Age, Gender and Handedness Prediction Using Handwritten Text
摘要
This work proposes an AI-based handwriting analysis system that combines Convolutional Neural Networks (CNN) and Google’s Gemini Generative AI for precise handwriting-based gender prediction and personality insights. This system utilizes the Flask Web Framework, allowing users to upload handwriting samples or capture handwriting samples via their webcam. The CNN model predicts gender based on handwriting samples. Gemini AI predicts characteristics such as handedness, age group, and style of writing. The results are then assessed for accuracy and the results from the two models are saved to a centralized MySQL database. There is also a contact module to allow users to reach out to the team with any feedback. We have developed a useful and accurate hybrid approach to use AI to predict human characteristics from handwriting, as it can be used for forensics, psychology, and education.