Research on Skin Disease Health Detection of College Students based on Deep Learning

Abstract

Skin disease is a common diseases. Although it is very common, the manifestations of skin diseases are various, and the diagnosis process is relatively difficult, and the requirements for dermatologists are also very high. And the recognition efficiency of detection also needs to be improved. To this end, this paper mainly investigates the potential of classifying skin diseases in deep learning convolutional neural networks to enhance the recognition accuracy. The VGG16 neural network was trained using the training data of the ISIC 2019 challenge, and the training data was augmented to increase the accuracy of the model, and accurately classify 3 skin diseases with 97% accuracy. This method can help doctors to make auxiliary diagnosis and provide technical support for skin disease diagnosis.