In this study, cases and limitations of each type of cadastral surveying results determination in the graphical area were reviewed. And as a way to overcome this, we tried to suggest a way to automatically match the land status boundary extracted from...
In this study, cases and limitations of each type of cadastral surveying results determination in the graphical area were reviewed. And as a way to overcome this, we tried to suggest a way to automatically match the land status boundary extracted from the UAV image with the cadastral map boundary. The research results are as follows.
After selecting the experimental area, an automatic matching experiment and accuracy analysis of the land status boundary and the cadastral map boundary were conducted. First, land status boundaries were extracted using UAV images. As a result of the accuracy analysis, the RMSE of the X-coordinate difference was 0.222m, the RMSE of the Y-coordinate difference was 0.105m, and the RMSE of the connection error was calculated as 0.246m. It was judged that this could be used as data for cadastral surveying results determination within the permissible range of boundary point error stipulated in the relevant laws and regulations.
For the implementation of the automatic matching algorithm, a 2D helmet/affine transformation matching algorithm was implemented using a total survey system. Using the implemented function, matching between land status boundary and cadastral map boundary was performed. For the application of the two methods, a reference matching point was selected and matching was performed based on this. In order to compare matching accuracy, RMSE was calculated using 10 check points and 5 matching points.
As a result of applying the 2D helmert algorithm, the RMSE of the X coordinate was calculated as 0.567m, the RMSE of the Y coordinate was calculated as 0.687m, and the RMSE of the connection error was calculated as 0.891m. As a result of applying the 2D affine algorithm using the same transformation points and checkpoints, the RMSE of the X coordinate was 0.547m, the RMSE of the Y coordinate was calculated as 0.700m, and the RMSE of the connection error was 0.888 m was calculated. As a result of comparing the amount of distortion between the two methods, it was found that the 2D helmert method had less distortion than the 2D affine method. When both matching accuracy and amount of distortion are considered, the 2D helmet method is judged to be a more consistent method to apply to cadastral surveying results determination. Therefore, it is judged that the most reasonable method is to perform matching in a 2D helmet) method and secure a large number of reference points for matching.
Based on the experimental results of the study, the application plan of the automatic matching method of land status boundary and cadastral map boundary was proposed. First, amendments to relevant laws and regulations for the extraction and use of UAV-based land status boundaries were presented. The advancement of land status boundary extraction by UAV was proposed. It was proposed that UAV LiDAR provides accurate elevation information together with the advantages of UAV surveying, so that higher accuracy can be achieved when used for extraction. Finally, it is suggested that the efficiency of the 2D helmet method application needs to be strengthened by implementing and adding functions to the current cadastral surveying system, the Landygo system.