Table of Content

25 July 2024, Volume 0 Issue 7

Previous Issue    Next Issue

A millimeter-wave radar odometry autonomous localization technique

LÜ Xuanfan, LIU Jingbin, MAO Jingfeng

2024, 0(7):  1-5.  doi:10.13474/j.cnki.11-2246.2024.0701

Asbtract ( )   HTML ( )   PDF (2641KB) ( )  

References | Related Articles | Metrics

By studying the fundamental characteristics of millimeter-wave radar data,we propose a millimeter-wave radar odometry autonomous localization technique.Our approach extracts AKAZE feature points from millimeter-wave radar images through feature matching.And we introduce three matching constraint conditions based on the geometric priors inherent in radar images,which significantly reduces feature point mismatches,ensuring the reliability of sensor pose and odometry calculation.Experimental results on the Oxford radar robot dataset show that our method achieves a relative positioning error of 6.8% for horizontal displacement and 0.89 degrees per meter for heading angle without the local mapping and loop closure detection optimization strategies,and the single-frame positioning time is only 0.081 s,which provides a new idea for indoor and outdoor navigation and positioning in harsh environments.

Performance analysis of BDS+Galileo single-epoch multi-frequency precision point positioning

XU Shengyi, GUO Jing, ZHAO Qile, LI Zhen, QIAO Lingna

2024, 0(7):  6-11,94.  doi:10.13474/j.cnki.11-2246.2024.0702

Asbtract ( )   HTML ( )   PDF (1775KB) ( )  

References | Related Articles | Metrics

Precise point positioning(PPP) typically requires tens of minutes to achieve centimeter-level positioning accuracy. However,the full utilization of multi-frequency and multi-constellation observations can accelerate PPP convergence. This study analyzes the BDS+Galileo multi-frequency single-epoch PPP performance based on the data from Australian regional reference stations. The results indicate that the extra wide-lane(EWL) and wide-lane(WL) ambiguity based on multi-frequency observations can improve the resolution of narrow-lane(NL) ambiguities significantly. Compared to dual-frequency PPP with ambiguity resolution(PPP-AR),the instantaneous ambiguity fixing rate increases significantly from 27.6% to 69.2% for triple-frequency PPP-AR. And it further increases to 78.9% and 76.9% for quad-and five-frequency solutions. The instantaneous positioning accuracy in the east,north,and up directions for BDS+Galileo quad-frequency solution can achieve 7.1,7.8 and 23.9 cm,respectively.

An integrated learning localization method fusing 5G CSI and geomagnetic data

CHENG Zhenhao, ZHAO Dongqing, GUO Wenzhuo, LAI Luguang, LI Linyang

2024, 0(7):  12-16.  doi:10.13474/j.cnki.11-2246.2024.0703

Asbtract ( )   HTML ( )   PDF (2345KB) ( )  

References | Related Articles | Metrics

A multi-input convolutional neural network (CNN) incorporating bidirectional long short-term memory neural network(BiLSTM) and attention mechanism is presented to address the issues of local convergence and poor heterogeneous fusion performance of deep learning algorithms in multi-sensor fusion positioning. Firstly, 5G channel state information (CSI) and geomagnetic data arepreprocessed separately. Then each of them is trained offline based on an independent branch network, and the spatial and temporal features of the fingerprint data are extracted at the sametime to append the attention mechanismlayer. Finally, the fusion of heterogeneoussensor data for localization is achieved at the fully connected layer.The experimental results in the conference room and the teaching building hall show that the average positioning error is 0.95 and 1.84 m respectively, which is 48.9% and 42.7% higher than that of the error backpropagation network(BPNN), and that positioning accuracy and system stability are both greatly improved.

An indoor positioning method based on geomagnetic sequence-assisted correction for PSO-PF

HE Zhengwei, SUN Bingyuan

2024, 0(7):  17-23.  doi:10.13474/j.cnki.11-2246.2024.0704

Asbtract ( )   PDF (1958KB) ( )  

References | Related Articles | Metrics

Indoor pedestrian positioning serves as a crucial foundation for location-based services, with geomagnetic signals being continuously perceived, making them a focal point in indoor pedestrian positioning research. Addressing the challenges of significant cumulative errors and low positioning accuracy in current particle filter fusion positioning, this paper proposes an indoor pedestrian positioning method: variable-length geomagnetic sequence assisted PSO-PF. Building upon traditional particle filter algorithms, this method integrates particle swarm optimization for optimal position to enhance real-time positioning accuracy. Subsequently, a DTW-A^* algorithm is established to correct cumulative errors over time for variable-length geomagnetic sequences, address the cumulative error problem associated with particle filter-based positioning methods. Experimental comparisons with existing mainstream positioning methods demonstrate that the proposed method achieves an average error of 0.90 m in indoor pedestrian positioning. The average error is reduced by 73.1%, 68.0%, and 63.8% compared to PDR, MaLoc, and Magicol methods, respectively. Notably, the proposed method achieves a 1.43 m positioning accuracy at 90%, showing a relative improvement of 75.1%, 68.4%, and 67.7% compared to PDR, MaLoc, and Magicol methods, respectively. Furthermore, experimental results on different models of smart phones indicate that the proposed research method is not only applicable,but also stable, offering potential support for indoor positioning across various devices.

