Deep Learning Projects for Multidisciplinary Engineering Design Students

Deep Learning is a form of AI machine learning that has gained a great deal of recognition in the past 10 years in a wide range of areas such as medical diagnosis, quality assurance, defect detection, face detection, autonomous vehicles, and many others. Deep learning networks, however, typically require large training databases of labeled images and often require specialized hardware and high-level software expertise. Techniques, such as transfer learning and the proper choice of software tools can mitigate some of these requirements. This paper describes a new, project-based course module to introduce deep learning and computer vision to undergraduate multidisciplinary engineering students in a robotics design and applications course using MATLAB software.