 3D maps, image maps and aerial map pictures as examples, and adopting the methods of questionnaire survey, eye movement experiment and scale analysis to study the cognitive roles of different types of maps, and at the same time to play attention to the gender differences therein. The results show that both map type and gender variables had a certain effect on the cognitive load of maps with significant differences between groups; 2D maps are more applicable, and the cognitive load and cognitive effect of men are higher than that of women.

Application of point cloud data in geometric detection of critical nodes of turnouts

WANG Dongyan, YU Cai, SHEN Kun, ZHANG Zhenjian, LI Yafeng

2024, 0(3):  107-112.  doi:10.13474/j.cnki.11-2246.2024.0318

Asbtract ( )   PDF (1884KB) ( )  

References | Related Articles | Metrics

The process of detecting the operational status of a turnout is complex, and traditional methods require the use of rail inspection vehicles, rail gauges, support gauges, and height gauges for measurement. The equipment types are varied, the measurement process is lengthy, and there is a high demand for the time window. To improve detection efficiency, this paper proposes a key node measurement method for single turnout based on point clouds. The method utilizes CAD graphic elements and graph convolutional neural networks to achieve the precise, automated identification, segmentation, and extraction of the three-dimensional point cloud data of the turnout structure, with an accuracy rate of 99.68%. At the same time, by combining the geometric prior information of the turnout structure, the key geometric position parameters such as the track gauge, curve radius, support spacing, and rail elevation drop are accurately extracted in a rapid and precise manner. Verified by examples, the measurement results of the key geometric position detection method for turnouts based on point cloud proposed in this paper have a sub-millimeter error with the real value, which meets the requirements of actual engineering inspection, eliminates a variety of inspection equipment and saves a lot of time for skylight, which has a high degree of practicability, and it is the development trend of future turnout measurements.

Application of multi-spectral remote sensing images in monitoring the water level of Yangtze River

WEI Xiang, TIAN Chengshuo, QIN Sixian, DUAN Mengmeng

2024, 0(3):  113-117.  doi:10.13474/j.cnki.11-2246.2024.0319

Asbtract ( )   PDF (4921KB) ( )  

References | Related Articles | Metrics

Human social life and production activities are closely related to changes in river water levels. River water level monitoring system can real-time get the changes in river water level, and is an important means for scientific warning of water hazards, ensuring the safety of ports and shipping, and improving flood control and drought resistance capabilities. This paper studies the correlation between the local water area of Yangtze River and the observed water level values, and establishes a regression equation. Therefore, using multi-spectral remote sensing images to extract the area of the Yangtze River water area, this regression equation can be used to calculate the water level. The results show that this method can provide a simple and feasible estimation method for water level monitoring of rivers, lakes, and reservoirs.

A method for calculating the height of a single tall building based on convolutional neural networks and drone oblique photography images

GAO Guitang, XI Lianxia, ZHONG Xiaolong

2024, 0(3):  118-122.  doi:10.13474/j.cnki.11-2246.2024.0320

Asbtract ( )   PDF (4937KB) ( )  

References | Related Articles | Metrics

In order to improve the height calculation effect of a single high-rise building, a method based on convolutional neural network for calculating the height of a single high-rise building using unmanned aerial vehicle oblique photography images is proposed. Firstly, drone tilt photography technology is used to capture images of high-rise buildings.Secondly, input high-rise building images into the convolutional neural network and perform enhancement processing on them to improve the clarity of the image.Finally, the shadow length is calculated using the band conversion index and shadow index, and the height of a single high-rise building is calculated based on this. The experimental results show that the proposed method has good image enhancement processing effect, high measurement accuracy, and efficiency.

Extraction of typical natural resource elements based on multi-source high-resolution remote sensing images

MA Jinshan, JIA Guohuan, ZHANG Sai, ZHANG Jiong

2024, 0(3):  123-126,150.  doi:10.13474/j.cnki.11-2246.2024.0321

Asbtract ( )   PDF (17575KB) ( )  

References | Related Articles | Metrics

Using high-resolution remote sensing data with high spatial resolution characteristics, typical natural resource elements are extracted based on traditional convolutional neural network deep learning algorithms using multi-source high-resolution remote sensing images of 0.3 and 1m in Xining, Qinghai province as data sources. The results show that the accuracy of extracting farmland and forest land from 0.3m remote sensing images is over 85%, with a recall rate of over 89%. The accuracy of extracting farmland and forest land from 1m remote sensing images is over 90%, with a recall rate of over 91%. The research results can be used for intelligent extraction of typical elements of natural resources in Xining.

