Computer vision technologies are continually reaching greater heights, incorporating imaging systems and their applications. These applications are available in fields like medicine, engineering, space, material science, and many others. Thus, these applications encompass vision metrics for the precise calculation of scientific data to achieve the desired results. The author focuses on the vision metrics involved in extracting the desired features from the images.

This book consists of eight chapters, four appendices, and an exhaustive bibliography with more than 500 references. A four-page introduction is an interesting addition that focuses on vision developments in the image features’ descriptions. There are historical surveys on image features and textural analysis. Along with these surveys, the robustness and vision metric taxonomies are special attractions of this book. There is also a survey on the feature descriptors that includes spectra, basic space, polygon, 3D, 4D, volumetric, and multimodal descriptors.

The book contains descriptions of the image capturing, representation, and preprocessing techniques. The section on feature extraction describes global, local, and regional features, as well as design concepts following classification and learning. An emphasis on ground truth data and analysis covers the details on optimizing computer vision computations.

The author claims this work is unique from similar works in that it avoids “how-to” source code examples, programming-based manuals, and illustrations on programming packages based on the functionalities in working with the images. However, some of the selected programming examples in OpenCV are an integral part of this work.

The author presents thoughtful insights on machine learning, classification techniques, and similarity measures of the featured metrics. This book is suitable for advanced readers who are well acquainted with vision metrics. Beginners in this field will need to revisit the fundamentals of computer vision and imaging techniques before reading it.

More reviews about this item: Amazon