Number Plate Recognition System Using ANN (Machine Learning)

NPRS is Machine Learning based number plate recognition system with a trained (Feed-Forward) neural network that is capable recognize only Punjab (Pakistan) number plates.

• Tested on MATLAB R2014a

Setting up

• Download zip file
  
• Extract it in Matlab folder
  
• Add this to path
  
• Run NPRS.m
  
• Put number plate image directory in text box
  

GUI display

Dataset display

Dataset Metadata

Title: Alphanumeric Character Recognition Dataset

Description:

This dataset contains images of alphanumeric characters (0-9 and a-z) for text recognition tasks. Each character is represented by 100 samples, resulting in a total of 3600 samples. The images are grayscale with a resolution of 20x20 pixels, which have been flattened into 400-dimensional vectors.

Content:

X: The feature matrix with dimensions (3600, 400). Each row corresponds to a 20x20 pixel image flattened into a 400-dimensional vector.

y: The label matrix with dimensions (3600, 36). Labels are one-hot encoded for 36 classes (0-9 and a-z), where each class has 100 samples.

Features:

X (3600, 400): Each row represents a flattened 20x20 pixel grayscale image of an alphanumeric character. Contains 3600 samples, with each sample having 400 features.

y (3600, 36): Each row represents the one-hot encoded label for the corresponding sample in X. Contains 3600 samples, with each sample having 36 features representing the alphanumeric classes (0-9 and a-z).

Classes:

36 classes representing the alphanumeric characters:

Digits: 0, 1, 2, 3, 4, 5, 6, 7, 8, 9

Letters: a, b, c, d, e, f, g, h, i, j, k, l, m, n, o, p, q, r, s, t, u, v, w, x, y, z

Usage:

This Dataset is primarily made for “Pakistan (Punjab) Number Plates Recognition specifically for German-style number plate font “ but it is not limited, it is suitable for tasks such as:

• Alphanumeric character recognition
• Optical character recognition (OCR)
• Machine learning and deep learning experiments for text classification

Source:

This dataset was created for research and educational purposes to facilitate the development and evaluation of text recognition algorithms.

Acknowledgements:

If you use this dataset in your research or projects, please cite the source appropriately.

NOTE

Simply load csv files mentioned below for loading data in dataframes

• X_NPRS_dataset_samples.csv
• y_NPRS_dataset_labels.csv

or load AlphaNumaric.mat for loading data in MATLAB