AUTOMATIC EXTRACTION INTERIOR ORIENTATION ON AERIAL PHOTOGRAPH USING IMAGE MATCHING METHOD

Wijaya, Helmy Mukti (2020) AUTOMATIC EXTRACTION INTERIOR ORIENTATION ON AERIAL PHOTOGRAPH USING IMAGE MATCHING METHOD. Masters thesis, Institut Teknologi Sepuluh November.

[img] Text
03311850010007_Master_Thesis.pdf - Accepted Version
Restricted to Repository staff only until 31 December 2021.
Available under License Creative Commons Attribution Non-commercial.

Download (3MB) | Request a copy

Abstract

Interior orientation is a set of parameters for the transformation of camera photo coordinates, which transforms the pixel coordinates to the photo coordinates. This parameter is used to calibrate the camera before an aerial photographic mission is carried out. The interior orientation parameter consists of the camera's focal length, lens distortion, principal point, location of the fiducial mark, camera resolution, and the level of the focal plane which has been calibrated and equivalent. All these parameters are attached to or presented on the camera sensor and the values of these parameters can usually be found from the camera's report page. The purpose of this research is to get the coordinates of the Fiducial Mark in the base image (Windows Patch) by automatic process. Herein, Fiducial Mark Template is created from the cropping photo frame to determine the value of the Fiducial Mark coordinates from the base image (Windows Patch). The Image Matching process is developed from the C ++ programming algorithm platform, which aims to obtain precise and accurate computational results. There are many techniques for performing Image Matching. The method used in this study Area Based Matching Normalized Cross-Correlation (NCC) whose function is precisely used in Image Matching computing. In conducting a comparative test in this case, 3 (three) templates are provided based on the dependent and independent variables which are used as test sample objects in determining the hypothesis of calculating the accuracy of Image Matching. Based on the calculated RMSE result of the sample test template, it is obtained as follows : Template 1 of 0.0072 mm, Template 2 of 0.0134 mm, and Template 3 of 0.0066 mm. As for the manual computation calculation as conventional calculation of the Fiducial Mark coordinates obtains the RMSE value of 0.0116 mm. Computation of 8 (eight) Fiducial Marks with a process of Normalized Cross-Correlation (NCC) of 346,034,176 is taken in 21 seconds. Figure out for Fiducial Mark coordinate with Image Matching method is very appropriate to be applied in the process of accelerating bundle adjustment computation to determine one of the interior orientation parameters.

Item Type: Thesis (Masters)
Uncontrolled Keywords: Interior Orientation, Fiducial Mark, Image Matching, Least Square, Normalized Cross-Correlation
Subjects: T Technology > TA Engineering (General). Civil engineering (General) > TA1637 Image processing--Digital techniques
Divisions: Faculty of Civil Engineering and Planning > Geomatics Engineering > 29101-(S2) Master Thesis
Depositing User: Helmy Mukti Wijaya
Date Deposited: 14 Aug 2020 03:41
Last Modified: 14 Aug 2020 03:41
URI: http://repository.its.ac.id/id/eprint/77855

Actions (login required)

View Item View Item