3D modelling method of high above ground rock art painting using multispectral camera

This paper presents a developed methodology of 3D modelling of rock art painting located at high above shelter floor based on close-range photogrammetric technique. Instead of using elevated devices to reach inaccessible paintings due to high elevation, the developed method proposed a simple techniq...

Full description

Saved in:
Bibliographic Details
Main Authors: Zainuddin, K., Majid, Z., Ariff, M. F. M., Idris, K. M., Darwin, N.
Format: Conference or Workshop Item
Language:English
Published: 2022
Subjects:
Online Access:http://eprints.utm.my/id/eprint/98423/1/MohdFarid2022_3DModellingMethodOfHighAboveGround.pdf
http://eprints.utm.my/id/eprint/98423/
http://dx.doi.org/10.5194/isprs-archives-XLVI-2-W1-2022-537-2022
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This paper presents a developed methodology of 3D modelling of rock art painting located at high above shelter floor based on close-range photogrammetric technique. Instead of using elevated devices to reach inaccessible paintings due to high elevation, the developed method proposed a simple technique. The camera was mounted on an expandable pole to acquire the panel with normal and tilted camera settings. Due to inaccessibility to place the control point marker on the panel surface, the distribution of photo control was positioned below the panel. Based on the configuration, the image of the panel was acquired using the low-spatial resolution multispectral camera in a two elevated strips position that imitates the aerial photogrammetry flight line. The camera was set parallelly to the painting panel on both strips, with additional tilted geometry included at the upper strip of the camera block. The acquired multispectral images were then processed using commercial SfM photogrammetry software to generate a 3D point cloud. The accuracy of constructed point cloud was then analysed by comparing it with the point cloud generated using a terrestrial laser scanner (TLS). The result has shown that the multispectral 3D point cloud has a small deviation against the TLS point cloud. The mean deviation was -0.43mm, indicating a slight downscaled on the multispectral point cloud.