Front-Vehicle Detection using Stereo Vision of Asynchronous Camera
碩士 === 國防大學中正理工學院 === 電子工程研究所 === 96 === Driver assistance system is an important topic for the intelligent vehicle, and the front-vehicle detection is the major research field for the driver assistance system. For the real-world environment, the vibrations from the rough road, working engine, and t...
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Others |
Language: | zh-TW |
Published: |
2008
|
Online Access: | http://ndltd.ncl.edu.tw/handle/57769031678203135754 |
id |
ndltd-TW-096CCIT0428028 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-TW-096CCIT04280282016-12-23T04:08:47Z http://ndltd.ncl.edu.tw/handle/57769031678203135754 Front-Vehicle Detection using Stereo Vision of Asynchronous Camera 非同步立體視覺之前方車輛偵測系統 Chen Wen Chung 陳文忠 碩士 國防大學中正理工學院 電子工程研究所 96 Driver assistance system is an important topic for the intelligent vehicle, and the front-vehicle detection is the major research field for the driver assistance system. For the real-world environment, the vibrations from the rough road, working engine, and the surrounding weather environment will influence the obstacle detection results. Parts of the researches use monocular detection system as the visual detection sensor for the front-vehicle detection, but this system cannot solve the distance calculation error under the vibrational influences. Thus, researchers start using the synchronous stereovision technology for a two-camera system to implement the obstacle detection. The cost and computing complexity obstruct the application in practice. This study uses two low-cost and asynchronous CMOS cameras as a platform of stereovision system and proposes an effective real-time front-vehicle detection algorithm. Images are captured by the asynchronous stereovision platform and stored into a computer. After a pre-process of the input image, the detection algorithm uses the edge information to locate the position of the front-vehicle. Then, get the disparity from a fast matching algorithm. According to the disparity value, system can convert the value to a relative distance between the front-vehicle and the detection system. In this study, the proposed algorithm can conquer the asynchronous exposure problem from the low-cost cameras and decrease the cost of the hardware system. In the meantime, the algorithm can speed up the computing time and obtain the accurate and real-time detection results 瞿忠正 2008 學位論文 ; thesis 66 zh-TW |
collection |
NDLTD |
language |
zh-TW |
format |
Others
|
sources |
NDLTD |
description |
碩士 === 國防大學中正理工學院 === 電子工程研究所 === 96 === Driver assistance system is an important topic for the intelligent vehicle, and the front-vehicle detection is the major research field for the driver assistance system. For the real-world environment, the vibrations from the rough road, working engine, and the surrounding weather environment will influence the obstacle detection results. Parts of the researches use monocular detection system as the visual detection sensor for the front-vehicle detection, but this system cannot solve the distance calculation error under the vibrational influences. Thus, researchers start using the synchronous stereovision technology for a two-camera system to implement the obstacle detection. The cost and computing complexity obstruct the application in practice.
This study uses two low-cost and asynchronous CMOS cameras as a platform of stereovision system and proposes an effective real-time front-vehicle detection algorithm. Images are captured by the asynchronous stereovision platform and stored into a computer. After a pre-process of the input image, the detection algorithm uses the edge information to locate the position of the front-vehicle. Then, get the disparity from a fast matching algorithm. According to the disparity value, system can convert the value to a relative distance between the front-vehicle and the detection system. In this study, the proposed algorithm can conquer the asynchronous exposure problem from the low-cost cameras and decrease the cost of the hardware system. In the meantime, the algorithm can speed up the computing time and obtain the accurate and real-time detection results
|
author2 |
瞿忠正 |
author_facet |
瞿忠正 Chen Wen Chung 陳文忠 |
author |
Chen Wen Chung 陳文忠 |
spellingShingle |
Chen Wen Chung 陳文忠 Front-Vehicle Detection using Stereo Vision of Asynchronous Camera |
author_sort |
Chen Wen Chung |
title |
Front-Vehicle Detection using Stereo Vision of Asynchronous Camera |
title_short |
Front-Vehicle Detection using Stereo Vision of Asynchronous Camera |
title_full |
Front-Vehicle Detection using Stereo Vision of Asynchronous Camera |
title_fullStr |
Front-Vehicle Detection using Stereo Vision of Asynchronous Camera |
title_full_unstemmed |
Front-Vehicle Detection using Stereo Vision of Asynchronous Camera |
title_sort |
front-vehicle detection using stereo vision of asynchronous camera |
publishDate |
2008 |
url |
http://ndltd.ncl.edu.tw/handle/57769031678203135754 |
work_keys_str_mv |
AT chenwenchung frontvehicledetectionusingstereovisionofasynchronouscamera AT chénwénzhōng frontvehicledetectionusingstereovisionofasynchronouscamera AT chenwenchung fēitóngbùlìtǐshìjuézhīqiánfāngchēliàngzhēncèxìtǒng AT chénwénzhōng fēitóngbùlìtǐshìjuézhīqiánfāngchēliàngzhēncèxìtǒng |
_version_ |
1718404988522201088 |