Deep learning and image processing techniques for skin disease identification are part of the suggested solution. Skin conditions are brought on by a variety of reasons, including DNA sequencing, radiation, and mutations, which result in skin defects. If skin conditions are not treated in a timely manner, they often spread to other parts of the body. In order to be treated, skin disorders must therefore be found in their early stages. These symptoms need to be identified early because skin disorders have been linked to mortality problems, lengthy, expensive therapies, and numerous characteristics. We are creating web application for early stage skin disease detection. In the suggested solution, skin illnesses will be identified from the provided image collection using image processing techniques which uses a color image's inputs. Depending on the training dataset, we have used the Convolutional Neural Network which has an excellent visual representation power for the recognition or detection task. Therefore in order to diagnose skin diseases early with more accuracy and efficiency transfer learning will be used . It will provide the best training to the model and give a high accuracy, precision, recall, specificity for the correct anticipation of skin diseases among all picked which help doctor for early detection and prevent chronic disorder as well as economic and mental loss

Keywords : Data Preprocessing, Outlier Detection, Image Classification, Deep Learning, Convolution Neural Network, Transfer Learning.