Face-Skin-disease

Face Skin disease detection comparison using CNN, CNN-LSTM, and Mobilenet To classify face skin diseases such as The categories including Acne, Actinic keratosis, basal cell, Lupus-Chronic-Cutaneous, Rosacea, Seborrheic, squamous cell - Neural Network such as CNN was implemented. One hot encoding implemented to classify each of the class from 0-5. To compare these models performance, pre-trained Mobilenet was used.CNN model got train accuracy 97% and the test accuracy is 82%. In contrast, Mobilenet Version - 2 has train accuracy 93% while the test accuracy is 78%. Proposed CNN model has outperformed the pre-trained mobilenet model.