Avoiding collision accidents is becoming more and more an important topic in the research
field of driver assistant systems. Especially for vision-based detection systems there are
various approaches, which are built upon many different methods.
This thesis deals with the avoidance of pedestrian accidents, caused by Blind Corner
view problems. The presented approach comprises a pedestrian detection subsystem, which
is part of a large camera system framework covering observation of the car environment.
Based on a Blind Corner Camera and a neural network classification method, research in
this thesis is focused on two aspects: detection improvement and danger level estimation.
Since vision-based classification methods usually are still not able to yield perfect results,
because of the complexity of this task, the detection result has to be improved by
preprocessing and post processing. In this work, first, effects of image enhancement
methods on detection are tested as preprocessing methods and, secondly, a new approach
for a simple tracking and estimation strategy is presented, which improves detection in the
way of a post processing method. Finally, information from tracking and prediction is used
to estimate a danger level for pedestrians, which provides information about how collisionprone
the current situations is.
Table of Contents
1 Introduction
1.1 Background
1.2 An Approach for Blind Corner Pedestrian Detection
1.2.1 Pedestrian Detection
1.2.2 Blind Corner Problem
1.3 A Detection and Estimation System for the Blind Corner Problem
1.3.1 System Description
1.3.2 Basic Components
1.3.3 Detection and Estimation
1.4 Outline of the Thesis
2 Basic Components
2.1 Blind Corner Camera
2.1.1 General Description
2.1.2 Geometric Information and field of view
2.2 Pedestrian Detection Method
2.2.1 Vision-Based Pedestrian Detection Methods
2.2.2 Neural Networks (NN)
2.2.3 Pedestrian Detection with a Convolutional Neural Network
3 Image Preprocessing
3.1 Detection Problems
3.1.1 CNN Testing
3.1.2 Evaluation of Test Results
3.1.3 Problem Summary
3.2 Image Padding
3.2.1 Method
3.2.2 Testing
3.2.3 Evaluation
3.2.4 Summary
3.3 Image Enhancement
3.3.1 Approaches
3.3.2 Methods and Testing
3.3.3 Evaluation
3.3.4 Summary
4 Tracking and Estimation
4.1 Tracking
4.1.1 Model and Approach
4.1.2 Testing
4.1.3 Evaluation & Improvement
4.1.4 Summary
4.2 Prediction
4.2.1 Model and Approach
4.2.2 Testing
4.2.3 Evaluation
4.2.4 Summary
4.3 Estimation
4.3.1 Model and Approach
4.3.2 Testing
4.3.3 Evaluation
4.3.4 Summary
5 Conclusion and Future Work
5.1 Conclusion
5.2 Future Work
A Algorithms
A.1 Image Enhancement Methods
A1.1 Laplacian Filter
A1.2 Averaging Filter
A1.3 Median Filter
A1.4 Gamma Correction
A1.5 Histogram Equalization
A.2 Regression Analysis: Least Squares Method (LSM)
Objectives and Thematic Focus
The primary goal of this thesis is to develop a driver assistance system that enhances road safety by identifying pedestrians in the "Blind Corner" area—a region often occluded from the driver's direct field of view. The work focuses on the research question of how to effectively detect and monitor vulnerable road users in challenging blind spots using vision-based methods and machine learning.
- Pedestrian detection using Convolutional Neural Networks (CNN)
- Image preprocessing techniques including padding and enhancement to improve detection accuracy
- Robust tracking of detected pedestrians across image frames
- Prediction strategies to handle temporary detection failures
- Estimation of collision danger levels to inform potential warning systems
Excerpt from the Book
3.1 Detection Problems
Before dealing with preprocessing there has to be outlined what, respectively where the specific problems of the system are. Thus, this section gives an account of detection problems and weaknesses of the CNN and some camera related difficulties in order to have a basis for thinking about solutions for these problems.
3.1.1 CNN Testing
Before the testing procedure is explained, it is important to mention that the CNN, which is used in this work as a basic component, is trained with features of the lower body (see also section 2.2.3). CNN training itself is not part of the thesis and thus shouldn’t be considered, except for the feature definition, since it is important for the evaluation procedure.
Recording and Capturing
The first step, which had to be made for CNN tests, was to record image material with the blind corner camera. Therefore, several scenes were recorded in outside areas of Tokyo. In order to make the later evaluation easier, most of the scenes were taken of predefined actions, such as approaching to the camera, walking away, crossing persons and a predefined amount of persons (e.g. approach of one single person).
For recording the camera was either fixed at a test car on top of the bumper or placed on a camera stand at bumper height. The height level of the camera to the street was always between 55cm and 65cm in these scenes. For first recordings, the camera was always adjusted with a water level. However, this was changed for later recordings to an adjustment parallel to the road, since it proved that many roads are tilted by some degrees.
Summary of Chapters
1 Introduction: Provides an overview of the motivation for pedestrian safety in blind corner situations and defines the scope of the proposed detection and estimation framework.
2 Basic Components: Introduces the Blind Corner Camera hardware and details the fundamental concepts of Convolutional Neural Networks used for pedestrian classification.
3 Image Preprocessing: Analyzes specific detection challenges and presents methods like image padding and enhancement to improve the input data quality for the CNN.
4 Tracking and Estimation: Describes the development of tracking algorithms, prediction functions, and the danger level model used to evaluate traffic situations.
5 Conclusion and Future Work: Summarizes the findings of the research and suggests potential future improvements, such as the use of stereo cameras or more advanced tracking filters.
Keywords
Blind Corner Camera, Pedestrian Detection, Convolutional Neural Network, CNN, Image Preprocessing, Image Padding, Image Enhancement, Tracking, Prediction, Danger Level Estimation, Driver Assistance Systems, Computer Vision, Threat Analysis, Feature Extraction, Collision Avoidance.
Frequently Asked Questions
What is the core purpose of this research?
The research focuses on preventing pedestrian-vehicle accidents that occur due to "Blind Corner" visibility issues, using a camera-based system to detect, track, and estimate danger levels for pedestrians.
What are the primary technical focus areas?
The work centers on using Convolutional Neural Networks (CNNs) for detection, refining this through preprocessing (padding/enhancement), and implementing post-processing tracking and threat assessment.
What is the ultimate goal of the system?
The system aims to categorize road scenes into danger levels, enabling a future driver assistance framework to warn drivers of potential collision risks in areas that are otherwise difficult to monitor.
Which scientific methodology is employed?
The approach utilizes an offline vision-based pipeline involving image acquisition, neural network-based object detection, custom tracking algorithms, least squares-based prediction, and fuzzy-logic inspired danger estimation.
What is covered in the main body of the work?
The main body examines component selection, preprocessing for CNN performance optimization, and the implementation of robust tracking, prediction, and situation analysis.
How are the key features described?
The key features include Blind Corner Camera, CNN, Image Padding, Tracking, Prediction, and Danger Level Estimation.
How does image padding solve the "margin problem"?
The CNN convolution process loses information at the edges of an image. Padding adds redundant pixels at the borders, allowing the network to detect pedestrians even when they are near the margin of the camera's field of view.
Why are "Tracking" and "Prediction" necessary?
Tracking links detections over time to analyze object behavior, while prediction fills gaps in detection—caused by system instability or temporary occlusion—to maintain a continuous assessment of a pedestrian's movement.
- Citation du texte
- Bastian Hartmann (Auteur), 2006, Vision-based pedestrian detection and estimation with a blind corner camera, Munich, GRIN Verlag, https://www.grin.com/document/176752
