Lane Recognition and Front-Vehicle Distance Detection Based on Fluctuating Weather Adaptation Threshold
碩士 === 國立臺北科技大學 === 電子工程系 === 107 === Traffic accidents often cause casualties and damage. Most traffic accidents are caused by driving negligence, vehicle deviation, or not lack of spacing. Therefore, the problems of driving safety are getting more and more attention. Installing a car camcorder in...
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Others |
Language: | zh-TW |
Published: |
2019
|
Online Access: | http://ndltd.ncl.edu.tw/handle/rjh36v |
id |
ndltd-TW-107TIT00427069 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-TW-107TIT004270692019-11-16T05:27:42Z http://ndltd.ncl.edu.tw/handle/rjh36v Lane Recognition and Front-Vehicle Distance Detection Based on Fluctuating Weather Adaptation Threshold 基於天氣變化適應閾值之車道辨識及前車距離偵測 YEN, SHANG-CHIH 嚴上智 碩士 國立臺北科技大學 電子工程系 107 Traffic accidents often cause casualties and damage. Most traffic accidents are caused by driving negligence, vehicle deviation, or not lack of spacing. Therefore, the problems of driving safety are getting more and more attention. Installing a car camcorder in the car has become widespread, and vehicle auxiliary systems are gradually booming. Advanced Driver Assistance System analyzes the changes of the environment outside the vehicle and the conditions of the vehicle while driving. To warn driver if the situation may cause damages. This paper proposes a front-vehicle and lane detection warning system that can adapt to the weather changes. This system can provide the driver with the warning. The image of the road condition are taken as input by the actual driving recorder device. Using the image processing algorithm to determine whether the lane is offset or the lack of the spacing to the front vehicle. In order to adapt to the different weather, the system uses the Region of Interest to define the environment, then get the relationship between the pixel intensity information after image processing and the binarization threshold. Using the algorithm to recognize the weather and get the features of horizontal, the shadow or light of the vehicle. The line of lane are obtained by the Hough transform. The angles between the lines of lane and parallel line of the vehicle are calculated to determine the lane offset is happened or not. After detecting the line of the lane and the images processed area, it is used for determine if a vehicle is in front. Last the system uses the front vehicle distance algorithm to get the distance to the front-vehicle. The paper introduces the environment of development, equipment, the correction rate and the result with different weather. TAN, SUN-YEN 譚巽言 2019 學位論文 ; thesis 71 zh-TW |
collection |
NDLTD |
language |
zh-TW |
format |
Others
|
sources |
NDLTD |
description |
碩士 === 國立臺北科技大學 === 電子工程系 === 107 === Traffic accidents often cause casualties and damage. Most traffic accidents are caused by driving negligence, vehicle deviation, or not lack of spacing. Therefore, the problems of driving safety are getting more and more attention. Installing a car camcorder in the car has become widespread, and vehicle auxiliary systems are gradually booming. Advanced Driver Assistance System analyzes the changes of the environment outside the vehicle and the conditions of the vehicle while driving. To warn driver if the situation may cause damages.
This paper proposes a front-vehicle and lane detection warning system that can adapt to the weather changes. This system can provide the driver with the warning. The image of the road condition are taken as input by the actual driving recorder device. Using the image processing algorithm to determine whether the lane is offset or the lack of the spacing to the front vehicle. In order to adapt to the different weather, the system uses the Region of Interest to define the environment, then get the relationship between the pixel intensity information after image processing and the binarization threshold. Using the algorithm to recognize the weather and get the features of horizontal, the shadow or light of the vehicle. The line of lane are obtained by the Hough transform. The angles between the lines of lane and parallel line of the vehicle are calculated to determine the lane offset is happened or not. After detecting the line of the lane and the images processed area, it is used for determine if a vehicle is in front. Last the system uses the front vehicle distance algorithm to get the distance to the front-vehicle.
The paper introduces the environment of development, equipment, the correction rate and the result with different weather.
|
author2 |
TAN, SUN-YEN |
author_facet |
TAN, SUN-YEN YEN, SHANG-CHIH 嚴上智 |
author |
YEN, SHANG-CHIH 嚴上智 |
spellingShingle |
YEN, SHANG-CHIH 嚴上智 Lane Recognition and Front-Vehicle Distance Detection Based on Fluctuating Weather Adaptation Threshold |
author_sort |
YEN, SHANG-CHIH |
title |
Lane Recognition and Front-Vehicle Distance Detection Based on Fluctuating Weather Adaptation Threshold |
title_short |
Lane Recognition and Front-Vehicle Distance Detection Based on Fluctuating Weather Adaptation Threshold |
title_full |
Lane Recognition and Front-Vehicle Distance Detection Based on Fluctuating Weather Adaptation Threshold |
title_fullStr |
Lane Recognition and Front-Vehicle Distance Detection Based on Fluctuating Weather Adaptation Threshold |
title_full_unstemmed |
Lane Recognition and Front-Vehicle Distance Detection Based on Fluctuating Weather Adaptation Threshold |
title_sort |
lane recognition and front-vehicle distance detection based on fluctuating weather adaptation threshold |
publishDate |
2019 |
url |
http://ndltd.ncl.edu.tw/handle/rjh36v |
work_keys_str_mv |
AT yenshangchih lanerecognitionandfrontvehicledistancedetectionbasedonfluctuatingweatheradaptationthreshold AT yánshàngzhì lanerecognitionandfrontvehicledistancedetectionbasedonfluctuatingweatheradaptationthreshold AT yenshangchih jīyútiānqìbiànhuàshìyīngyùzhízhīchēdàobiànshíjíqiánchējùlízhēncè AT yánshàngzhì jīyútiānqìbiànhuàshìyīngyùzhízhīchēdàobiànshíjíqiánchējùlízhēncè |
_version_ |
1719291700232323072 |