Current and Evolving Applications to Video and Imaging
Daniel Minoli, Benedict Occhiogrosso · 2023
This chapter explores machine learning techniques for imaging and computer vision (CV). It focuses on the use of artificial intelligence to interpret or enhance images in support of applications such as classification, detection, CV, face recognition, surveillance, situational awareness, and medical imaging. Neural networks (NNs), deep neural networks (DNNs), and ultra deep NNs play important roles in automatic processing of large bodies of data, especially in the video/CV arena. There are various types of NNs including feed-forward networks and convolutional neural networks (CNNs). A CNN is a DNN with a convolutional structure. CNNs have become prevalent in the CV field in recent years: they are now often used for vision and image recognition applications. Convolution has applications that include signal processing, image processing, CV, among others. In imaging applications, convolutions are used for extracting shapes and curves in an image.