Deep Learning was developed as a Machine Learning approach to deal with complex input-output mappings. While traditional ML methods successfully solve problems where final value is a simple function of input data. On the contrary, Deep Learning techniques are able to capture composite relations between air pressure recordings and English words, millions of pixels and textual description, brand-related news and future stock prices.In conclusion, Deep Learning has a great advantages vs shallow learning, because deep nets can learn very complex functions which we even hardly understand.This book features the application of various deep learning techniques like Artificial Neural Network and Machine Vision in Friction Stir Welding process. It also contains few research papers already published by the authors in the form of chapters to make you fully understand the concepts.