| DC 欄位 |
值 |
語言 |
| DC.contributor | 資訊工程學系 | zh_TW |
| DC.creator | 周恩鋒 | zh_TW |
| DC.creator | En-fong Chou | en_US |
| dc.date.accessioned | 2010-7-26T07:39:07Z | |
| dc.date.available | 2010-7-26T07:39:07Z | |
| dc.date.issued | 2010 | |
| dc.identifier.uri | http://ir.lib.ncu.edu.tw:444/thesis/view_etd.asp?URN=975202088 | |
| dc.contributor.department | 資訊工程學系 | zh_TW |
| DC.description | 國立中央大學 | zh_TW |
| DC.description | National Central University | en_US |
| dc.description.abstract | 近年來，車輛輔助安全駕駛系統的議題愈來愈受重視。在安全駕駛的研究上，我們架設一部相機於車前，用來偵測前車或障礙物，分析前車和已車的行車狀況，以確保駕駛人行車時的安全。
前車偵測中最常使用的方法就是利用車輛在地面上的陰影偵測前車，但是在雨天或天候不良的狀況下，無法有效的偵測到陰影。為了在不同的天氣下都能穩定的偵測前車，我們提出以水平邊及垂直邊的互動組合偵測前車。首先，我們動態的二值化水平邊及垂直邊；並且利用水平邊的真實寬度濾除過短的水平邊。接著，估算水平邊上方的垂直邊的顯著程度，如果水平邊上方的垂直邊夠顯著，我們判斷這條水平邊為車輛的最底部的水平邊，並在水平邊的左右兩側找到垂直的邊緣，產生候選車輛。如果水平邊上方的垂直邊不顯著，我們則是在水平邊的左右找一組對稱的垂直邊配對作為垂直邊緣。這時我們無法以水平邊找到車輛的底部，我們改以垂直邊配對的底部作為候選車輛的底部，產生候選車輛。最後，我們利用支持向量機對候選車輛做驗證判斷候選車輛是否為車輛或障礙物。
在實驗中，我們利用不同的天氣如晴天、陰天、雨天、向陽、與傍晚的影片測試演算法。我們在晴天、陰天、和向陽的天候狀況下，有超過 90% 的偵測率。此外，在雨天及傍晚的天候狀況下，也有超過 80% 的偵測率。
| zh_TW |
| dc.description.abstract | Recently, the issue on real-time driver assistance system becomes more and more popular. In this study, we propose a method to detect preceding vehicles with the road scene which is captured by a camera.
To detect preceding vehicle robustly under variant weather conditions, the proposed method utilizes the horizontal and vertical edges instead of underneath shadow which can’t be detected while the weather is bad such as rainy day or night. We check the vertical edge above horizontal edge to confirm the location of horizontal edge beside object and generate a candidate vehicle with vertical borders while horizontal edge is located at the bottom of object. If the bottom of object can’t be found by a horizontal edge, we provide a method to find the vertical borders and bottom of the vehicle by finding symmetric vertical edge pair. Then, we estimate the actual width of the vehicle to filter out objects which are too small or big. At last, we use SVM to verify the detected vehicle.
In the experiments, the proposed method was evaluated on images of different weather conditions such as sunny day, cloudy day and rainy day. The detection rate of preceding detection is over 90% in sunny day and cloudy day and is over 80% in rainy day.
| en_US |
| DC.subject | 前車偵測 | zh_TW |
| DC.subject | 天候變化 | zh_TW |
| DC.subject | 支持向量機 | zh_TW |
| DC.subject | weather-adapted | en_US |
| DC.subject | vehicle detection | en_US |
| DC.subject | support vector machine | en_US |
| DC.title | 可適應天候變化的前車偵測技術 | zh_TW |
| dc.language.iso | zh-TW | zh-TW |
| DC.title | Weather-adapted Preceding Vehicle Detection | en_US |
| DC.type | 博碩士論文 | zh_TW |
| DC.type | thesis | en_US |
| DC.publisher | National Central University | en_US |