TRAFFIC SIGN BOARD DETECTION AND RECOGNITION FOR AUTONOMOUS VEHICLES
Our project is an AI project which is based on some computer vision techniques which can recognize traffic signboards. To detect any traffic sign board we are building a Convolutional Neural Network trained model which is based on the GTSRB (German Traffic Sign Recognition Benchmark) database which
2025-06-28 16:29:51 - Adil Khan
TRAFFIC SIGN BOARD DETECTION AND RECOGNITION FOR AUTONOMOUS VEHICLES
Project Area of Specialization RoboticsProject SummaryOur project is an AI project which is based on some computer vision techniques which can recognize traffic signboards. To detect any traffic sign board we are building a Convolutional Neural Network trained model which is based on the GTSRB (German Traffic Sign Recognition Benchmark) database which has more than 40 classes and a total number of 50000+ images. Traffic Sign Recognition (TSR) is an advanced driver assistance system that recognizes and relays traffic sign information to drivers via the instrument panel, multimedia, or heads-up display. In the United States, TSR systems can recognize speed limit signs, do not enter signs, and stop signs. Some can even detect yield signs, though it depends on the model. TSR aims to help make drivers more aware and able to make better safer driving decisions.
Project ObjectivesThe main idea of our traffic sign recognition project is to reduce driver error which can cause major accidents on roads. Like if there is a signal ahead traffic signboard and the driver didn’t slow down his vehicle, our system will automatically detect the signboard which will be very helpful not only for the driver but for other vehicles too. In 2019, 26% of road accidents in the US happed when the driver was not aware of any traffic signboard. By 2025, it will be a standard feature in Russian vehicles to recognize the warning signboards to assist the driver and reduce accidents ration
Project Implementation MethodOur project is based on modern artificial intelligence (AI) which focuses on its objective of assisting drivers to get rid of accidents. We will download a dataset that has more than 50,000 images and is divided into 43 classes of the most important traffic signs. We are building Convolutional Neural Network in our project. Our project is written in Python which is a very perfect fit for computer vision projects. We will use Tensorflow, Keras, Matplotlib, and a few other neural network libraries/dependencies to build our trained model. In real-time, our project will recognize traffic signs and also show us the accuracy we are getting in our project.
Benefits of the ProjectThis project is utilizing artificial intelligence techniques which will be helpful for Automobile manufacturing industries.
Reduce accident.
Reduce Driver’s mistakes.
Technical Details of Final DeliverableOn printed pages, we will capture these pages in front of the camera to get the recognition algorithm output.
It will threshold images and convert them into greyscale to get better results.
After capturing the images it will show how much accurate our traffic sign is. It will be between 80 to 95 %.
Final Deliverable of the Project HW/SW integrated systemCore Industry ManufacturingOther Industries IT Core Technology RoboticsOther Technologies OthersSustainable Development Goals Quality EducationRequired Resources| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
|---|---|---|---|---|
| Total in (Rs) | 76000 | |||
| LED Electric Board/Camera | Equipment | 1 | 37000 | 37000 |
| Android Studio | Equipment | 1 | 20000 | 20000 |
| fram | Equipment | 1 | 9000 | 9000 |
| others | Miscellaneous | 1 | 1000 | 10000 |