Vehicle Detection using Camera
Vehicle Detection Using Cameras for Self-Driving Cars |
Using machine learning and computer vision I create a pipeline that detects nearby vehicles from a dash-cam.
why an Opensea Clone Script might be your perfect match.pdf
Vehicle Detection using Camera
1. Mid Semester Presentation 2018
Vehicle Detection using Camera
Babu Banarasi Das University
Presented By:
Shubham Agrahari
Ankita Kumari
2. Problem Statement
• When driving a car, two very important things to do is to…
a) stay in your lane.
b) avoid other cars.
• Amongst the top causes for road accidents the most prominent ones are due
to distracted driving, recklessness and weather adversities.
• Due to road accidents we are making the vehicle detection to control these
types of accidents we are making this project. Through vehicle detection the
car will automatic control the car speed and we will be safe.
3. Planning statement
Task 4 Jan.-
30 Jan.
31 Jan.-
9 Feb.
10 Feb.-
12 Mar.
13 Mar.-
16 Apr.
17 Apr.
22 Apr.
23 Apr.-
28 Apr.
Develop Project
proposal
27 Days
Analysis 10 Days
Designing 30 Days
Coding 34 Days
Unit Testing 5 Days
Implementation 5 Days
4. Introduction
• Vehicle detection cars have been a long-lasting dream of robotics
researchers. Vehicle detection promise to bring a number of
benefits to society, including prevention of road accidents.
• Detection of road obstacles is an important task in the intelligent
transportation.
• In today’s time in multinational company’s the machine’s do many
work’s instead of humans.
• In these machine the artificial intelligence has been use through
which they work.
• It is the part of self driving car in this project we are using object
detection.
5. • The data we are going to process to achieve vehicle detection are
images taken from a camera embedded in a moving car.
6. Objective
• The objective of our project is to design features like vehicle & lane
detection for a vehicle.
• The challenge was to create an algorithm that detects other vehicles
on the road, using video acquired using a front-face camera
• An image of the vehicle contains a digital image of the vehicle to
validate the platform for the detection of a vehicle's convenience.
• A camera will be set on the vehicle for image processing algorithm
the camera will detect the vehicle’s convenience and then it will alert
the driver to control the vehicle speed so that people will safe from
the accidents.
7. • For detecting vehicles we need to know what differentiates them from
the rest of the image also captured by the camera. Colors and gradients
are good differentiators but the most important features will depend on
the appearance of the objects.
8. Document convention
• Font Family
• Times New Roman
• Algerian
• Font Size
• Heading –32
• Paragraph-20
• Font Colour
• Content-Black
9. Minimum Hardware Requirement :
• System Processor : 1Ghz or more
• System RAM : 2Gb or more
• Others : Webcam, High speed Internet
HARDWARE INTERFACE
10. Software Requirements :
• Library : Numpy, Scipy, Pandas, jupyter, orange, etc.,
• Operating system : Windows 10 or Linux
• IDE : Spyder or Pycharm
• Database : SQLite
• Interpreter : Python with Anaconda package
• Documentation tools : MS-word, MS-PowerPoint
SOFTWARE INTERFACE
16. • INTERFACE MANAGER
To develop a platform for designing a set of effective features
to be used in vehicle detector.
• REQUIREMENT INSTALLER
In this is Function that Manage all calculation of a function
of gradient of image to validate the function is a sum of an
absolute value of the log gradient over all pixels in the first
masked sub domain.
MODULES OF THE PROJECT
17. • FILE MANAGEMNET SYSTEM
It is a system that is responsible for controlling or administering or
organizing a group of files.
• DATABASE MANAGER
Manages all different type of data stored in a single place to detecting the
vehicle.
18. • CAMERA MANAGER
In this project, your goal is to write a software pipeline to detect
vehicles in a video (start with the test_video.mp4 and later
implement on full project_video.mp4)
19. • VIDEO STREAM ANALYSER
In this project our goal is to write a software pipeline to detect vehicles
in a video ( start with the test video.mp4 and later implement on full
project video.mp4)
20. • PREVIEW SYSTEM
In this vehicle detection , we detect in a video pipeline,
potential boxes, via a sliding window boxes and it will be
preview in the users system.
21. •TRANINIG SYSTEM
First, we need to get training data(cars). we can get car images from GTI
vehicle image database, Over 1500 images per each is good for this project.
• TEST SYSTEM
To test all the subordinate of system hypotheses of a vehicle’s appearance
• ALGORITHM & MODEL SELECTION
To increase or decrease the performance of the software using the
algorithm and model
23. Functional Requirement
Sr.No. Req. No. Priority Requirement
1. R1.1 R Open Camera
2. R1.2 O Watch preview
3. R1.3 C Train Dataset
4. R1.4 O Check CSV
5. R1.5 O Change Settings
R – Required ; C- Conditionally Required; O-Optional
24. Column Name Data Type Size Constraint Description
ID Integer 10 Primary key Unique id
File Name varchar 50 _ Name of file
Created on Date _ _ Creation Date
Updated on Date _ _ Updation Date
Database Schema..
Video message
25. Conclusion
• This new vehicle detection method is based on association notion
described above. In order to select more reliable features, the corner
detector is used. Based on horizontal and vertical projection of the
associated edge points, the focused sub-region is selected as region of
interest.
• We will validated the efficiency of our method, and then we will show
that this method is capable to work in real time. In the future, we plan
to improve our vehicle detection method, which will be tested to detect
much more.
• In future the vehicle will be computer with tiers.