Extraction of maize lodging range from remote sensing image based on canopy height model

ZHAO Lian, YU Yajie, LIANG Zhihua

2024, 0(3):  127-133.  doi:10.13474/j.cnki.11-2246.2024.0322

Asbtract ( )   PDF (5787KB) ( )  

References | Related Articles | Metrics

Accurate extraction of maize lodging area is the basis of accurate field management and estimation of maize yield loss, and the remote sensing image acquired by UAV is flexible, which is a popular method for crop lodging measurement.However, most of the existing researches use spectral and texture features, which are easily affected by shooting time, terrain, angle and so on. The method of extracting maize lodging range based on canopy height difference is developed by using unmanned technology. Firstly, the background soil distribution is extracted by the visible light band differential vegetation index. And then the height of maize is extracted. Finally, the maize lodging range is extracted based on SVM and OSTU automatic threshold method. The experimental results show that the classification accuracy of SVM for three samples is 88.84%,89.52% and 90.80%,respectively, and for OSTU automatic threshold method is 94.61%,89.74% and 97.20%,respectively, which is slightly better than the former. In this study, crop lodging is extracted based on crop height as a structural parameter. The mechanism is clear and the effect of UAV imaging instability is eliminated to some extent.

Remote sensing monitoring of cultivated land occupancy in Guizhou province based on deep learning technology

WANG Honglei, YAN Wenpu

2024, 0(3):  134-139.  doi:10.13474/j.cnki.11-2246.2024.0323

Asbtract ( )   PDF (6520KB) ( )  

References | Related Articles | Metrics

Guizhou province is confronted with a relative deficiency of arable land resources, limited scope for expanding its cultivated areas,posing a challenge to its regional agricultural production and food security. Timely monitoring of land occupation is of great significance for farmland protection and loss reduction. Remote sensing technology can play an important role in monitoring cropland occupation, however, due to the complexity of the surface structure, high-precision monitoring of cropland occupation faces greater difficulties. In order to improve the monitoring accuracy, this paper investigates the use of deep learning technology to monitor the cultivated land occupation in Guizhou province. Firstly,multi-type and high-frequency high-resolution satellite images are utilized to obtain a large number of samples in the whole area of Guizhou province, according to which the information of cultivated land occupation in remote sensing images is mined. Then,convolutional neural network and recurrent neural network are jointly used to construct a deep learning network model for monitoring cropland changes, and cropland changes are extracted from the spectral, spatial and temporal phase information of remote sensing images. Finally,typical areas are selected to verify the accuracy of the monitoring results. The results show that the method can quickly monitor the areas of occupied cropland in Guizhou province, and provide decision-making references and regulatory tools for relevant departments.

Application of high-precision BDS+GPS deformation monitoring technology in geological hazard monitoring of UHV transmission lines

LI Long, MIAO Chengguang, LI Xiang, ZHANG Peng, YUE Lingping

2024, 0(3):  140-144.  doi:10.13474/j.cnki.11-2246.2024.0324

Asbtract ( )   PDF (1685KB) ( )  

References | Related Articles | Metrics

BDS+GPS deformation monitoring technology is of great significance for high precision monitoring of UHV transmission towers. On the basis of building the BDS+GPS high-precision positioning mathematical model, according to the consistency of the epoch before and after, the paper gives a robust GNSS observation data preprocessing strategy. Considering the high accuracy and fast response of the monitoring algorithm, the real-time sliding window solution strategy is used to monitor the displacement changes of high-voltage transmission towers in real time. BDS+GPS deformation monitoring software is developed with B/C architecture, which supports data acquisition, data analysis, data solution and front-end display. The precision guide rail test proves that the horizontal RMS of the algorithm is less than 3mm, and the elevation RMS is less than 5mm, meeting the deformation monitoring requirements of centimeter level and millimeter level.