Combined GNSS+5G positioning based on adaptive optimal selection-anti-differential adaptive Kalman filter hybrid model

HU Xiangxiang, SONG Bao, SHI Yaya, PANG Dongdong, WU Chengyong, ZHANG Lili, LI Yifei

2024, 0(7):  24-29.  doi:10.13474/j.cnki.11-2246.2024.0705

Asbtract ( )   PDF (3168KB) ( )  

References | Related Articles | Metrics

The construction of the PNT system is a crucial research topic in communication and navigation. Developing a comprehensive PNT system compatible with different types of PNT means and providing better flexibility and environmental adaptability has become an important topic that can be completed on time. The emergence and development of the 5G and BeiDou systems provide new ideas and guidance for the PNT system to be more comprehensive and flexible. Accordingly, this paper proposes an AOS-RAKF method based on the combined GNSS+5G data to realize high-precision localization estimation in urban complex environments. The algorithm mainly consists of two main modules. They are AOS-based 5G base station measurement data optimization and combined GNSS+5G positioning based on the AOS-RAKF algorithm. The AOS-based 5G base station measurement data optimization module achieves better observation data re-selection using adaptive optimal selection factors. The combined GNSS+5G positioning utilizes the optimized 5G data and GNSS to establish a coupled structural model to achieve high-accuracy localization of the mobile vehicle using the RAKF method. The results of semi-physical simulation tests show that the proposed AOS-RAKF method significantly improves the positioning accuracy in complex urban environments compared to GNSS, single 5G, and traditional combined GNSS+5G positioning using raw measurement data.

Spillway automatic extraction based on GF-7 satellite data

CHEN Jia, ZHANG Wen, LI Junjie, YANG Zhiwen, MENG Lingkui, LI Linyi

2024, 0(7):  30-34.  doi:10.13474/j.cnki.11-2246.2024.0706

Asbtract ( )   HTML ( )   PDF (5256KB) ( )  

References | Related Articles | Metrics

Spillways are important hydraulic structures. Intelligent detection of spillways in remote sensing images are of great scientific significance and application values. However, due to the small scale and intricate surroundings of spillways, it is difficult to automatically detect spillways in remote sensing images. In this study, the domestically developed high-resolution imagery from the GF-7 satellite is used, which has stereoscopic imaging capability. The Spillway Geometry Stereoscopic Index (SGSI) is constructed and an automatic spillway recognition method is proposed. Firstly, data preprocessing is conducted to obtain fused imagery and high-precision DSM. Next, the Mean-Shift algorithm is employed for segmenting the fused imagery, extracting multi-dimensional features of segmented objects, and determining target objects through multi-feature joint decision-making. Finally, integrated and post-processing of spillway patches are performed based on the spatial context relationship of target objects to output the final recognition results, achieves automatic recognition of spillways. The proposed method is validated on three dam reservoirs containing spillways, and experimental results show good matching between the extracted spillway boundaries and reference boundaries, with an overall accuracy of 89.23%. The proposed method is proved to be efficient and accurate for automatic spillway recognition.

Analysis of surface movement characteristics and seasonal annual variation of Petermann Glacier in Greenland

TIAN Kang, WANG Zhiyong, LI Zhenjin, ZHANG Baojing, SUN Shichang

2024, 0(7):  35-40,54.  doi:10.13474/j.cnki.11-2246.2024.0707

Asbtract ( )   PDF (9745KB) ( )  

References | Related Articles | Metrics

The Greenland ice sheet, which accounts for 10% of the world's total ice, is surrounded by many overflowing glaciers, and melting glaciers will cause more ice to be pumped into the sea. Based on the data of 96 Sentinel-1A, the offset tracking algorithm is used to obtain the surface flow velocity of Petermann Glacier in 2019—2022, and the characteristics of glacier surface movement, seasonal and annual variations were analyzed. The results show that: ①The surface flow velocity of Petermann Glacier shows obvious seasonal variations, with high flow velocity in summer and low flow velocity in spring, autumn and winter, and the flow velocity of the glacier profile line is significantly faster than that on both sides. ②In 2019—2022, the average flow velocity of the main glacier area is stable, and the change trend of the upstream part is consistent, and the ice flow velocity of the main part of the glacier in 2021 and 2022 was faster than that in 2019 and 2020, and the high flow rate lasted longer. ③Temperature, precipitation and wind speed promote the movement of glacier surface velocity, while sea level pressure inhibited glacier velocity. From the perspective of correlation, the effect of temperature is the most significant, followed by wind speed, and the effect of precipitation is the least.

High-precision terrain-based on-orbit geometric calibration method for laser altimeter

ZHANG Hao, XU Qi, HUANG Peiqi, CHEN Gang, XIE Huan

2024, 0(7):  41-47.  doi:10.13474/j.cnki.11-2246.2024.0708

Asbtract ( )   HTML ( )   PDF (5064KB) ( )  

References | Related Articles | Metrics

The digital calibration method for laser altimeters has attracted increasing attention due to its prominent advantages of high calibration frequency and low consumption of manpower and resources. This paper proposes a “two-step” digital calibration method for laser altimeters, focusing on the geometric positioning model of linear regime laser altimeters. A geometric calibration model concerning roll angle, pitch angle, and range system error is constructed, and calibration tests are conducted using publicly available reference terrain data SRTM and high-precision digital calibration fields. Comparing and analyzing with the actual field detector experimental calibration data of the GF-14. The results show that the geometric calibration method proposed in this paper has an angle difference within 0.5 arcseconds and a ranging difference within 0.2 m compared to the ground detector calibration parameters. By independently validating the elevation accuracy of the calibrated results, both the proposed digital calibration method and detector calibration method exhibit good consistency in footprint elevation bias. The RMSE between the footprint elevations obtained from both methods and the high-precision airborne LiDAR point cloud elevations is less than 0.3 m. The method proposed in this paper is not only ensures accuracy, but also reduces the consumption of human and material resources, providing a new approach and method for on-orbit normalization calibration of spaceborne laser altimeters.

