消防通道场景下的车辆检测技术分析-analysis of vehicle detection technology in fire passage scene.docxVIP

重庆大学硕士学位论文 重庆大学硕士学位论文 英文摘要 PAGE PAGE IV criterion. The simulation results for the fusion of the components in the vehicle sequence images show that the effect of the multi-features fusion method based on the Fisher criterion is the best. ③ In the validation phase of the interesting region of the vehicle, a vehicle detection method based on multi-features fusion cascading classifier is proposed. First, the corner features, edge features and textural features are integrated by a multi-feature integration algorithm based on the Fisher criterion, and the strong classifier is generated utilizing the Adaboost algorithm by training the characteristics derived by the Fisher criterion and the Harr-like characteristics derived by the “integral image”. Then, the interesting region of the vehicle is verified by the trained cascading classifier. The experimental results show that compared with the traditional methods of vehicle detection, the proposed method can self-adaptively detect vehicles in the fire engine access with a better detection rate and false positive rate in different climatic conditions. Keywords: Fire engine access;Vehicle detection; Shadow feature; Multi-feature fusion; Cascade classifier 重庆大学硕士学位论文目 重庆大学硕士学位论文 目 录 目 录 中文摘要 I 英文摘要II 1 绪 论 1 1.1 课题研究背景及意义 1 1.2 国内外研究现状 1 1.3 本文的主要研究内容及技术路线 5 1.4 本文的结构安排 6 2 车辆感兴趣区域提取方法 8 2.1 图像预处理 8 2.1.1 图像灰度化 8 2.1.2 图像去噪处理 9 2.1.3 图像形态学处理 12 2.2 基于车辆底部阴影特征的感兴趣区域提取 14 2.2.1 车辆底部阴影自适应阈值方法 14 2.2.2 车辆底部阴影边缘提取 17 2.2.3 车辆感兴趣区域确定 19 2.3 本章小结 19 3 车辆多特征融合算法 20 3.1 车辆图像的特征提取 20 3.1.1 角点特征提取 20 3.1.2 边缘特征提取 21 3.1.3 纹理特征提取 22 3.2 基于加权马氏距离的车辆多特征融合算法 23 3.2.1 加权马氏距离算法描述 23 3.2.2 多特征融合算法验证 24 3.3 基于类间方差的车辆多特征融合算法 27 3.3.1 类间方差算法描述 27 3.3.2 多特征融合算法验证 27 HYPERLINK \l _TOC_250009 3.4 基于 Fisher 准则的车辆多特征融合算法 29 HYPERLINK \l _TOC_250008 3.4.1 Fisher 准则算法描述 29 3.4.2 多特征融合算法验证 30 3.5 本章小结 33 4 基于多特征融合级联分类器的车辆检测方法 34 HYPERLINK \l _TOC_250007 Harr-like 特征 34 HYPERLINK \l _TOC_250006 Harr-like 矩形