Number Plate Recognition System

Project Title

Number Plate Recognition System

Project Area of Specialization Artificial IntelligenceProject Summary

Our Product Is Car Number Plate Recognition System and this will be the first release of our product, our basic scope is to identify registered cars that have valid entries and to make sure that any unregistered or illegal car can be caught at the spot. This can be implemented in housing societies, parking lots, government offices.

Project Objectives

Number plate recognition is realized by acquiring images of either the front or the rear of vehicles with cameras and then by image processing to identify license plates. To help a society or a compound have their cars parked only in their area and we can also can help the governing authorities find the illegal or unregistered cars present in the city. For this we will use different methods in which first one is Number plate identification and localization in this segment the visual scene is improved with image processing using Convolutional Neural Network(CNN). Second one is Optical Character Recognition in which text is transferred into encoded text information.

Project Implementation Method

First of all the system will be able to detect the number plate and it will tell whether it is registered or not and will display the details to the monitoring guard’s computer. The image will be scanned by the system and then displayed on the screen.
Secondly there will be a CRUD in the database when the administrator will login he will be able to add,edit and delete.The whole list of details will be visible to the administrator.

We will be working with Python using Keras which will be used for training our Convolutional Neural Network(CNN) and TensorFlow Library which is used for processing and running our model, the platform will be JUPYTER and our database will be on FIREBASE. The hardwares will include the camera, a computer, lights.

Benefits of the Project

The benefits of the projects are that the cars which are not registered will not be able to enter the area, hence proving that any unauthorized person cannot enter the society, parkinglot.

The workforce is less only one guard will be enough because the system wis already doing the authorized numberplate work.

Technical Details of Final Deliverable

Feature 1 : 

Entering the right username and password for the login should redirect to the dashboard. If the entry is written wrong. It shows an error message of wrong input and doesnt direct you to the dashboard. Same for managing profiles. The number plate should have 3 digits and 3 characters. 11 numbers should be for a phone number. Name should be in characters.

Feature 2 : 

Having a database is important for saving the customer profiles and for confirmation whether it is valid or not. The database we are using is firebase which can be accessed at all time.

Feature 3 : 

We will first get the image and then process it to RGB, then it will be converted to our specified length and width basically it will resize our image. It will draw rectangular box around the number plate where the rectangle edges are given by the dataset.

Feature 4 : 

We will send an image to the system to detect the number plate area in the image by using our model and trained weights, it will give us the rectangular box in the image where no. plate is located.

Feature 5 : 

The specified area where the rectangle is drawn i.e. on the numberplate will be cropped and then it will be converted in grayscale and after that it will go into MNIST and OCR extraction.

Feature 6 : 

We will get the individual characters for extraction and convert it into text.

Feature 7 : 

We will create a firebase connection that will compare the output text with the relevant stored information of the desired number plate. If it succeed than it will print success on the console.

Feature 8 : 

The guard/admin can see which car is entering the toll and if the numberplate matches the gate will open automatically and it will also display “Success” on the screen. When the barrier is opening it will also display “Opened” whether it is opened manually or automatically.

Final Deliverable of the Project HW/SW integrated systemCore Industry ITOther Industries Education , Security Core Technology Artificial Intelligence(AI)Other TechnologiesSustainable Development Goals Industry, Innovation and InfrastructureRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	28000
Camera	Equipment	1	20000	20000
SSD	Equipment	1	8000	8000