Digital Elevation Models (DEMs) are commonly used for environment, engineering, and architecture-related studies. One of the most important factors for the accuracy of DEM generation is the process ...of spatial interpolation, which is used for estimating the height values of the grid cells. The use of machine learning methods, such as artificial neural networks for spatial interpolation, contributes to spatial interpolation with more accuracy. In this study, the performances of FBNN interpolation based on different parameters such as the number of hidden layers and neurons, epoch number, processing time, and training functions (gradient optimization algorithms) were compared, and the differences were evaluated statistically using an analysis of variance (ANOVA) test. This research offers significant insights into the optimization of neural network gradients, with a particular focus on spatial interpolation. The accuracy of the Levenberg–Marquardt training function was the best, whereas the most significantly different training functions, gradient descent backpropagation and gradient descent with momentum and adaptive learning rule backpropagation, were the worst. Thus, this study contributes to the investigation of parameter selection of ANN for spatial interpolation in DEM height estimation for different terrain types and point distributions.
Interpolation of a spatially continuous variable from point samples is an important field in spatial analysis and surface models for geosciences. In this study, spatial interpolation methods which ...are Inverse Distance Weighted (IDW), Ordinary Kriging (OK), Modified Shepard's (MS), Multiquadric Radial Basis Function (MRBF) and Triangulation with Linear (TWL), and Multi-Layer Perceptron (MLP) which is an Artificial Neural Networks (ANN) method were compared in order to predict height for different point distributions such as curvature, grid, random and uniform on a Digital Elevation Model which is an USGS National Elevation Dataset (NED). This study also aims to quantify the effects of topographic variability and sampling density. Errors of different interpolations and ANN prediction were evaluated for different point distributions and three different cross-sections on the characteristic parts of the surface were selected and analyzed. Generally, OK, MS, MRBF and TWL gave promising results and were more effective in terms of characteristics of surface than MLP and IDW. Although MLP simplified the contours obtained from predicted heights, it was a satisfactory predictor for curvature, grid, random and uniform distributions.
The aim of this article is to try to determine whether there’s a statistically significant difference between coordinates obtained under different elevation angles and measuring epochs through ...different correction methods in Network Real Time Kinematic Applications. For this purpose, analysis of variance was used. When the statistical data obtained according to different variables are compared, the accuracy seems to be less in the direction of h coordinate axis. There is no a statistically significant difference among the values obtained according to different epoch numbers in all directions of the coordinate axes. The subsets formed according to satellite elevation angles show similar characteristics along Y and h coordinate axes while they differ 30° according to other angles in the direction of X coordinate axis. Based on the correction methods, while FKP and VRS method show different characteristics with respect to each other, MAC method shows similar characteristics compared to other methods.
This paper describes an experiment carried out with the GNSS RTK method to combine GPS and GLONASS measurements in obstructed and unobstructed sites. This study investigates the RTK achievable ...accuracy and repeatability under different satellite constellations and site conditions in Samatya (coastal area), Istanbul, Turkey. The RTK (GPS and GPS/GLONASS) measurement techniques are described and compared in terms of productivity and precision. Surveyed area, control and test points as well as the impact area of trees where the precision decreases are visualised using Autodesk LandXplorer Studio ProfessionalTM Software. These obtained results indicate that integrating RTK GPS system with GLONASS is favoured for surveying applications.
Interpolation of a spatially continuous variable from point samples is an important field in spatial analysis and surface models for geosciences. In this study, spatial interpolation methods which ...are Inverse Distance Weighted, Ordinary Kriging (OK), Modified Shepard's (MS), Multiquadric Radial Basis Function (MRBF) and Triangulation with Linear (TWL), and Multi-Layer Perceptron (MLP) which is an Artificial Neural Networks (ANN) method were compared in order to predict height for different point distributions such as curvature, grid random and uniform on a Digital Elevation Model which is an USGS National Elevation Dataset (NED). Errors of different interpolations and ANN prediction were evaluated for different point distributions and three different cross-sections on the characteristic parts of the surface were selected and analyzed Generally, OK MS, MRBF and TWL gave promising results and were more effective in terms of characteristics of surface than MLP and IDW. Although MLP simplified the contours obtained from predicted heights, it was a satisfactory predictor for curvature, grid, random and uniform distributions.
