Custom-Generated Synthetic Aadhar Card Dataset for Robust Identity Authentication Research

Published: 25 March 2024| Version 1 | DOI: 10.17632/wk6n8fhx3c.1
Contributors:
Syeda Bibi Javeriya,

Description

This dataset comprises 1000 synthetic images of fake Aadhar cards, meticulously crafted for research and development in the domains of face recognition and Aadhar card verification systems. Each image in the dataset is meticulously generated using a custom fake Aadhar card app, ensuring a diverse range of transformations and variations. For each entry, both front and back sides of the Aadhar card are provided, yielding a total of 200 unique templates. Despite the fictitious nature of the Aadhar card details, the dataset accurately emulates real-world scenarios, facilitating robust algorithm development and testing. Key Features: Image Variations: The dataset includes 1000 images, each with dimensions of 1188 pixels by 748 pixels. These images undergo various transformations such as hue saturation adjustment, contrast enhancement, blur effects, and scaling, simulating the diverse conditions encountered in real-world applications. Face Recognition: Researchers can leverage this dataset to develop and refine face recognition algorithms. By extracting facial regions from the Aadhar card images, algorithms can be trained to identify and verify individuals based on their facial features, enhancing the accuracy and robustness of face recognition systems. Aadhar Card Details Extraction: Using optical character recognition (OCR) techniques, developers can extract text information from the Aadhar card images. Despite the dummy details, this process facilitates the development of OCR algorithms tailored for Aadhar cards, extracting fields such as name, Aadhar number, address, and other pertinent details. Aadhar Card Verification: The extracted details enable the implementation of Aadhar card verification systems. These systems compare the extracted information against a database of genuine Aadhar card details to verify the authenticity of the card, contributing to enhanced security measures in various domains. In conclusion, this dataset serves as a valuable resource for advancing the capabilities of face recognition algorithms, OCR technology, and Aadhar card verification systems. Its synthetic nature allows for comprehensive testing under diverse conditions, fostering innovation in identity verification solutions and bolstering security measures.

Files

Institutions

Akkamahadevi Women's University

Categories

Computer Vision, Image Processing, Optical Character Recognition, Face Image Identification

Licence