First of all we express our heartiest thanks and gratitude to the almighty God for his divine blessing which enables us to successfully complete the final year project. We are truly grateful and express our sincere gratitude to Nazmun Nessa Moon, Assistant Professor, Department of CSE, Daffodil International University, Dhaka. Deep knowledge and great interest of our supervisor in the field of "Embedded Systems and Digital Image Processing" to carry out this project.
Syed Akhter Hossain, Head of CSE Department, for his kind assistance in completing our project and also to other faculty members and staff of CSE Department of Daffodil International University. We would like to thank our entire coursemate at Daffodil International University who participated in this discussion while completing the course. Our project titled “An Embedded System for Driver Drowsiness Detection” aimed to develop a real-time embedded system that detects driver drowsiness.
When closing the eyes for a certain period of time exceeds the threshold value, the driver is determined to be drowsy. This system has several OpenCV libraries and a dlib library and the method we use is eye aspect ratio and facial landmark localization. To locate the facial structure, we use facial landmarks and using the eye aspect ratio we can determine the closure value of the eye.
We use various software like Raspbian Operating System, Python IDLE, CMake, Advance IP Scanner, MobaXterm to develop this system.
INTRODUCTION
- Introduction
- Motivation
- Objectives
- Expected Outcome
- Report Layout
With this in mind, we wanted to create a cost-efficient embedded system that everyone can use. To build an embedded system where the camera can monitor the face, detect the eye area and calculate the eye aspect ratio.
Background
Requirement Specification
Methods and Library
Implementation and Testing
Conclusions and Future Scope
BACKGROUND
Introduction
Related Works
- Mercedes-Benz (Attention Assist): Mercedes unveiled a complex structure where driver fatigue is being monitored using various high-end sensors named
If it is drowsy, it will vibrate, beep and flash to alert the driver.
Comparative Studies
REQUIREMENT SPECIFICATION
- Work-Flow Diagram
- Hardware Specification
- Camera
- Audio Module
- Software Requirements
- Python 3 IDLE
- Visual Studio (For testing)
- Advance IP Scanner
- MobaXterm
- CMake
- Introduction
- Facial Landmark
- Image Localization
- Detecting key Structure
- Eye Aspect Ratio
- SciPy Package
- Dlib Library
- PyGame Module
Because the project is in an embedded system, it must be a combination of hardware and software. The Raspbian operating system is the primary operating system of Raspberry Pi singleboard computers. CMake is an open source, cross-platform, extensible software application that helps build C/C++ projects on any platform.
If there is an image, the shape predictor focuses on locating key points of interest with the shape. We can locate different facial recognition detectors and label the next facial region. The method by which we calculate the aspect ratio of the eyes was introduced by Tereza Soukupová and Jan Čech in their article Real-Time Eye Blink Detection using facial landmarks [2].
The equation derived from the work done by Tereza Soukupová and Jan Čech in their paper is called the eye ratio equation. Dlib is one of the essential open source cross-platform tools for implementing component-based, free software. Because this library written in C++, we used CMake to convert the library with python links.
We mainly used it to play sound as an alarm when camera detects driver drowsiness.
IMPLEMENTATION AND TESTING
- Introduction
- Hardware Setup
- Raspberry Pi 3 Model B
- Camera Setup
- Library Installation
- SciPy Package Installation
- Installing Dlib Library
- Installing Imutils Package
- PyGame Module for Sound
- Algorithm Implementation
Reset your swap file size, boot options and memory partitioning The six steps were followed when we installed dlib on our Raspberry pi. If the function's return value is approximately constant, it is considered eyes open. This means that if the return value from the function is lower than the threshold value, it will consider the frame as a drowsy detected frame.
If this happens consecutively (more than 5 seconds), the system will treat the condition as sleepiness and play an alarm sound. If the value of the next 5 frames is equal to or lower than the threshold value, the system will detect drowsiness.
CONCLUSION AND FUTURE SCOPE
Obstacles and Discussion
Limitation
Use of glasses: If the user wears glasses, it becomes difficult to determine the condition of the eye. Camera distance: For best results, the user's face should be close to 100 cm.
Future Work
Rafael Hossen ID This report is presented in partial fulfillment of the requirements for the degree of. Rafael Hossen of the Department of Computer Science and Engineering, Daffodil International University, has been accepted as satisfactory for the partial fulfillment of the requirements for the degree of B.Sc. In recent decades, insight has increased into the factors that cost more than 1.2 million lives each year as a result of traffic accidents.
Driver fatigue is becoming one of the most important factors in preventable traffic accidents. FIGURES ©Daffodil International University PAGE viii Figure 3.1 Workflow diagram Figure 3.2 Raspberry Pi 3 Model B Figure 3.3 Raspberry Pi camera module Figure 4.1 Visualization of the 68 facial recognition coordinates Figure 4.2 Eye aspect ratio comparison Figure 4.3 Open and closed eyes with face. All these accidents are not only due to driver fatigue but are also one of the important factors.
Conclusions and Future Scope It gives an overall idea, limitations of the project also the future aspect of this project. Studying most of the system, they use different sensors that detect and analyze driving behavior. Requirement Specification 3.1 Workflow Diagram Technically, our project is about detecting the human eye aspect ratio.
The list of hardware we use is below: 3.2.1 Raspberry Pi 3 Model B One of the most commonly used microcontrollers at the lowest cost. To better understand the ports of Raspberry pi 3 model B are given below - Figure 5.1 Raspberry pi 3 model B ports 5.3 Library installation Installation process of the library we used below - 5.3.1 SciPy package installation For the installation we used Python and pip installed our system. During this project we also collected information about road safety and the impact of road accidents on our society.
So, a high processor is needed which will ultimately increase the cost of the product. Some of the features are given below- • Processor speed will be increased for faster action. An audible alarm is not enough to avoid the accident, so this system must be incorporated into a motorized system that can control the speed of the vehicle.