The network real-time kinematic (NRTK) positioning technique is currently used in numerous applications. The aim of this study was to better understand the process of obtaining accurate positions by ...statistically evaluating the significance of differences between repeated measurements for a single point at different times of the day (morning, noon, and evening) using the Virtual Reference Station (VRS), Flächen Korrektur Parameter (FKP), and Master Auxiliary Concept (MAC) correction methods. An analysis of variance (ANOVA) was used to this effect. Further analysis was carried out to determine the accuracy and precision of the coordinate differences obtained via a static GNSS (global navigation satellite system) and by averaging the repeated measurements. It was determined that the accuracy and precision of the vertical component of the coordinates were lower than that of the horizontal component. The FKP correction method yielded the best results. It was observed that the accuracy and precision of the measurements taken at noon were the lowest. The ANOVA showed that the differences between repeated measurements were statistically significant and that there were outlier measurements. The results of this study are important for NRTK users to be able to statistically evaluate different measurement configurations and obtain positions with the desired accuracy and precision.
This article describes an experiment carried out with the GNSS RTK method to combine GPS and GLONASS measurements in obstructed and unobstructed sites. This study investigates the RTK achievable ...accuracy and repeatability under different satellite constellations and site conditions in Samatya (coastal area), Istanbul, Turkey. The RTK (GPS and GPS/GLONASS) measurement techniques are described and compared in terms of productivity and precision. Surveyed area, control and test points as well as the impact area of trees where the precision decreases are visualised using Autodesk LandXplorer Studio ProfessionalTM Software. These obtained results indicate that integrating RTK GPS system with GLONASS is favoured for surveying applications.
To assess the accuracy of vertical components obtained by Classical RTK and Network Real Time Kinematic positioning with different correction methods with setting of different number of epoch, height ...differences obtained from precise geometric leveling was used as reference height differences along with those obtained from static GPS only solutions, and combined solutions of GPS and GLONASS on different GNSS days. One-way variance analysis (ANOVA) were applied to the height differences to test them whether they are significantly different or not. The significant differences, if any, were further tested by using Post Hoc test to reveal the effects of the method used, the epoch number, or the measuring time, which is responsible for the significant differences. The results from ANOVA showed that the number of epoch had not much effect on the vertical components but on the correction methods. MAC method presented the most deviated results compared to the other methods with different GNSS days’ solutions. In general, VRS method was found to be the most consistent method among others used. This study is important in both; an accuracy assessment with respect to the most precise method in heights (i.e., precise geometric leveling) was carried. The techniques (ANOVA and Post Hock Test) used in the accuracy assessment.
•The Displacement Monitoring System was designed for accuracy evaluation for NRTK Methods.•Interferometer was used to determine the reference displacements for analyses.•6 mm accuracy achieved for ...overall assessment by VRS, FKP and MAC real time positioning methods.
Nowadays, Global Navigation Satellite Systems (GNSS) are widely used for deformation monitoring of structures (dams, skyscrapers, etc.), rail systems and landslides. In this study, a displacement monitoring system was designed to evaluate the precision and accuracy of FKP (Flächen Korrektur Parameter), VRS (Virtual Reference Station) and MAC (Master Auxiliary Concept) Network Real Time Kinematic (NRTK) GNSS methods for determining the vertical and horizontal displacements. For this purpose, the distances (simulated displacements) were measured on the X, Y and h coordinate axes using NRTK methods and Static GNSS Technique and compared with the displacements measured by an interferometer (0.1 mm precision) on this designed system. Displacements measured by using the displacement monitoring system were accepted as true values. In general, NRTK methods provides sufficient precision for determining the displacements respect to displacement monitoring system results. Performance of the FKP correction method is better especially on horizontal axes, while MAC method on vertical. In the overall assessment, average differences of displacements between VRS, FKP and MAC methods and the displacement monitoring system were found 6.8 mm, 5.1 mm and 5.2 mm, respectively. Mean square errors of NRTK method was found 4.7 mm in horizontal and 7.9 mm in vertical for overall assessment.
PDF
