![References
[1] IEEE Vehicle Number Recognition system for automatic toll tax
collection-Shoaib Rehman Soomro; Mohammad Arslan Javed;
Fahad Ahmed Memon
[2] Vehicle Number Plate Recognition for automatic toll tax collection
Using IoT and Machine Learning by Aheraaz Tamboli and Shounak Sugave
[3] Automatic Toll Collection using Vehicle Number Recognition System
by P. Sai Kiran; Yalanati Ayyappa; Maddila Kavya Sree; Jaswanth Jampani;
Mounisha Raavi; Karthik Aluru
[4] Intelligent toll collection system for moving vehicles in India by
Rajeev Kumar Chauhan a, Kalpana Chauhan
[5] OpenCV: Automatic License/Number Plate Recognition (ANPR) with
Python by Adrian Rosebrock](https://image.slidesharecdn.com/ipprpbl-250720053938-71e7eff6/85/Image-Processing-and-Pattern-Recognition-ANPR-pptx-12-320.jpg)
Image Processing and Pattern Recognition ANPR.pptx
- 1. Automatic Toll System Using License Plate Recognition IPPR Project by : Reet Shinde 1032211433 Siddhi Amale 1032211692
- 2. Aims and Objectives The aim of this project is to design and implement an automatic toll collection system that uses licence plate recognition to identify vehicles and calculate tolls automatically. This system aims to streamline the toll collection process, reduce traffic congestion at toll booths, and improve overall efficiency. Objectives are: ● Utilise Existing Infrastructure: ● Dataset Creation: Single images for displaying working of code. ILP for model training (kaggle dataset) ● Implement Recognition Models and filtering techniques ● Evaluate Performance:
- 3. Literature Survey Automatic number plate recognition, the recent trend in current environment, various systems have already been developed by existing authors to detect the license plate as well as number plate using machine learning and deep learning algorithms. Internet of things (IoT) and some machine learning techniques for image processing has also contributed many classical systems. Modern machine learning approaches for pattern recognition in the field of license plate recognition is an effective method to improve the traditional OCR only approaches. Typical VNR System consists of four modules: image acquisition, license plate extraction, character segmentation, and character recognition. There are several common searching algorithms to locate vehicle license plate. Searching algorithm rely on color information. In this method a color search algorithm is used to extract the likelihood ROI in an image. These algorithms are usually fast but can detect only single colored standardized number plate. These algorithms are usually fast but can detect only single colored standardized number plate. High license plate extraction rate is achieved in based on vertical edging and mathematical morphology operations; because of having vertical edges in English characters & digits, they can be easily classified. Presently, there are several common algorithms for the segmentation of license plate characters, such as segmentation through dilation, template matching & projection analysis. In the segmentation through dilation, characters of number plate are dilated vertically for separating each character & smearing algorithm is used for finding character region.
- 4. Methodology Dataset: Dataset is divided into images and it’s labels. the image contains 943 train images and 181 validation test images. Similarly labels are categorized according to its respective image. Framework: When a frame is captured, it undergoes preprocessing: it’s converted to grayscale, then a bilateral filter is applied to reduce noise while preserving edges. Edge detection is subsequently applied to highlight contours. The program then identifies contours and examines the characteristic of a license plate. If a potential license plate region is detected, the area is cropped for OCR analysis. Using Tesseract OCR, the program reads and extracts text from the cropped license plate region, printing the detected text and drawing a bounding box around the license plate on the captured image for visual feedback.
- 5. YOLO v5 YOLO v5 (You Only Look Once, version 5) is a state-of-the-art, real-time object detection model designed to be fast, accurate, and relatively easy to deploy. It's highly popular in the field of computer vision for tasks like object detection, instance segmentation, and even some aspects of image classification. Key Features 1. Real-Time Detection: YOLO v5 is optimized for real-time object detection with high frame rates and fast processing times. 2. Efficiency: It offers good trade-offs between speed and accuracy and is suitable for deployment on resource-constrained devices. 3. Pre-trained Models: YOLO v5 has several model sizes (e.g., YOLOv5s, YOLOv5m, YOLOv5l, YOLOv5x) optimized for different hardware and performance needs. 4. Easy to Use and Extend: The PyTorch-based implementation is designed to be user-friendly and customizable. 5. Onnx Export and TFLite Compatibility: YOLOv5 supports export to various formats, making it deployable on platforms like TensorFlow Lite, CoreML, and others for cross-device compatibility.
- 6. Software: 1. Python (3.x) ● Description: The programming language used to implement the solution. ● Installation: Download from Python.org. 2. Tesseract OCR ● Description: An OCR engine to perform text recognition on the license plate images. 3. NumPy: For array handling and mathematical operations. 4. OpenCV: For image processing and video capture functionality. 5. imutils: For convenience functions, like resizing the frame 6. pytesseract: Python wrapper for Tesseract OCR.
- 7. Flow Chart
- 9. Predicted License Plates Actual License Plate
- 11. Conclusion
- 12. References [1] IEEE Vehicle Number Recognition system for automatic toll tax collection-Shoaib Rehman Soomro; Mohammad Arslan Javed; Fahad Ahmed Memon [2] Vehicle Number Plate Recognition for automatic toll tax collection Using IoT and Machine Learning by Aheraaz Tamboli and Shounak Sugave [3] Automatic Toll Collection using Vehicle Number Recognition System by P. Sai Kiran; Yalanati Ayyappa; Maddila Kavya Sree; Jaswanth Jampani; Mounisha Raavi; Karthik Aluru [4] Intelligent toll collection system for moving vehicles in India by Rajeev Kumar Chauhan a, Kalpana Chauhan [5] OpenCV: Automatic License/Number Plate Recognition (ANPR) with Python by Adrian Rosebrock