Detection of skin diseases from skin image by using deep learning algorithms in python

Worldwide data storage, processing, and interpretation work are becoming vital regardless of the industry. This has become an important part of the new technological scheme based on visuality. With the developing hardware and the increasing speed of the internet, image processing works prepared the basis for the formation of complex, detailed and modernized structures. Increasing the quality of photographic and video images allows us to make a determination in many different areas from the visual world. In many fields such as cybersecurity, finance, statistics, defense industry, and health, visuals can be collected and processed for diagnosis, determination, and inference. The most effective and fastest way to do this is by using deep learning methods for large data sets. Tests classify and self-learning platforms are created by parallel processing method. The health sector has been teaching machine learning and so on for many years. alternative methods to the experts in the field are tried to be developed. In this context, significant successful studies have been conducted. In particular, studies on the images show a lot of them, one of which is the skin images evaluated for the diagnosis of skin cancer. Skin cancer is seen in serious numbers compared to the worldwide population. Early detection plays an important role in the treatment of a high percentage of the disease. The study represents a successful and easy alternative to early detection of skin cancer. Image processing methods are very effective in this context. However, in addition to the increasing image sizes, the increasing amount of data and the presence of self-learning structures, deeper learning methods become more prominent. One of the most common methods of constructing these structures is the Convolutional Neural Network (ESA). This structure is one of the most effective methods used to classify the data to be examined. In this study, the skin lesions were diagnosed with different layers using the ESA structure.