Dyslexia is a significant issue that affects an individual's writing, reading, and speaking skills, making it difficult for them to integrate into society and handle regular tasks, especially in school. Our project aims to aid individuals with dyslexia in improving their handwriting, which will indirectly enhance their reading skills and overall communication. By refining their handwriting, we hope to make daily life and academic experiences smoother for them.
Our website is designed to help users practice their handwriting through a structured process. We will provide them with a set of sentences to copy down on paper. Once they have completed the task, they will take a picture of their writing and upload it to our website. Using OpenCV, the uploaded image will be processed to recognise the handwritten letters. The system will utilise a dataset of characters to interpret the text and convert it into digital form. Due to the nature of dyslexia, certain characters may not be processed correctly, and these will be highlighted for further analysis. The extracted text will then be compared with the original sentence provided to the user, and any discrepancies will be identified. The errors in handwriting will be categorised, and the user will receive detailed feedback on their mistakes and how to correct them. Additionally, users can choose to receive feedback through text-to-speech using gTTS, or, if they are at a more advanced level, they can attempt to read the feedback themselves.
The backend of our project involves several crucial tasks. First, we need to compile a dataset of properly written handwriting characters. This dataset will be used to train our model to accurately recognize different letterforms. The model must achieve at least 75% efficiency in recognizing handwritten characters to ensure reliability. Once trained, the model will be integrated with OpenCV to process user-submitted handwriting images and extract text from them. After extraction, the system will compare the extracted text with the given correct sentence and determine which letters are incorrect or missing. The system will then present this analysis in a clear and accessible format, providing users with constructive feedback to help them improve their handwriting.
Our project aims to create an inclusive and supportive learning environment for individuals with dyslexia. By leveraging technology, we provide a practical tool to help them refine their handwriting, which in turn enhances their reading and writing skills. With continuous improvements and user feedback, we hope to make this platform as effective and user-friendly as possible, ultimately making a meaningful difference in the lives of those affected by dyslexia.
Developed by: Chocolate Vanilla Swirl (Santino, Bryan and Dhanvin (not in photo))