Road surface points extraction from vehicle LiDAR point cloud based on radial gradient

MAO Jingyi, WANG Jingxue, DONG Xiao

2024, 0(7):  48-54.  doi:10.13474/j.cnki.11-2246.2024.0709

Asbtract ( )   PDF (1679KB) ( )  

References | Related Articles | Metrics

To address the challenges of road surface extraction from large-scale vehicle LiDAR point clouds, including difficulties caused by obstacles occluding the road surface and variations in road topography leading to inaccuracies and noise in the extraction results, this paper introduces a vehicle LiDAR point cloud road surface extraction method based on radial gradients. The method involves initial data preprocessing, including pass-through filtering and voxel downsampling. Subsequently, the point cloud data is transformed into polar coordinates,which is projected onto a fan-shaped grid based on the hardware parameters of the vehicle's LiDAR system. Using the fan-shaped grid as a foundation, a radial gradient road surface point extraction is performed within a moving window, and a plane fitting technique is employed, optimized using the least-squares method to refine the extraction results. The KITTI dataset is chosen for road surface point extraction experiments. Results indicate that compared to other road surface extraction methods this approach exhibits robustness and high accuracy, with an average accuracy of 91.85%, an average completeness of 80.63%, and an average precision of 75.25% in road surface point extraction.

Backpack LiDAR damage pavement deformation detection method

WU Jingdong, CAI Lailiang, ZHANG Bingjie, WANG Xin

2024, 0(7):  55-59,82.  doi:10.13474/j.cnki.11-2246.2024.0710

Asbtract ( )   PDF (2661KB) ( )  

References | Related Articles | Metrics

In order to efficiently monitor the road damage phenomenon, this paper adopts backpack 3D laser scanning technology to observe the road surface, and proposes a method to detect the deformation of damaged road surface. For the stretching type of damage, a point cloud damaged pavement extraction algorithm based on the mean value of the colour field is established. Firstly, a two-dimensional grid is constructed for the initial extraction of damage, and then the weighted mean value of the RGB colour gamut of the point cloud is calculated, and the point cloud of road damage is obtained by fine screening through the differentiation of the ratio; for the extrusion-type damage, the normal vector pinch angle standard deviation method is used. Then, using Euclidean clustering, the classified tensile and extruded damages are calculated as length, width, damage interval, height, and finally the pavement inclination. A comprehensive pavement damage evaluation model is established through the 5D information. The experimental results show that the accuracy of tensile damage extraction reaches 96.4%, and the accuracy of extrusion damage extraction reaches 100%, and the comprehensive evaluation model of pavement damage indicates that the road is a third-class damage, which needs to be repaired. It basically meets the required accuracy of road damage extraction, and provides technical support for the formulation of road traffic control and repair measures.

Geological disaster identification and monitoring along the Ordos platform pipeline based on SBAS-InSAR

HUANG Xitao, HU Zhifeng, QIAO Pei, ZHANG Yu, WANG Xinshuang

2024, 0(7):  60-64.  doi:10.13474/j.cnki.11-2246.2024.0711

Asbtract ( )   PDF (5317KB) ( )  

References | Related Articles | Metrics

The geological environment of Ordos platform is fragile, and geological disasters such as landslide, collapse and debris flow occur frequently, which pose a serious threat to people's production and life and economic construction. Based on 76 Sentinel-1 images of a local area of the Ordos platform from January 2021 to August 2022, combined with SBAS-InSAR and CR-InSAR technologies, this study analyzes the abnormal surface deformation trends in the study area through time series deformation information and verifies the accuracy. By integrating the interpretation of geological hazard points from GF-2 satellite remote sensing images and conducting field validations, a comprehensive analysis is performed on two identified typical hazard points. The results show that this technology can achieve millimeter-level accuracy in surface deformation. According to the 《Geological Disaster InSAR Monitoring Technology Guide》(T/CAGHP 013—2018), the accuracy has been upgraded from Level II to Level I. During the monitoring period, 83% of the study area's deformation rate was within -20 to 20 mm/a, indicating a stable state, while the remaining 17% had potential hazards. A total of 41 geological hazard points were interpreted, and with field verification, the accuracy rate was 73.2%. Thirty geological hazard points were identified, including 21 landslide hazards, 4 collapse hazards, and 5 unstable slopes. Two typical landslide hazard points showed significant abnormal deformation, with deformation rates ranging from -21 to 21 mm/a and maximum cumulative deformation per point from -25 to 25 mm. These areas are significantly affected by climate, topography, and human activities, posing high-risk hazards. It is essential to strengthen manual inspections, increase monitoring efforts and frequency, and early identification of geological disasters to ensure the safety of engineering construction and other works.

