Converting Images of Mathematical Equation into Latex Equivalent

Recognizing handwritten mathematical expressions is challenging due to the similarity of symbols and their two-dimensional structure, unlike linear text. This paper presents a deep learning-based approach that directly converts images of handwritten mathematical expressions into LaTeX code without requiring predefined rules or explicit symbol segmentation. The proposed model employs a Convolutional Neural Network (CNN) for image feature extraction and a Recurrent Neural Network (RNN) with an attention mechanism to generate mathematical expressions sequentially. The model demonstrates superior performance compared to traditional methods on the CROHME dataset, achieving high accuracy in handwritten mathematical expression recognition.