SURF Feature Based Monocular Simultaneous Localization and Mapping
碩士 === 國立臺灣科技大學 === 電機工程系 === 97 === Due to the development of the area of intelligent robots which are given many abilities, such as basic actions of picking up and putting down, motion, and moving by itself. It is an important basis for building the spatial relationship between robots and surround...
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Others |
Language: | zh-TW |
Published: |
2009
|
Online Access: | http://ndltd.ncl.edu.tw/handle/85413562733382610711 |
id |
ndltd-TW-097NTUS5442069 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-TW-097NTUS54420692016-05-02T04:11:39Z http://ndltd.ncl.edu.tw/handle/85413562733382610711 SURF Feature Based Monocular Simultaneous Localization and Mapping 以SURF特徵為基礎的單眼視覺同步定位與地圖建置 Ming-nan Yen 顏銘男 碩士 國立臺灣科技大學 電機工程系 97 Due to the development of the area of intelligent robots which are given many abilities, such as basic actions of picking up and putting down, motion, and moving by itself. It is an important basis for building the spatial relationship between robots and surrounding environment, so simultaneous localization and mapping (SLAM) is a key point of scene perception. Vision based SLAM technique is recently a popular issue. Vision based SLAM technique is majorly in capturing scene images through vision sensors, and extract basic representative features, such as edges and corners to localization and mapping. However, image sensors are not fixed for feature observation and angles, and these features may cause difficulties of matching and errors of SLAM due to differences of scale, rotation, and shape diversity. This study discusses how to use cameras to implement SLAM, and we propose SURF to increase robustness of SLAM for robots according to feature extraction and matching problems in references. Hsin-Teng Hsu 許新添 2009 學位論文 ; thesis 73 zh-TW |
collection |
NDLTD |
language |
zh-TW |
format |
Others
|
sources |
NDLTD |
description |
碩士 === 國立臺灣科技大學 === 電機工程系 === 97 === Due to the development of the area of intelligent robots which are given many abilities, such as basic actions of picking up and putting down, motion, and moving by itself. It is an important basis for building the spatial relationship between robots and surrounding environment, so simultaneous localization and mapping (SLAM) is a key point of scene perception.
Vision based SLAM technique is recently a popular issue. Vision based SLAM technique is majorly in capturing scene images through vision sensors, and extract basic representative features, such as edges and corners to localization and mapping. However, image sensors are not fixed for feature observation and angles, and these features may cause difficulties of matching and errors of SLAM due to differences of scale, rotation, and shape diversity. This study discusses how to use cameras to implement SLAM, and we propose SURF to increase robustness of SLAM for robots according to feature extraction and matching problems in references.
|
author2 |
Hsin-Teng Hsu |
author_facet |
Hsin-Teng Hsu Ming-nan Yen 顏銘男 |
author |
Ming-nan Yen 顏銘男 |
spellingShingle |
Ming-nan Yen 顏銘男 SURF Feature Based Monocular Simultaneous Localization and Mapping |
author_sort |
Ming-nan Yen |
title |
SURF Feature Based Monocular Simultaneous Localization and Mapping |
title_short |
SURF Feature Based Monocular Simultaneous Localization and Mapping |
title_full |
SURF Feature Based Monocular Simultaneous Localization and Mapping |
title_fullStr |
SURF Feature Based Monocular Simultaneous Localization and Mapping |
title_full_unstemmed |
SURF Feature Based Monocular Simultaneous Localization and Mapping |
title_sort |
surf feature based monocular simultaneous localization and mapping |
publishDate |
2009 |
url |
http://ndltd.ncl.edu.tw/handle/85413562733382610711 |
work_keys_str_mv |
AT mingnanyen surffeaturebasedmonocularsimultaneouslocalizationandmapping AT yánmíngnán surffeaturebasedmonocularsimultaneouslocalizationandmapping AT mingnanyen yǐsurftèzhēngwèijīchǔdedānyǎnshìjuétóngbùdìngwèiyǔdetújiànzhì AT yánmíngnán yǐsurftèzhēngwèijīchǔdedānyǎnshìjuétóngbùdìngwèiyǔdetújiànzhì |
_version_ |
1718254165784788992 |