Car model recognition using Image Processing

Published on 11 May 2021. Written by Vinay Raj

Advancement in image processing, machine learning, and automation had led to the improvement of the intelligent transport system and this intelligent transportation system helps us to get more information and make things easy for us. An example of that this project. In this project, we will be using image processing techniques and will try to find the car maker and model number from a CCTV video or an image, etc… not only car you can use the same method to train the system using any image and get the results according to the user requirements.

Read more..

SLNOTE


Project Description

The main aim of this project is to make use of image processing and detect the car model and manufacturer. The input of the project may be a video or and real-time image file, it must process and give the desired output the user wants i.e. the model and manufacturer of the car. This is done by using machine learning and image processing technology. For example, if you give a Honda city car video to the system it will process and show the manufacturer “Honda” and model as “City”. 

Numerous methods of image processing techniques are used like SIFT, SURF, HOG, correlation, edge detection, etc… even machine learning algorithms like CNN, DNN, SVM classifiers are used.  

Concepts Used

• Scalar invariant feature transform
• Speeded up robust features
• Histogram of oriented gradients
• Convolutional Neural Networks
• Deep Neural Networks

Software and hardware components

• MATLAB

SLLATEST
Project Implementation 

The first step done in this project is to extract the information from the image or video feed. This is done by the above-mentioned image processing techniques. Then the images are keenly processed for the extra features that will detect the model of the vehicle. Moreover, you can even include information related features like logo, etc… which will help you in recognizing the model and manufacturer easily. The project can be used in the toll plaza, Traffic monitoring, security surveillance, etc.… 

The working process of the project as follows:

1. The image of the vehicle or video feed is loaded in the system
2. The identifiable features are extracted by the image processing techniques.
3. The BOF and SURF are used for obtaining the result of the image.
4. These extracted images are checked with the BoSURF data which is used to train the system.
5. Here SVM is used to check the accuracy of the image with the manufacturer and model of the car.

We can use any datasets for this project or also use different machine learning and image processing methods as well. You can even use the same method for the recognition of the number plates in the car also. To be at the pace of the development present today, you should have good practical skills and skills from the multidisciplinary subjects that will also be helpful for you in achieving your dream project or dream job.

SLDYK
Kit required to develop Car model recognition using Image Processing:

• Camera

Technologies you will learn by working on Car model recognition using Image Processing:

• Image Processing
• Car recognition

Any Questions?