Project proposal

1. HELLO!
Here 1Crore Projects
I am here because I love to give presentations.
You can find me at @1CROREPROJECTS
1

2. HELMET AND NUMBER PLATE
DETECTION OF IMAGES USING
PYTHON

3. AIM OF PROJECT
3
The aim of this project is to detect the motorcyclist wearing helmet and capturing the name plate of
the vehicle Under violation of motor rules and informing to the department of transport.

4. ABSTRACT
4
 In today’s world, the increasing use of Motorcycles has prompted increment in road accidents and injuries. Helmet
not used by the motorcycle rider is one of the major cause. Currently, one procedure is to physically check use of
helmet at the pavement junction or through the CCTV footage video, which requires human intervention to detect
motorcyclists without helmet. The proposed framework presents a computerization machine structure to distinguish
the motorcycle rider with or without helmet from images.
 And these process will be executed with YOLO v7 modules.

5. INTRODUCTION
5
Helmets are one of the essential key safety factor for motorcyclists. Unfortunately their use is not increased particularly
where the rules for not wearing helmet is strict. The goal of this procedure is to provide an automated system approach to
detect motorcyclists with and without helmet and their corresponding number plates respectively. India accounts 15 percent
of overall traffic deaths typically, with world’s 1 percent of automobiles, according to the World Bank. Space acquired by
private vehicles has shown increase from 59 percent to 77 percent in the past twenty years, while the road acquired by buses
has marginally reduced from 6.2 percent to 2.2 percent, a report by the Mumbai Environmental Social Network. To increase
the helmet usage, Government of India has proposed, Various Penalties under Motor Vehicles (Amendment) Bill – 2019.
According to section 194D a motorcyclist not wearing helmet will be fined rupees 1000 and disqualification of license for 3
months. Currently, the traffic Police monitor motorcyclists manually whether they are with or without helmet. To check
manually is inadequate, time-consuming as well as there are limitations of human errors. Also, in major semi-urban and rural
areas CCTV surveillance based methods are not automated and require human efforts. The increasing amount of motorcycles
and accidents caused by not wearing helmet, has led to increase in research field of road safety surveillance department.

6. EXISTING SYSTEM
6
Existing system and tested automated system for detection of motorcycle riders not wearing helmet was done by
Chiverton. The system uses SVM classifier trained on features derived from image data near head region of
motorcyclists. The features selected capture the shape and reflective property of helmets where the top half of helmet
surface is found to be brighter than the bottom half of the helmet. It also takes into account the circular arc-shape of
helmet. The system uses circular arc detection technique based on Hough transform. The main shortcoming of this
approach is that it leads to a lot of misclassification, as some objects which look similar to helmet get classified as
helmets and some helmets which are different do not get classified as helmets. Another limitation is that it does not
first identify motorcyclists in the frames, which should have been done, since helmet is only relevant in case of
motorcyclists.

7. DISADVANTAGE
7
▰ A small amount of data, may over fit or under fit.
▰ Low accuracy

8. PROPOSED SYSTEM
The YOLO V7 Model was used for real-time detection of Number plate for a non-helmeted motorcyclist. In the system,
the video frame is taken as the input, and the expected output is the localized Number plate for a non-helmet
motorcyclist. In our approach, the system checks the presence of a helmet on the motorcyclist. If the motorcyclist is
without a helmet, the Number plate is extracted so that it can be given to Automatic number plate recognition (ANPR)
technology for recognition of characters and fine the defaulters.
8

9. ADVANTAGES
▰ Achieving good accuracy for detection of motorcyclists not wearing helmets.
▰ Automatic detection of motorcycle riders without helmet from CCTV video and automatic retrieval of vehicle
license number plate for such motorcyclists.
9

10. SYSTEM ARCHITECTURE
10

11. SYSTEM MODULE
 Module 1: Dataset (video input)
 Module 2: Data Pre-processing
 Module 3: Model Implementation
 Module 4: Result
11

