Classification 19 Type of Skin Condition by Using Convolution Neural Network

Abstract

Convolutional Neural Network (CNN) is a vision computerization analysis that assists in understanding the deep learning models. This study will carry out to purpose one method for machine learning files that can be used for classifying skin disease through deep learning based. This study will determine the precision(p), recall(r) and score(F1) of model to classify the type of skins into own categories. The focus is to generate the architecture hdf5 type file that can be automated for the diagnosis of 18 common skin diseases and 1 normal skin by using data from public clinical images dataset and patient information using deep learning pre-trained EfficientNetB7 model. The implement a simple image classification model using TensorFlow to rebuild one of parameter hospitality for nursery robot development in future. The image classification model will classify images of various skin disease problems into labelled classes.