Date of Award

2014

Degree Name

Physical and Applied Science

College

College of Science

Type of Degree

M.S.

Document Type

Thesis

First Advisor

Ralph Oberly

Second Advisor

Anne Axel

Third Advisor

James Wolfe

Abstract

A process is described to convert aerial photographs from flat images to 3D point clouds and then convert into height maps to be used as pseudo digital elevation models for surface modeling. All software used in the process is either free or open source. The process uses a DJI Phanton multicoper and two Canon point and shoot digital cameras. One camera is unaltered, and a second camera is modified to produce infrared images. A DJI Phantom FC-40 multicopter is used as the aerial platform to carry the cameras. Multiple paths are described to convert from still images (or video to still images) to N-view matches, followed by sparse point clouds then dense point clouds. Point clouds are distinct 3D points charted in an XYZ coordinate system. The dense point clouds can be converted into 3D models for viewing and analysis. A height map is extracted from the point cloud and surface images (in raster format) are created and then used in QGIS or ArcMap as pseudo digital elevation models for surface modeling. Finally, the digital elevation models are evaluated in comparison to similar LIDAR images.

Subject(s)

Remote sensing - Atmospheric effects.

Aerial photogrammetry.

Download

Included in

Physics Commons

Share

COinS