Agricultural greenhouse information extraction based on SFNet-F land feature recognition technology

FU Lizhao, YANG Qinggang, CHEN Yongli, HAN Jinting, YANG Xinjia

2024, 0(7):  65-70.  doi:10.13474/j.cnki.11-2246.2024.0712

Asbtract ( )   PDF (13331KB) ( )  

References | Related Articles | Metrics

Agricultural greenhouses are an essential component of modern agricultural facilities, and accurate identification and dynamic monitoring of their distribution provide a reliable scientific basis for the government to carry out agricultural subsidy accounting and agricultural production decision-making. In this paper SFNet-F image processing technology is proposed to address the issue of low accuracy in traditional image recognition methods. By collecting agricultural greenhouse datasets of different types, periods, and regions, the FixMatch semi-supervised learning module is combined with SFNet to improve the efficiency and quality of sample library establishment, reduce costs, and achieve high-precision semi-supervised adaptive segmentation. In order to evaluate the feasibility of this method,multiple accuracy evaluation indicators were selected for accuracy validation in Pingquan city,Hebei province,and compared with U-Net,HBRNet,and DeepLabV3+. The results show that the deep learning model based on the SFNet-F embedded SuperMap platform can identify agricultural greenhouses on a large scale quickly and accurately. The recognition effect is the best in all accuracy indicators compared to several popular methods.

Study on NPP dynamic change characteristics and driving factors response in mining area of Shanxi province:taking Xuangang mining area as an example

WANG Wenwen, LI Jiafang, YANG Wenfu

2024, 0(7):  71-76,163.  doi:10.13474/j.cnki.11-2246.2024.0713

Asbtract ( )   PDF (2972KB) ( )  

References | Related Articles | Metrics

Over the years, the development of Shanxi province has been mainly driven by coal resources, but the ecological environment has been destroyed. The net primary productivity of vegetation (NPP) is an important index to measure the regional ecological environment. The study of NPP of mining area vegetation is of great significance for regional ecological restoration, sustainable development and high-quality development. Taking Xuangang mining area in Shanxi province as an example, this study used Landsat, MOD17A3H, DEM, meteorological and population data, and integrated CASA model, monadic linear regression, Hurst index and geographical detector methods to reveal the spatio-temporal variation characteristics of NPP in the mining area, and analyzed its driving factors. The results show that: ①In terms of time scale, the NPP high value area of vegetation in Xuangang mining area shows a significant upward trend from 2000 to 2021, and the average annual value is 370.79 gC/(m²·a); ②In spatial scale, NPP of vegetation in the mining area has significant heterogeneity, showing a gradual decrease from southwest to northeast, with the median area (53.63%)、high value area (41.99%)、low value area (4.38%),from the largest to the smallest; ③In terms of changing trend, NPP of vegetation in mining area is mainly increasing, and will continue to intensify in the future, and the vegetation growth environment still faces great challenges; ④From the perspective of driving factors, the NPP of vegetation in mining area is affected by land use (0.109)、elevation (0.088)、population density (0.075)、slope (0.042)、precipitation (0.020)、air temperature (0.014)、slope direction (0.042),from the largest to the smallest.

Direct correction model for UWB coordinate error based on artificial neural network

WANG Yifan, LI Zengke, JIANG Shizheng, CHEN Yuan, HUANG Linchao, JI Liya, DENG Weifang

2024, 0(7):  77-82.  doi:10.13474/j.cnki.11-2246.2024.0714

Asbtract ( )   PDF (5305KB) ( )  

References | Related Articles | Metrics

The one-step ultra-wideband (UWB) coordinate error correction models based on the generalized regression neural network (GRNN) and back-propagation neural network (BPNN) were proposed to address the difficulty of correcting the coordinate error existing in UWB positioning based on conventional means. The correction models took the UWB original positioning coordinates,the distance between it and different base stations as inputs,and the UWB relative high-precision reference value error as output. The correction models were trained with GNSS RTK point coordinates as the dynamic experimental reference values and total station point coordinates as the static experimental reference values,respectively. Besides,the correction models were employed to correct the UWB coordinates of the non-modeled sample points. Then a comparative analysis of the accuracies before and after correction and the accuracies of the different correction models was conducted. The results show that the method of using artificial neural networks to construct the one-step UWB coordinate error correction models is feasible and it is easier and faster without the need to solve the coordinates using the corrected distance. The correction models can effectively improve the dynamic and static positioning coordinate accuracy of UWB overall. Among them,the correction performance of the GRNN-based correction model is the most significant. Moreover,the GRNN-based correction model can correct the UWB coordinate error more effectively than the BPNN-based correction model. The accuracy of the corrected UWB dynamic positioning planar coordinate can reach the centimeter level,and the accuracy of the static positioning planar coordinate is as high as the millimeter level.

Experimental research and application of rapid measurement method of CPⅢ plane network on main beam of five-tower continuous cable-stayed bridge based on GNSS

LI Kunpeng, LI Qilin, YANG Xuefeng, SUN Hongbin, QU Shufeng

2024, 0(7):  83-87.  doi:10.13474/j.cnki.11-2246.2024.0715

Asbtract ( )   PDF (1598KB) ( )  

References | Related Articles | Metrics

The long continuous beam of the long-span bridge will produce large longitudinal and vertical displacement with the change of temperature, resulting in the multi-value of the CPⅢ coordinates on the continuous beam bridge, and the station accuracy cannot meet the relevant accuracy requirements of the high-speed railway track plate measurement. This paper proposes a new method for rapid measurement of CPⅢ point plane coordinates on continuous beams based on GNSS static measurement, and a practical measurement experiment is designed based on the Changqing Yellow River Bridge of Zhengzhou-Jinan high-speed Railway to verify the feasibility of the new method. The experimental results show that this method for rapid measurement of CPⅢ point plane coordinates based on GNSS. In the case of observation duration of 30 min, it can meet the requirements of CPⅢ planar network measurement accuracy in the construction of base plate and track plate.

