MINI PROJECT
Helmet detection and license
plate number recognition
GROUP MEMBERS :
[Link] -411419205015
[Link] -411419205021
[Link] -411419205023
GUIDED BY :
Ms. S. Kanmani Jebaseeli
Assistant Professor / IT Dept
ABSTRACT :
•Nowadays, road accidents are one of the major causes that lead to human death.
•Motorbike accidents can cause severe injuries. The helmet is important for every motorcyclist.
However, many fail to conform to the law of wearing helmets.
•Here is the software using CNN, to recognize the motorbike drivers who are not obeying the helmet
law.
•The system consists of motorbike detection, helmet vs no helmet classification, and motorbike
license plate recognition.
Continue…..
• The motorbikes are scanned using the feature vector HOG.
• Once the motorbike is detected, by CNN, it identifies whether the motorcyclist is wearing a
helmet or not.
• If the motorcyclist is identified as not having a helmet, then the license plate of the motorcycle
is detected using Tesseract OCR.
EXISTING SYSTEM :
• Circle arc detection technique supported Hough remodel. They applied it to find the presence of
helmet that did not offer correct result.
• Combination of image process and Optical Character Recognition to find vehicle variety plate
underneath totally different background however it's worked on static i.e. non-moving Images in
Malaysia.
PROPOSED SYSTEM:
• In this project, we are detecting whether a two-wheeler rider wearing a helmet or not if he is not wearing a
helmet then we are extracting the number plate of that two-wheeler.
• To extract the number plate we have the YOLO CNN model with some train and test images and if you want to
add some other images then send those images to us so we can include those images in the YOLO model with
annotation to extract the number plate of those new images.
MODULES :
Image Acquisition: Input of the proposed system is an image or an image sequence
captured by a general CCD camera
Recovering: With the extracted features, several plate candidates can be selected.
License Plate Verification: Once a set of region candidates has been selected, the confidence
of each candidate for being a license plate is verified according to the geometrical properties
and the number of characters appearing in this candidate.
Feature Extraction Using Morphological Operations In the past, color and vertical edges
are two common features used to detect the locations of license plates.
Preprocessing It is well known that image intensities are very sensitive to different lighting
variations.
Features for License Plate Detection In this paper, we use several morphological operations
to find the high contrast area as an important feature to detect license plates.
License Plate Segmentation After labeling, a set of potential license plates can be extracted
from the images processed by morphological operations.
Back Propagation Neural Network BP (Back Propagation) Neural Network is a supervised
neural network, with three layers input layer, output layer and hidden layer [15].
Learning Vector Quantization Neural Network Learning Vector Quantization (LVQ) is a
hybrid network. In this paper LVQ uses supervised learning to form classification [16][17][18].
Candidate Extraction Let R denote the extracted plate region with the size w h × where w
and h are the width and height of R, respectively.
ARCHITECTURE:
SOFTWARE REQUIREMENTS:
• A powerful N-dimensional array object
• Sophisticated (broadcasting) functions
• Python 3.8
• Useful linear algebra, Fourier transform, and
random number potential
ADVANTAGES :
• A Non-Helmet Rider Detection system is developed where
a video file is taken as input.
• If the motorcycle rider in the video footage is not wearing
helmet while riding the motorcycle, and then here we are
uploading an image to identify the license plate number
of that motorcycle is extracted from image and displayed.
OUTPUT
CONCLUSION
The main purpose of this research paper is to investigate the possibilities of getting all the
details related to vehicle using license plate recognition. To improve the quality of the vehicles
images we can use fusion technique, then extract the license plate and isolate the characters
which are present on the number platisolatee and then identify the characters on the license
plate using artificial neural network. The experimental have shown that the ability to recognize
the characters correctly on license plate by neural network has probability of 95% even in
presence of noise with 50% density.
This proposed approach of LPR can be implemented by police to detect speed violators, no-
parking zone voilators, parking areas, highways, bridges or tunnels. Also the prototype of the
system is going to be integrated and tested as part of the sensor network being developed by
Ayers Island, LLC as part of their Intelligent Island system.
In this paper, the Automatic Number Plate Recognition system using vehicle license plate
is presented. This systemsystem use image processing techniques for recognition of the
vehicle from the database stored in the computer. The system works satisfactorily for wide
variation of conditions and different types of number plates. The system is implemented
and executed in Matlab and performance is tested on genuine images.
