Abstract: | 隨著生活品質的提昇,車輛輔助安全駕駛的議題,也就變得越來越重要。在安全駕駛的研究議題上,我們將一部相機架設在車上,用來偵測及追蹤前車,並且偵測前車的方向燈與煞車燈,來分析前車的行車狀況,幫助駕駛人避免危險的發生。 到目前為止我們的系統已有十個偵測及分析功能 : 車道線偵測、多車道估計、虛實道路線分類、己車方向估計、己車左右位置估計、偏離車道警示、前車偵測、前車距離估計、煞車燈偵測、及方向燈偵測。其中的前車偵測與追蹤、方向燈與煞車燈偵測是本論文所提供的。在前車的偵測中,我們以車子的陰影來偵測前車的可能位置,然後利用車輛的垂直邊來偵測前車。之後,再以一個小矩形框來代表我們所找到的前車區塊。最後再利用車寬和車道寬的比例、對稱性、及區塊色彩標準差來驗證是否真的是車輛。在車燈偵測方面,我們根據所得到的前車位置,用前後張影像車燈之相對位置關係、對稱性、及出現次數來偵測煞車燈。用前後張影像車燈之相對位置關係、車燈的位置、群聚性、及閃爍頻率來偵測方向燈。 我們以多種的天候狀況影像 ; 例如,晴天、陰天、多雲、和雨天等,來測試我們的偵測效能。從實驗結果顯示,我們所提出的系統可以在不同的天候狀況下,即時且快速的偵測前車、煞車燈、及方向燈。 Recently, the issue on safe driving becomes more and more important. To aid the safe driving, we use a camera mounted on a vehicle to capture road scenes for detecting preceding vehicles, brake light, and turn signal of the vehicles. The proposed system consists of ten functions: lane detection, multiple lane estimation, classification of solid/dashed lane, direction estimation, vehicle lateral offset estimation, lane departure warning, preceding vehicles detection, distance of preceding vehicles estimation, brake light detection, and turn signal detection. In this thesis, the detection methods of preceding vehicles, brake light, and turn signal are proposed. We use cast shadow, left/right borders of a vehicle to detect preceding vehicle and use ratio of the road width and vehicle width, symmetry, and standard deviation to verify, use temporal difference, symmetry, and appearance times to detect brake light, and use temporal difference, position check, cluster, and frequency to detect turn signal. In the experiments, the proposed detector is evaluated on several different weather conditions such as sunny day, misty day, dusky day, cloudy day, and rainy day. From the experiment results, we find that the proposed approach can stably detect or track the preceding vehicles and their rear lamps in real time. |