Research and application of multi-scale population spatial big data aggregation model in map visualization

LI Yayun, XIN Jing, CONG Jing

2024, 0(3):  145-150.  doi:10.13474/j.cnki.11-2246.2024.0325

Asbtract ( )   PDF (4429KB) ( )  

References | Related Articles | Metrics

In order to solve the technical problems such as long rendering time of maps and unresponsive pages caused by massive population and building data, a multi-scale population spatial big data aggregation model is proposed in this paper.By studying the hierarchical nested tree logical structure of massive data and spatial relationship within the viewable area, the statistical analysis and visualization of population spatial data in the multi-scale administrative areas of megacities are realized. It has been well applied in multi-source heterogeneous attribute fusion, which better meets the needs of fine management of megacities.

Research and practice on ultra long distance cross river elevation measurement technology Based on BeiDou

LIANG Yong, LIU Chaosong

2024, 0(3):  151-155,178.  doi:10.13474/j.cnki.11-2246.2024.0326

Asbtract ( )   PDF (1930KB) ( )  

References | Related Articles | Metrics

At present, the frequently used methods for cross-river elevation measurement include leveling, ranging trigonometric elevation, and GNSS cross-river elevation measurement. Conventional leveling has a relatively short sight distance, and ranging trigonometric elevation measurement requires very high site conditions, requiring mutual visibility and synchronous alignment observation between measurement points. The measurement accuracy is greatly affected by wind, atmospheric refraction, and distance. As the measurement distance increases, the influence of vertical deviation also increases, and the accuracy of trigonometric leveling has decreased. When the span exceeds 3500m, there is no relevant standard specification to follow, which is a considerable challenge for cross-river elevation transfer measurement with a span of more than 10km. This paper studies the use of measurement method for elevation measurement of ultra-long-distance cross-river projects. The basic principle of BeiDou cross-river leveling method is firstly introduced in the paper, taking the elevation transmission of Chongming-Taicang river-crossing tunnel of Shanghai-Chongqing-Chengdu high-speed railway as an example, the layout scheme of leveling control network, field measurement quality control, data processing, and accuracy analysis are elaborated during the implementation of cross-river leveling measurement. The practical results show that: In the survey area with relatively flat terrain and relatively gentle change rate of elevation anomaly, the control network is optimized by laying compulsory observation pier, and the two-line cross-river leveling method is adopted. All the accuracy indexes attached inside and outside meet the requirements of second-order leveling. The 10.2km ultra-long-distance cross-river leveling technology has been successfully applied to the second-class level transmission in Chongming-Taicang cross-river tunnel of Shanghai-Chongqing-Chengdu high-speed railway, which has very good reference significance for similar engineering projects.

Research and application of the ground-underground integration in linear engineering site selection demonstration

LIU Chao, Lü Sanhe, ZHOU Shengchuan, QIAO Xin, LIU Pengchao

2024, 0(3):  156-161.  doi:10.13474/j.cnki.11-2246.2024.0327

Asbtract ( )   PDF (11088KB) ( )  

References | Related Articles | Metrics

In this paper, we aim to address the challenge of quickly grasping the natural conditions along linear projects during the planning and demonstration stages. This difficulty arises due to the long routes, wide ranges, and complex natural environments involved. To overcome this, we utilize 3D real scene data as the above-ground spatial base, 3D geological model data as the underground spatial base, and thematic data as the data element base. Through this approach, we construct a 3D spatial scene that seamlessly integrates both above-ground and underground aspects.Subsequently,we have developed a process for conducting 3D site selection demonstration analysis in linear engineering based on this 3D scene.To illustrate the effectiveness of our approach, consider the proposed construction of expressways in Qingdao as a case study. This analysis can quickly understand the terrain and topography, actual situation, and geological fault structures along the line, significantly enhance the efficiency and quality of site selection for linear engineering projects. Overall, our work presents a novel model for conducting site selection demonstrations in three-dimensional spatial scenarios for linear engineering projects.

Geological hazard monitoring based on the perspective of innovative smarter city: a case study of Guangzhou

LIU Guochao, PENG Weiping, LIU Wei

2024, 0(3):  162-167.  doi:10.13474/j.cnki.11-2246.2024.0328

Asbtract ( )   PDF (8159KB) ( )  