GNSS-IR soil moisture inversion integrating isolated forest and deep learning

YANG Xiaofeng, WEI Haohan, ZHANG Qiang, XIANG Yunfei

2024, 0(7):  88-94.  doi:10.13474/j.cnki.11-2246.2024.0716

Asbtract ( )   PDF (6973KB) ( )  

References | Related Articles | Metrics

Aiming at the problems of uneven quality,poor reliability and unstable model inversion results of single characteristic parameter data in GNSS reflected signal remote sensing,this paper proposes a GNSS-IR soil moisture inversion method that combines isolated forest and deep learning. The experimental results show that the frequency characteristic parameters of GNSS SNR are not suitable for the inversion of soil moisture,while the amplitude and phase characteristic parameters are highly correlated with soil moisture,which can be used for the inversion of soil moisture. The inversion results of the fusion amplitude and phase characteristic parameters of the three deep learning models of CNN,DBN and GRU are in good agreement with the measured soil moisture. Compared with the single feature parameter inversion method using only amplitude or phase,the inversion accuracy of the proposed method is improved by 21.4%～55.8%,and the correlation coefficient is improved by 4%～9.1%.

UWB indoor localization algorithm based on KF-LSTM

TIAN Yalin, LIAN Zengzeng, WANG Penghui, WANG Mengqi, LU Li

2024, 0(7):  95-99,151.  doi:10.13474/j.cnki.11-2246.2024.0717

Asbtract ( )   PDF (3394KB) ( )  

References | Related Articles | Metrics

As a new wireless localization technology, UWB has attracted much attention in the field of indoor localization.In order to improve localization accuracy in ultra-wide band, this paper combines the advantages of Kalman filtering and LSTM networks and proposes a long short-term memory neural network (KF-LSTM) algorithm that incorporates Kalman filtering. Firstly, the UWB timing data is processed by Kalman filtering to weaken the Gaussian white noise in the data, and then the data is put into the LSTM network for training, which takes advantage of the LSTM network's processing of timing features to deal with the non-Gaussian noise and then obtains a more accurate label location.The final measured data show that the average localization accuracy of the KF-LSTM algorithm is improved by 70.21%,37.28% and 38.23% compared to the BP, KF-BP and LSTM network algorithms respectively, and the KF-LSTM algorithm performs more stably.

Rapid detection method of abnormal roundness area of large diameter steel cylinder

LÜ Weirong, WU Jiaqiang, YAO Shuai, QI Jingjing, LU Beirong, DING Shibao, GAN Dejia

2024, 0(7):  100-104,128.  doi:10.13474/j.cnki.11-2246.2024.0718

Asbtract ( )   PDF (1322KB) ( )  

References | Related Articles | Metrics

Large diameter steel cylinders may have one or more roundness anomalies for some reason, during fabrication, transport, installation and operation, which don't not meet the relevant requirements for equipment acceptance and operation. At present, the detection methods for large diameter cylinders are complex and not suitable for rapid detection. In this paper, based on the principle of coordinate measurement, combined with the non-contact automatic measurement mode of high-precision total station, an automatic search procedure is prepared by Matlab program, and a rapid detection method of large diameter steel cylinder roundness anomaly area is proposed. The simulation by AutoCAD and the results of field measurements show that the inspection method is not only fast, but also can significantly improve the detection accuracy by increasing the density of measurement points, providing accurate data for subsequent repair and improvement work.

YOLOv7 shield tunnel water leakage detection method based on CutMix data augmentation and multi-constraint loss function

GAO Xianjun, LIU Zhenyu, XU Lei, HUANG Yifan, TAN Meilin, XIONG Wenhao, YANG Yuanwei

2024, 0(7):  105-110.  doi:10.13474/j.cnki.11-2246.2024.0719

Asbtract ( )   PDF (6011KB) ( )  

References | Related Articles | Metrics

Since the size of leakage water in the intensity projection image of the shield tunnel is inconsistent and the proportion of pixels is limited, the learning ability of key features of object detection models is weak. As a result, the detection accuracy of leakage disease targets is too low to meet the requirements of the application. Therefore, an improved YOLOv7 shield tunnel leakage water detection method based on CutMix data enhancement and multi-constraint loss function is proposed to address the issue in this paper. Firstly, the tunnel images are enhanced using the embedded CutMix approach. Various training samples are randomly combined into new samples with comprehensive features. Secondly, the YOLOv7 network is employed as the skeletal structure, and an efficient channel attention module is introduced to enhance the ability of crucial leakage features to learn and express themselves autonomously. Finally, a loss function incorporating multi-constraint geometric conditions is designed to improve the accuracy of the geometric shape of the prediction box, thereby improving the model's predictive accuracy. The four algorithms included Fast R-CNN, SSD, YOLOv5, and YOLOv7 are chosen for comparison in complex environments with good lighting, poor lighting, and occlusion. The experiments show that our algorithm achieves a leakage detection accuracy of 85.90%. The average accuracy of the proposed method is higher than Fast R-CNN, SSD, YOLOv5, and YOLOv7 by 5.55%, 8.89%, 3.93%, and 2.75%, respectively. It exhibits good robustness and generalization ability.

