PharmAI:A Comprehensive Image Dataset of Tablet Names
Description
This dataset offers a comprehensive collection of images featuring tablet names, perfectly suited for applications like OCR, image classification, and drug recognition. It's neatly organized into 26 folders, one for each letter of the alphabet (A-Z), with each folder containing 100 unique images of tablet names starting with that letter. Additionally, the dataset includes over 10,000 augmented images, derived from the original set using various techniques, significantly boosting the diversity and size of the training pool for more robust machine learning model development.
Files
Steps to reproduce
The dataset was created through a systematic process. Firstly, images of tablet names were captured using mobile phones during visits to various pharmacies across Hassan. Subsequently, each image underwent a meticulous cropping process to isolate and focus solely on the tablet name, excluding any extraneous elements. Finally, all images were uniformly resized to a standardized dimension of 470 pixels in width and 100 pixels in height, ensuring consistency and compatibility across the entire dataset. This approach ensured that the final dataset consists of high-quality images specifically focused on the tablet names, providing a valuable resource for subsequent analysis and machine learning applications.