Artech House author Souvik Hazra, along with Avik Santra, wrote Deep Learning Applications of Short-Range Radars explains the importance of this book:
With the well establishment of deep learning and its use in real world , it has shown ground breaking performance in Computer Vision tasks (CV), Natural Language Processing & Understanding (NLP & NLU) and etc. However, application of DL in short-range radars was limited although has a growing interest. There was no central volume, thus we decided to write this book to spur and cater to the research needs and push that the field needs.
Deep Learning Applications of Short Range Radars is focused on providing a comprehensive guide to deep learning and using it for developing industrial and consumer solution using short-range radars to professionals and students pursuing advanced degrees on machine learning , signal processing and other relevant topics .
The book covers a smooth transition from conventional signal processing pipeline to machine learning/ deep learning pipeline for real-world applications such as gesture recognition and sensing, human activity classification, air-writing, material classification, vital sensing, people sensing, people counting, people localization and in-cabin automotive occupancy and smart trunk opening with multiple deep learning techniques in use for each application. It caters to a wide audience coming from different backgrounds and thus entails the basic blocks of deep learning and signal processing algorithms. Each technique used in different applications is justified on the basis of accuracy, computational power and memory size, which enables such algorithms to be deployed in on-the-edge systems. Furthermore, the book touches on advance deep learning deployment methodologies such as Federated Learning, which allows readers to think, design and develop an end-to-end deep learning solution using Short Range Radars.
The most important aspect of the book is that it provides a balanced view between multiple data representation and deep learning algorithms used for each application, encouraging readers to use different blocks for most appropriate idealization of their target application.
Each chapter concludes with a series of questions and references, which enables the readers to evaluate their understanding and have access to additional resources on the given topic.