Cotton growth monitoring combined with coefficient of variation method and machine learning model

YANG Sijia, WANG Renjun, ZHENG Jianghua, ZHAO Pengyu, HAN Wanqiang, MAO Xurui, FAN Hong

2024, 0(7):  111-116.  doi:10.13474/j.cnki.11-2246.2024.0720

Asbtract ( )   PDF (9221KB) ( )  

References | Related Articles | Metrics

In order to obtain the growth information of the key phenological period of cotton more accurately, the cotton planting area is extracted through the cotton mapping index. Secondly, five indexes, including plant height, SPAD value, leaf wet weight, leaf dry weight and leaf area, reflecting cotton growth, are constructed into a comprehensive growth Index, namely Flowering and boll cotton growth index (FBCGI), using the coefficient of variation method. Finally, the optimal characteristic variables are selected and the inverse model of cotton growth is constructed by combining with random forest model. The results showed that: ①The overall classification accuracy of cotton reached 81.65%.②Compared with the five single growth indicators, the constructed FBCGI had a higher correlation with vegetation index. ③The R² and RMSE of the cotton growth monitoring model based on the optimal characteristic variables and random forest model in the modeling set and validation set are 0.74, 0.07 and 0.51, 0.10, respectively. The results can provide important reference for cotton growth monitoring.

Ecological sensitivity analysis and assessment of Kubuqi desert based on GIS

DONG Jiaqi, YAN Min, ZUO Hejun, XI Cheng, YAN Yu

2024, 0(7):  117-122.  doi:10.13474/j.cnki.11-2246.2024.0721

Asbtract ( )   PDF (2679KB) ( )  

References | Related Articles | Metrics

Ecological sensitivity assessment is an important means and link of regional ecological environment protection and construction work. Taking Kubuqi desert as the research site, six evaluation factors including elevation, slope, slope direction, water buffer, vegetation coverage and land use type are selected to construct the ecological sensitivity assessment system of Kubuqi desert in the paper. The ecological sensitivity of Kubuqi desert is analyzed and evaluated by using analytic hierarchy process and GIS technology, and five sensitivity levels are divided. The results show that the overall ecological sensitivity of Kubuqi desert is low, mainly mildly sensitive. Vegetation coverage affects the most, followed by land use type. Insensitive, mildly sensitive, moderately sensitive, highly sensitive and extremely sensitive are 27.37%, 46.66%, 22.00%, 2.20% and 1.77% of the Kubuqi desert area, respectively.The areas with high sensitivity are mainly distributed in the northern end of Kubuqi desert, mainly lakes, rivers and vegetation reserves. The research results could provide some scientific suggestions for the regional ecological environment construction and protection work in the Kubuqi desert.

Application and discussion of real estate surveying based on unmanned aerial vehicle oblique photogrammetry technology

FU Huiwei, CEN Ming, LIAO Chaoming, LÜ Huaquan

2024, 0(7):  123-128.  doi:10.13474/j.cnki.11-2246.2024.0722

Asbtract ( )   PDF (3401KB) ( )  

References | Related Articles | Metrics

In response to the problems of long construction period, low work efficiency, and high cost in traditional real estate measurement field work, this article conducts research on real estate measurement technology based on unmanned aerial vehicle tilt photogrammetry technology. Taking the survey project of integrated real estate ownership in a rural area of Guangxi as an example, drone tilt photogrammetry technology is applied to the survey of real estate ownership. Traditional measurement methods are used to collect verification data on site and external accuracy analysis is conducted with the real estate measurement results obtained from drone tilt photogrammetry. The results indicate that the real estate measurement results obtained using unmanned aerial vehicle tilt photogrammetry technology can fully meet the first level accuracy requirements of the “Technical Rules for the Investigation of Real Estate Registration in Rural Areas of Guangxi(Trial)”, providing technical support for the smooth implementation of the rural real estate registration investigation project, and also providing beneficial technical references for personnel engaged in real estate surveying and mapping operations.

Locking degree and seismic risk analysis of yishu fault zone constrained by InSAR data

YAN Bingdun, YIN Haitao, YANG Yuyong, LIAN Kaixuan, WEN Liyuan, GUO Zongbin

2024, 0(7):  129-133.  doi:10.13474/j.cnki.11-2246.2024.0723

Asbtract ( )   PDF (11201KB) ( )  

References | Related Articles | Metrics

In this paper, the Sentinel-1A data covering 60 scenes of the Yishu fault zone from October 2022 to April 2023 are used to obtain large-scale surface LOS deformation results by SBAS-InSAR technology, and the InSAR deformation field is down-sampled.Then, based on the negative dislocation model, TDEFNODE software is used to calculate the fault locking degree and slip loss distribution of Yishu fault zone, and combined with the temporal and spatial distribution of earthquakes, the movement characteristics of fault plane are described in detail, and the segmental seismic risk of the fault is quantitatively determined. The results show that the fault locking degree is high and the locking depth is deep in the area north of Anqiu in the northern segment of Ju county fault, the maximum locking depth is about 26°, and the fault locking is weak in the segment from Anqiu to Tancheng, which is basically in creep slip state.The area north of Anqiu in Yishu fault zone has a dextral slip loss rate of 0.5～1.5 mm/a, which is the unruptured segment of 1668 Tancheng earthquake. It has the characteristics of seismicity gap and is easy to accumulate stress. The seismic risk of this segment deserves attention.Compared with the previous studies, it is found that the locking degree of the section north of Tancheng in Yishu fault zone is weakened, which is mainly related to the influence of Miyagi Mw9.0 earthquake on March 11,2011. The occurrence of Japan earthquake plays a certain role in relieving the westward subduction and compression of Pacific plate, resulting in the change of regional stress-strain field of Tanlu fault zone.

