Linked e-resources
Details
Table of Contents
Preface; Contents; Symbols; 1 Vision-Based Driver-Assistance Systems; 1.1 Driver-Assistance Towards Autonomous Driving; 1.2 Sensors; 1.3 Vision-Based Driver Assistance; 1.4 Safety and Comfort Functionalities; 1.5 VB-DAS Examples; 1.6 Current Developments; 1.7 Scope of the Book; 2 Driver-Environment Understanding; 2.1 Driver and Environment; 2.2 Driver Monitoring; 2.3 Basic Environment Monitoring; 2.4 Midlevel Environment Perception; 3 Computer Vision Basics; 3.1 Image Notations; 3.2 The Integral Image; 3.3 RGB to HSV Conversion; 3.4 Line Detection by Hough Transform; 3.5 Cameras
3.6 Stereo Vision and Energy Optimization3.7 Stereo Matching; 4 Object Detection, Classification, and Tracking; 4.1 Object Detection and Classification; 4.2 Supervised Classification Techniques; 4.2.1 The Support Vector Machine; 4.2.2 The Histogram of Oriented Gradients; 4.2.3 Haar-Like Features; 4.3 Unsupervised Classification Techniques; 4.3.1 k-Means Clustering; 4.3.2 Gaussian Mixture Models; 4.4 Object Tracking; 4.4.1 Mean Shift ; 4.4.1.1 Mean Shift Tracking; 4.4.2 Continuously Adaptive Mean Shift; 4.4.3 The Kanade-Lucas-Tomasi (KLT) Tracker; 4.4.4 Kalman Filter
4.4.4.1 Filter Implementation4.4.4.2 Tracking by Prediction and Refinement; 5 Driver Drowsiness Detection; 5.1 Introduction; 5.2 Training Phase: The Dataset; 5.3 Boosting Parameters; 5.4 Application Phase: Brief Ideas; 5.5 Adaptive Classifier; 5.5.1 Failures Under Challenging Lighting Conditions; 5.5.2 Hybrid Intensity Averaging; 5.5.3 Parameter Adaptation; 5.6 Tracking and Search Minimization; 5.6.1 Tracking Considerations; 5.6.2 Filter Modelling and Implementation; 5.7 Phase-Preserving Denoising; 5.8 Global Haar-Like Features; 5.8.1 Global Features vs. Local Features
5.8.2 Dynamic Global Haar Features5.9 Boosting Cascades with Local and Global Features; 5.10 Experimental Results; 5.11 Concluding Remarks; 6 Driver Inattention Detection; 6.1 Introduction; 6.2 Asymmetric Appearance Models; 6.2.1 Model Implementation; 6.2.2 Asymmetric AAM; 6.3 Driver's Head-Pose and Gaze Estimation; 6.3.1 Optimized 2D to 3D Pose Modelling; 6.3.2 Face Registration by Fermat-Transform; 6.4 Experimental Results; 6.4.1 Pose Estimation; 6.4.2 Yawning Detection and Head Nodding; 6.5 Concluding Remarks; 7 Vehicle Detection and Distance Estimation; 7.1 Introduction
7.2 Overview of Methodology7.3 Adaptive Global Haar Classifier; 7.4 Line and Corner Features; 7.4.1 Horizontal Edges; 7.4.2 Feature-Point Detection; 7.5 Detection Based on Taillights; 7.5.1 Taillight Specifications: Discussion; 7.5.2 Colour Spectrum Analysis; 7.5.3 Taillight Segmentation; 7.5.4 Taillight Pairing by Template Matching; 7.5.5 Taillight Pairing by Virtual Symmetry Detection; 7.6 Data Fusion and Temporal Information; 7.7 Inter-vehicle Distance Estimation; 7.8 Experimental Results; 7.8.1 Evaluations of Distance Estimation; 7.8.2 Evaluations of the Proposed Vehicle Detection
3.6 Stereo Vision and Energy Optimization3.7 Stereo Matching; 4 Object Detection, Classification, and Tracking; 4.1 Object Detection and Classification; 4.2 Supervised Classification Techniques; 4.2.1 The Support Vector Machine; 4.2.2 The Histogram of Oriented Gradients; 4.2.3 Haar-Like Features; 4.3 Unsupervised Classification Techniques; 4.3.1 k-Means Clustering; 4.3.2 Gaussian Mixture Models; 4.4 Object Tracking; 4.4.1 Mean Shift ; 4.4.1.1 Mean Shift Tracking; 4.4.2 Continuously Adaptive Mean Shift; 4.4.3 The Kanade-Lucas-Tomasi (KLT) Tracker; 4.4.4 Kalman Filter
4.4.4.1 Filter Implementation4.4.4.2 Tracking by Prediction and Refinement; 5 Driver Drowsiness Detection; 5.1 Introduction; 5.2 Training Phase: The Dataset; 5.3 Boosting Parameters; 5.4 Application Phase: Brief Ideas; 5.5 Adaptive Classifier; 5.5.1 Failures Under Challenging Lighting Conditions; 5.5.2 Hybrid Intensity Averaging; 5.5.3 Parameter Adaptation; 5.6 Tracking and Search Minimization; 5.6.1 Tracking Considerations; 5.6.2 Filter Modelling and Implementation; 5.7 Phase-Preserving Denoising; 5.8 Global Haar-Like Features; 5.8.1 Global Features vs. Local Features
5.8.2 Dynamic Global Haar Features5.9 Boosting Cascades with Local and Global Features; 5.10 Experimental Results; 5.11 Concluding Remarks; 6 Driver Inattention Detection; 6.1 Introduction; 6.2 Asymmetric Appearance Models; 6.2.1 Model Implementation; 6.2.2 Asymmetric AAM; 6.3 Driver's Head-Pose and Gaze Estimation; 6.3.1 Optimized 2D to 3D Pose Modelling; 6.3.2 Face Registration by Fermat-Transform; 6.4 Experimental Results; 6.4.1 Pose Estimation; 6.4.2 Yawning Detection and Head Nodding; 6.5 Concluding Remarks; 7 Vehicle Detection and Distance Estimation; 7.1 Introduction
7.2 Overview of Methodology7.3 Adaptive Global Haar Classifier; 7.4 Line and Corner Features; 7.4.1 Horizontal Edges; 7.4.2 Feature-Point Detection; 7.5 Detection Based on Taillights; 7.5.1 Taillight Specifications: Discussion; 7.5.2 Colour Spectrum Analysis; 7.5.3 Taillight Segmentation; 7.5.4 Taillight Pairing by Template Matching; 7.5.5 Taillight Pairing by Virtual Symmetry Detection; 7.6 Data Fusion and Temporal Information; 7.7 Inter-vehicle Distance Estimation; 7.8 Experimental Results; 7.8.1 Evaluations of Distance Estimation; 7.8.2 Evaluations of the Proposed Vehicle Detection