References | Related Articles | Metrics

The construction of the innovative smarter city provides new opportunities and new ideas for the geological disaster monitoring in Guangzhou. Combining the scientific research and engineering practice of the geological disaster monitoring in recent years, the author briefly discusses some thoughts on the geological disaster monitoring in Guangzhou from the perspectives of perception, dispatch and co-governance. In terms of perception: the geomorphic structure of Guangzhou is complex, and the northern hilly platform is a disaster-prone area. It is suggested to construct a three-inspection system of sky-air-ground in key areas to identify potential risk sources in time. Early warning and dispatching: The occurrence of ground disasters in Guangzhou has a strong temporal correlation and spatial coupling with rainfall. It is suggested to construct a refined meteorological early warning grid in the rain nest and rainfall concentration area to improve the accuracy of monitoring and early warning; In view of the phenomenon of heavy monitoring and light warning in geological disaster monitoring and warning, it is suggested to build a data-knowledge-driven early warning model to achieve high-precision and interpretable disaster prediction modeling; For the difficult slope, an expert research and judgment system should be established to achieve accurate "pulse" and accurate "treatment". Joint construction and governance: strengthen cross-departmental and cross-level cooperation, strengthen the assessment of new construction projects, reduce the stock of local disasters, and control the increase of local disasters.

Practical exploration on the deep integration of innovation entrepreneurship training and smart city curriculum

LI Shaoying, ZHANG Xinchang, WU Zhifeng, CHEN Chengjing, RUAN Yongjian

2024, 0(3):  168-172.  doi:10.13474/j.cnki.11-2246.2024.0329

Asbtract ( )   PDF (1308KB) ( )  

References | Related Articles | Metrics

The innovation and entrepreneurship training program is an educational concept and practice formed to adapt to the development of economic society and higher education. How to embed it into the teaching process to cultivate innovative talents is an important topic of colleges' teaching reform. With the increasing intelligence of urban management, this paper puts forward the teaching reform framework of the integration of innovation and entrepreneurship training plan and smart city curriculum. Guided by the output of "Internet+" innovation and entrepreneurship training, students are guided to find the pains of smart construction and put forward feasible solutions in combination with smart city technology and cases, which effectively cultivated students' innovative thinking and stimulated their entrepreneurial enthusiasm, and provided useful references for the organic integration of curriculum teaching and students' innovation and entrepreneurship training.

UAV image road crack detection method based on improved YOLOv5

ZHU Weigang, WANG Lun, CHEN Tian, ZOU Bowen

2024, 0(3):  173-178.  doi:10.13474/j.cnki.11-2246.2024.0330

Asbtract ( )   PDF (2017KB) ( )  

References | Related Articles | Metrics

The emergence of road cracks has an obvious impact on the road service life and the safety of people and vehicles,so it is necessary to detect road cracks in time.Aiming at the problems of low detection accuracy caused by small crack target and complex image background in UAV image,this paper takes the crack image collected by UAV as research data,and proposes a deep learning road crack detection method based on improved YOLOv5 model.The attention mechanisms of CBAM,SimAM and CA are added to the backbone network of YOLOv5 model to improve the crack recognition ability and detection accuracy of the model.Comparative analysis is carried out through ablation experiments.At the same time,adaptive spatial feature fusion algorithm is incorporated into the YOLOv5 model to improve the ability of crack feature extraction.The research shows that the accuracy of the improved YOLOv5 network model is significantly higher than that of the original model,and the mean average accuracy(mAP) is increased by 20.6%.It not only ensures the accuracy but also effectively improves the detection accuracy,and can provide a new method for road crack detection.

Application of 3D laser scanning technology in monitoring bridge deformation

HU Yang, HAN Yang, LIANG Wenpeng

2024, 0(3):  179-182.  doi:10.13474/j.cnki.11-2246.2024.0331

Asbtract ( )   PDF (2393KB) ( )  

References | Related Articles | Metrics

With the acceleration of China's economy, the construction of urban transportation facilities is constantly increasing. As an important part of urban transportation, bridges are also constantly updating and improving the technology in the aspects of construction and testing. After long-term operation due to the uneven load and natural uncontrollable external force, bridge has local deformation, cracks and other diseases. Because the bridge across the water, the influence of traffic and natural factors, conventional GNSS receiver, all station such as measuring instrument is difficult to quickly and accurate obtain bridge full space information data, and 3D laser scanning due to its high operation efficiency, non-contact has the advantages of traditional measuring equipment.Taking Runyang bridge in Jiangsu province as an example, this paper discusses the application of 3D laser scanning technology in bridge deformation monitoring, and introduces the steps of field scanning, point cloud splicing denoising, achievement map drawing, and deformation analysis in detail, which provide reference experience for the application of new means of bridge monitoring.