Time series InSAR deformation monitoring of Dongting Lake based on mini-stack technology

AI Hui, WU Wenhao, YIN Mu, ZHANG Tengxu, YU Li, GE Zixuan, CAO Rui

2024, 0(7):  134-139.  doi:10.13474/j.cnki.11-2246.2024.0724

Asbtract ( )   PDF (7435KB) ( )  

References | Related Articles | Metrics

The Dongting Lake region, as an important water source protection area and a key wetland ecological region, is facing dual threats of tectonic subsidence and excessive human development, which pose challenges to its sustainable development and ecological environment. Considering the relatively low overall deformation rate in the Dongting Lake region, this study utilized long-term Sentinel-1A satellite image data for time series InSAR processing to accurately determine the surface subsidence rate in the area. Additionally, to mitigate the impact of decorrelation noise, this study performed time-domain compression on the images, generating a compressed image stack with a higher signal to noise ratio, thereby achieving efficient time-series processing. In comparison with classical time-series InSAR techniques, this approach not only enhances processing efficiency but also increases the spatial density of coherent points by 100 times. The research findings reveal the presence of two distinct subsidence areas within the study region.

Soil moisture inversion in highland areas based on Sentinel-1 and Landsat 8 remote sensing data

WANG Xiaying, SHE Yulin, ZHANG Shuangcheng, XIA Yuanping, NIU Yufen

2024, 0(7):  140-146.  doi:10.13474/j.cnki.11-2246.2024.0725

Asbtract ( )   PDF (4887KB) ( )  

References | Related Articles | Metrics

Soil moisture is a crucial parameter for agricultural production, water resource management, global climate, and other related fields. Synthetic aperture radar (SAR) serves as a significant mean for acquiring soil moisture, among which, vegetation and surface roughness are two key influencing factors. Therefore, this paper focuses on several aspects. Firstly, using Sentinel-1 radar data to obtain overall backscattering coefficients and which of vegetation are separated utilizing three vegetation indices (NDVI, NDWI, MSAVI) deriving from Landsat 8 optical data using the water-cloud model. Subsequently, the Dobson model is employed in conjunction with the advanced integral equation model (AIEM) to establish a table of backscattering coefficients lacking surface roughness, determining optimal roughness parameters through a minimum-cost function. Finally, a least squares method is used to determine the coefficients of the empirical equation for soil moisture inversion. Experimental results demonstrate that in highland areas, the model's inversion results exhibit good consistency with ground-truth measurements. Among these, the use of the normalized difference water index (NDWI2) as input for the water-cloud model combined with optimal roughness inversion yields the best results, with a fitting coefficient of 0.840 2 and a root mean square error of 0.027 21 cm³/cm³.

BeiDou-INS precision measurement and tamping method for ordinary speed railway

FANG Bole, LIANG Yong, LOU Liangwei

2024, 0(7):  147-151.  doi:10.13474/j.cnki.11-2246.2024.0726

Asbtract ( )   PDF (2748KB) ( )  

References | Related Articles | Metrics

The skylight time for precise measurement and tamping of existing ordinary speed railway track is short, which requires high operation efficiency. The operation effect directly affects the track quality and passenger comfort. The use of the measurement system provided by large-scale road maintenance machinery for both measurement and tamping has poor rectification effects, and can only control the overall unevenness of the track. Using relative rail inspection instruments to measure data to guide large-scale road maintenance machinery in digital tamping has improved the tamping effect, but it is still unable to control absolute linear and long wave irregularities. In this paper, the principle of the BeiDou-inertial navigation system(INS) precision measurement and tamping method of the ordinary speed railway is described, and the actual line is measured using the BeiDou-INS track geometry state measuring instrument. Through the BeiDou data solution, integrated navigation fusion processing and track alignment fitting optimization, the track amount and track setting amount that can be used to guide the digital tamping of large road maintenance machinery are obtained. After the completion of the actual operation of precise measurement and tamping, statistical analysis is conducted on the improvement of the static TQI of the track. The results show that the BeiDou-INS precision measurement and tamping method for the ordinary speed railway has high operational efficiency and significant improvement in track quality.

Surveying method for the main construction of elliptical high-rise buildings

CUI Gang

2024, 0(7):  152-157,167.  doi:10.13474/j.cnki.11-2246.2024.0727

Asbtract ( )   PDF (1651KB) ( )  

References | Related Articles | Metrics

In the process of modern city construction, the number of high-rise buildings is increasing, the building appearance curve is rich and changeable,and the construction measurement of the main body of the building should also adapt to the new demand.At present, the traditional elliptic curve sampling method is generally used. Although there are few tracing points and simple calculation, there are many problems such as low curve fit, less rounded and beautiful elliptic curve and low measuring accuracy.This paper presents a method for measuring rectangular coordinates of elliptic curves.This method uses the biaxial symmetry of ellipse, combined with the principle of rectangular coordinate fixed point and curve determination of series key points, calculates the rectangular coordinates of series curve points by elliptic equation and coordinate calculation formula, and uses rectangular coordinates to depict series curve points, and does not require a large area of flat site, so that smooth and beautiful elliptic curves with high fit can be obtained. The measurement accuracy meets the requirements of the measurement specification.Combined with engineering examples, the effectiveness of the proposed method is verified, and it can be applied widely.

