The Development Of Computer Vision
Computer vision is how computers can be made to gain a high-level understanding of digital images or videos. From an engineering perspective, this field seeks to automate objects that the human visual system can do. Computer vision is concerned with the automatic extraction, analysis, and understanding of useful data from a single image or series of pictures. The computer vision models involve developing theoretical and algorithmic foundations to achieve automatic visual understanding.
As a scientific domain, computer vision is affected by the idea of artificial methods that extract data from images. Image data can bear many states, such as video series, views from numerous cameras, or multi-dimensional data from medical scanners. In the late 1960s, Computer Vision began at the university that pioneered artificial intelligence. The technology is intended to imitate the human eyesight system, as a stepping stone to grant robots with intelligent demeanor. In 1966, it was believed that this could be achieved via a summer project, by connecting a camera to a computer and holding it “draw what it saw”.
What distinguished computer vision from the field of digital image processing prevalent at that time was the desire to extract three-dimensional structures from images to achieve full scene understanding. Studies in the 1970s formed the early foundations for many of today’s computer vision algorithms, including edge extraction from images, line labeling, non-polyhedral and polyhedral modeling, representation of objects as interconnections of smaller structures, optical flow, and estimation.
The next decade was marked by studies based on more rigorous mathematical analysis and quantitative aspects of computer vision. These include spatial-scale mathematical concepts, shape inference from various cues such as shadow, texture, and focus, as well as contour models known as snakes. The researchers also realized that many of these mathematical concepts could be treated in the same optimization framework as regularization and Markov random fields.