Automatic Radiometric Improvement of Moon Images for Shadow Segmentation

Thumbnail Image

Files

master_Kumar_Rachit_2015.pdf (23.93 MB)

URL

Journal Title

Journal ISSN

Volume Title

Sähkötekniikan korkeakoulu | Master's thesis
Unless otherwise stated, all rights belong to the author. You may download, display and print this publication for Your own personal use. Commercial use is prohibited.

Authors

Date

2015-08-24

Department

Major/Subject

Space Robotics and Automation

Mcode

AS3004

Degree programme

SST - Space Science and Technology (TS2005)

Language

en

Pages

73 + 11

Series

Abstract

Kaufmann et al. [18] have proposed a method for pose estimation of the spacecraft during the descent phase by matching the shadows between real time images and reference images. The shadow segmentation of the real time images is largely affected by the lunar surface reflectance, the lunar surface features and the illumination conditions. The thesis investigates various radiometric enhancement methods to reduce the effect of these artefacts on the shadow segmentation. An enhancement pipeline was designed to enhance the contrast of the images. An automated classification of images was also implemented in the pipeline based on topographical information and mathematical parameters. The Narrow Angle Camera (NAC) images from Lunar Reconnaissance Orbiter (LRO) mission were used, to develop the automated classification logic of the pipeline using the training data set and to validate the performance of the pipeline using the test data set. The reference images were rendered from the Digital Terrain Model (DTM) files of the corresponding NAC images. The result shows that enhancement of the descent images increases the amount of segmented shadows, when compared with the shadow segmented original image. The percentage of correct shadow match between the shadow segmented virtual image and shadow segmented enhanced images are higher compared to the shadow segmented original image. Further, it is observed that the applied enhancement method depends on the surface reflectance and the incidence angle.

Description

Supervisor

Visala, Arto

Thesis advisor

Lingenauber, Martin

Keywords

image processing, optical navigation, shadow segmentation, lunar image classification

Other note

Citation

Permanent link to this item

https://urn.fi/URN:NBN:fi:aalto-201509184365

Collections

[dipl] Sähkötekniikan korkeakoulu / ELEC