Multi-source POI fusion method based on composite feature rule library

WANG Qingshe, YANG Chuanshi, GUO Sihui

2024, 0(7):  158-163.  doi:10.13474/j.cnki.11-2246.2024.0728

Asbtract ( )   PDF (2805KB) ( )  

References | Related Articles | Metrics

Multi-source POI fusion refers to integrating POI datasets from different sources into a unified, accurate, and comprehensive database. POI data, as one of the crucial data resources for the National Geographic Information Public Service Platform (Tianditu), poses a key technological challenge in the construction of Tianditu—namely, how to effectively merge various POI data sources to continuously enhance the timeliness, accuracy, and richness of Tianditu's POI data. Research on methods for POI data fusion has garnered widespread attention in the fields of surveying and mapping, geographic information systems, as well as big data and artificial intelligence. Substantial progress has been made in this regard. However, due to the complexity and diversity of POI textual attributes, effectively determining the similarity of POI attributes remains a challenge in practical engineering applications. Given the urgent need for rapid updates to Tianditu's POI data and the shortcomings of current multiple-source POI data fusion methods, this paper proposes a multi-source POI fusion method based on a composite feature rule library. This method aims to provide technical support for the data updates and master database construction of Tianditu.

Research on the construction and hierarchical expression of 3D geo-name entity

LI Xiaoqiang, GAO Hua, WANG Ling, CAI Qi, YANG Chao

2024, 0(7):  164-167.  doi:10.13474/j.cnki.11-2246.2024.0729

Asbtract ( )   PDF (2922KB) ( )  

References | Related Articles | Metrics

Fundamental geo-entity is the positioning framework and bearing foundation of other geo-entity and related information. The new fundamental surveying and mapping system for the first time includes the geo-name entity into the classification of fundamental geo-entity,associates the geometric abstract model of geographic objects with attribute information,and realizes the comprehensive expression of the geo-name information of geographic objects.This paper collects and sorts out the latest electronic map database and point of interest database of Xiaoshan District,as the data source of the extraction of the geo-name information of geographic objects such as residential land,territory,transportation,water system,airport,housing estate,buildings,hotels,parks and scenic spots. The geo-entity is constructed by using the geo-entity abstract modeling and the entitative transformation of stock data. The geo-name entity elevation is assigned through the digital elevation model of Xiaoshan District,and the 3D geo-name entity data of Xiaoshan District is obtained,providing support for the special work of digital enabling social governance in Xiaoshan District.

The ocean literacy of GIS major students

LI Lianying, REN Fu, YANG Min, ZHAO Jianjun

2024, 0(7):  168-172.  doi:10.13474/j.cnki.11-2246.2024.0730

Asbtract ( )   PDF (1358KB) ( )  

References | Related Articles | Metrics

This article analyzes the connotation of ocean literacy and the content that should be included in the ocean literacy of GIS major students. It explores ways to cultivate ocean literacy, including participation in ocean-related scientific research projects, academic competitions, and learning and practice in ocean-related courses and other systematic training. Finally, taking Wuhan University's geographic information science major as an example, the article analyzes the ocean literacy of GIS major students from the aspects of course teaching, entrepreneurial innovation training, and employment trends. It is believed that students majoring in geographic information science have outstanding GIS practical ability and good ocean literacy.

Enhancing few-shot UAV image classification with laser navigation-assisted decision fusion

XIE Xingsheng, ZHANG Yongting, DING Zongbao, JIANG Yuhuan, LIU Jian

2024, 0(7):  173-177.  doi:10.13474/j.cnki.11-2246.2024.0731

Asbtract ( )   PDF (1660KB) ( )  

References | Related Articles | Metrics

In recent years, the integration of laser navigation technology with few-shot UAV aerial image classification has provided unprecedented precise spatial positioning and valuable information for fields such as land use survey, urban planning, and environmental monitoring, significantly enhancing the importance of classification technology applications. This study proposes a method of few-shot UAV aerial image classification that integrates laser navigation and decision fusion techniques, aimed at improving classification performance and spatial positioning accuracy. By utilizing the high-precision geographic location information provided by the laser navigation system, the feature extraction process of aerial images is optimized. The study adopts self-supervised learning to construct auxiliary tasks, enhancing the generalization ability of feature extractors through rotation and flipping techniques. Moreover, combining feature extractors trained with two self-supervised paradigms and utilizing a logistic regression classifier for the classification task. A novel decision fusion module is designed to automatically adjust the weights of each decision, enhancing classification accuracy. Experimental results on the NWPU-RESISC45 and UC Merced datasets validate the effectiveness and advanced nature of the proposed method, demonstrate the potential of laser navigation technology in enhancing few-shot UAV aerial image classification.

Application of BeiDou Navigation System in the numerical weather model

DU Mingbin, CAO Yunchang, WANG Xiaofeng, GUO Wei, TU Manhong, YIN Yue, GU Wen, LIANG Hong

2024, 0(7):  178-181.  doi:10.13474/j.cnki.11-2246.2024.0732

Asbtract ( )   PDF (2273KB) ( )  

References | Related Articles | Metrics

The meteorology is one of the earliest important fields to apply the BDS in. An unprecedented pattern has gradually formed for the comprehensive application of BDS to meteorology with the development of BDS. The construction and application of meteorological industry systems based on BDS ground-based augmentation systems can effectively improve the meteorological observation capabilities of GNSS in China, and also reduce the dependence of current GNSS applications on the GPS and GLONASS. This is of great significance for the development of meteorological application and the support for BDS industrialization. The current advances in the GNSS application in the numerical weather models are reviewed in this paper. The observation error is estimated for real-time inversion of precipitable water vapor (PWV) with BDS, based on the application of BDS in the numerical weather model. Through BDS-PWV assimilation application experiments in regional numerical weather model, the preliminary evaluation of the improvement effect is conducted for BDS-PWV assimilation application in the regional numerical weather model of the East China.