Skin Disease Identification using online and Offline Data Prediction using CNN Classification

In this study, a convolution neural network (CNN) is used to classify images for the detection of skin illness. We collected in a database from the government medical hospital in Aurangabad and the HAM10000 online data. The seven classes are in the skin diseases dataset such as Basel Cell Carcinoma, Psoriasis, Ringworm, Impetigo, Leprosy, and Eczema. The seven additional categories of skin disease are in the database. We have used pre-processing techniques to improve the model accuracy such as resizing images, and normalization of a dataset. We have used a deep learning algorithm for the classification of skin diseases in the database. We have used a deep learning algorithm for the classification of skin disease. It is given an 80.2% percent accuracy rate and its overall accuracy is 78%. Acne disease identification is got 100 accuracies while testing for it is 97.6% accurate. For the classification of skin diseases, we used a deep learning system. Its total accuracy is 78% and it has an accuracy rate of 82.2%. Identification of the acne disease has a 100 accuracy rating, while testing for it has a 97.6% accuracy rating