Smart Guard For Automobiles

Smart Guard For Automobiles via Artificial Neural Network (ANN) is proposed due to increase in number of vehicles which is triggering an increase in more security and traffic problems. In addition human reliability is reducing day by day. So, we proposed this sys

Project Title

Smart Guard For Automobiles

Project Area of Specialization

Artificial Intelligence

Project Summary

Smart Guard For Automobiles via Artificial Neural Network (ANN) is proposed due to increase in number of vehicles which is triggering an increase in more security and traffic problems. In addition human reliability is reducing day by day. So, we proposed this system to increase efficiency in the field of security and control as the system will only allow the authorized vehicles to pass from a certain entry point. It works by first extracting license plate features using image processing algorithms, through special cameras installed at a certain distance from the entry point. The trained ANN will take decision according to the fetched data whether to allow the authorized vehicle to pass by opening the barrier and displaying on the screen via wireless connectivity or transmission.

Project Objectives

The main objective of this project is to keep monitoring the vehicles that are entering and exiting the building. It only allows authorized/registered vehicles to access in to the building. The system can be implemented on the entrance for security control of a highly restricted area like private organization, military zones or area around top government offices.

Project Implementation Method

Training Process:

Real Time Video: 

The proposed system will have the HD camera that continuously capturing the video/image of the vehicle which further sends for the detection.

Preprocessing:

In preprocessing step, we utilized LAB transformation and L channel which represents luminance and a, and b channels describes color component. After transformation, a luminance channel is extracted which is later utilized in the segmentation steps.

Segmentation:

In the segmentation step the processed luminance channel, extracted in the previous step, is converted into a binary image. Firstly, we utilized Otsu’s segmentation which is a clustering based algorithm. After performing Otsu’s segmentation, Operations of erosion and dilation is implemented. The purpose of erosion and dilation is to make digits in license plate more accurately visible. We subtract the erosion image from dilated image then 2D-convolution operation is performed on subtracted image, to make the character in the license plate more visible. After performing 2D-convolution operation, enhance the intensity value of segmented image. Lastly, we implemented some morphological operations to remove the extra regions to get the final segmented regions.

Features Extraction:

 This process will separate the number plate area from the vehicle.

Artificial Neural Network Training:

In this process the artificial neural network will train by learning features extracted from the image.

Testing Process:

Real Time Video, Preprocessing, Segmentation and Features Extraction will be same as in the Training process, further processes are:

Trained Artificial Neural Network:

After the completion of training process the ANN will be able to recognize the number plate and will forward the result to wireless module for further operations.

Wireless Module: 

It will send the instruction wirelessly to the dedicated controller to open the barrier or to generate a signal of unauthorized number plate. Also it will transmit the data to the screen which will monitor the car entrance time etc.

Benefits of the Project

• This system is more accurate and secure than any human guard.
• No unauthorized vehicle can cross the entrance.
• This system prevents thefts attempts.
• This project will maintain the overall data of vehicles entering and exiting time.
• Can be used as anti-theft device on the roads and streets by tracing the number plate.

Technical Details of Final Deliverable

It works by first extracting numbers of license plate using image processing, through ip cameras installed at a distance from the entrance. The trained ANN will take decision according to the data whether to allow the authorized vehicle to pass by opening the barrier and displaying on the screen via wireless connectivity or transmission.

Final Deliverable of the Project

HW/SW integrated system

Core Industry

Telecommunication

Other Industries

Security

Core Technology

Artificial Intelligence(AI)

Other Technologies

Sustainable Development Goals

Industry, Innovation and Infrastructure

Required Resources

If you need this project, please contact me on contact@adikhanofficial.com