12. DATASET COLLECTION
▰ A dataset (or data set) is a collection of data, usually presented in tabular form. Each column represents a
particular variable. Each row corresponds to a given member of the dataset in question. It lists values for each
of the variables, such as height and weight of an object. Each value is known as a datum.
▰ In this project we collected the vehicle (motorcycle) datasets which are categorized as wearing helmet and
without wearing helmet.
12

13. DATA PRE-PROCESSING
• Data preprocessing is a data mining technique
• It is used to transform the raw data in a useful and efficient format.
• The collected data it goes through a rigorous process of data cleaning and data processing.
1. Data cleaning: This process involves cleaning the noise from raw data and making up the missing parts
to get useful data.
2. Data transformation: This process helps to transfer the data to make suitable for mining process.
3. Data reduction: Data mining takes huge time to process due to large data, in order to avoid such things
data reduction technique.
13

14. MODEL IMPLEMENTATION
• TRACKING:
Video is collection of frames. The negligible time gap between frames makes the stream of photos looks
like flow of scenes. When designing algorithm for video processing. Videos are classified into two classes. Video
stream is an ongoing process for video analysis. The processor is not aware of future frames. Video sequence is
video of fixed length. All the consecutive frames are obtained prior to processing of current frame. Motion is
distinct factor that differentiates video form frame. Motion is a powerful visual Que. Object properties and action
can be realized by noticing only sparse points in the image.
14

15. MODEL IMPLEMENTATION
• DETECTING OBJECT:
Frames are captured from camera at regular intervals of time. Difference is estimated from the consecutive
frames. Optical Flow This technique estimates and calculates the optical flow field with algorithm used for optical
flow. A local mean algorithm is used then to enhance it. To filter noise a self-adaptive algorithm takes place. It
contains a wide adaptation to the number and size of the objects and helpful in avoiding time consuming and
complicated pre-processing methods. Background subtraction (BS) method is a rapid method of localizing objects
in motion from a video captured by a stationary camera. This forms the primary step of a multi-stage vision system.
This type of process separates out background from the foreground object in sequence in images.
15

16. RESULT
• Once the detection is made, there are two possible outcomes whether the motorcyclist wear helmet or not using
Boundary boxes. And also detect the number plate of the motor cycle.
16

17. SOFTWARE REQUIREMENTS
▰ Operating System : Windows 7,8,10 (64 bit)
▰ Software : Python
▰ Tools : Anaconda (Jupyter notebook IDE)
17

18. HARDWARE REQUIREMENTS
▰ Hard Disk : 500GB and above
▰ RAM : 4GB and above
▰ Processor : I5 and above
18

19. CONCLUSION
In conclusion, the proposed system based on YOLOv7 for real-time detection of helmets and number plates on
vehicles is a significant contribution to enhancing road safety and enforcing traffic regulations. The use of a
pre-trained YOLOv7 model fine-tuned on a custom dataset results in high accuracy and speed. The proposed
system has the potential to be deployed in traffic monitoring and surveillance systems to improve road safety.
The experimental results demonstrate that the proposed system outperforms existing state-of-the-art methods
for helmet and number plate detection. Overall, the proposed system is a promising approach for real-time
object detection in traffic-related applications.
19

20. REFRENCES
▰ N. Dalal and B. Triggs, "Histograms of oriented gradients for human detection," 2005 IEEE Computer Society
Conference on Computer Vision and Pattern Recognition (CVPR'05), San Diego, CA, USA, 2005, pp. 886-893
vol. 1.
▰ J. Chiverton, "Helmet presence classification with motorcycle detection and tracking," in IET Intelligent
Transport Systems, vol. 6, no. 3, pp. 259-269, September 2012.
▰ R. Silva, K. Aires, T. Santos, K. Abdala, R. Veras and A. Soares, "Automatic detection of motorcyclists without
helmet," 2013 XXXIX Latin American Computing Conference (CLEI), Naiguata, 2013, pp. 1-7.
20

21. 21
THANKS!
Any questions?
You can find me at
Reach us – 1croreprojects@gmail.com
Contact / Whatsapp: 7708 150 152 / 9751 800 789 / 790 432
0834