Tongue Image Enhancement for Ayurvedic Diagnosis

Ayurveda thinks that looking at the tongue is an important way to figure out what is going on inside the body and to see if the doshas like Vata, Pitta and Kapha are out of balance. The problem is that the traditional way of looking at tongues to diagnose problems is done by people so it is not always accurate and can be wrong. This study is trying to create a system that can automatically make tongue images clearer and classify them using a kind of computer program that combines Convolutional Neural Network and Support Vector Machine models. Ayurveda and the doshas, like Vata, Pitta and Kapha can really benefit from this system. The proposed method enhances tongue images through preprocessing techniques such as noise removal, contrast enhancement, and segmentation, followed by feature extraction using CNN. The extracted deep features are classified using SVM to improve diagnostic accuracy. Experimental results demonstrate that the CNN–SVM hybrid model achieves superior performance compared to standalone classifiers, with an accuracy of up to 93%. The system provides a reliable, objective, and efficient approach to assist Ayurvedic practitioners in diagnosis