Skin Disease Recognition and Classification Using Machine Learning and Deep Learning in Python

This chapter reviews the concept of skin disease recognition and classification based on computerized techniques, that is, machine learning and deep learning-based methods. It begins by discussing about the skin disease image acquisition methods and the available datasets as the main and primary requirement for computer-aided systems. Applying pre-processing techniques to improve the image quality and clarity is an important step in designing the skin disease recognition and classification systems. By applying pre-processing, the input images are resized all in the same image size and their quality and clarity are improved by color space conversion, contrast enhancement, denoising, and deblurring. The exact area of the skin disease is separated from the healthy skin by applying different techniques of segmentation which improves the recognition performance. When extracting features from image data, the histogram of oriented gradients is often employed as feature extraction. It is commonly used in computer vision applications, such as skin disease detection and recognition.