You get a bonus - 1 coin for daily activity. Now you have 1 coin

Computer vision

Lecture



Computer vision is the theory and technology of creating machines that can see. As a scientific discipline, computer vision refers to the theory and technology of creating artificial systems that receive information from images. Video data can be represented by a variety of forms, such as video sequences, images from different cameras, or three-dimensional data from a medical scanner.

As a technological discipline, computer vision seeks to apply theories and models of computer vision to the creation of computer vision systems. Examples of the use of such systems can be:

  • Process control systems (industrial robots, autonomous vehicles)
  • CCTV systems
  • Systems for organizing information (for example, for indexing image databases)
  • Object or environmental modeling systems (medical imaging, topographic modeling)
  • Interaction systems (eg, input devices for a human-computer interaction system)

Computer vision can also be described as an addition (but not necessarily the opposite) to biological vision. In biology, the visual perception of humans and various animals is studied, as a result of which models of such systems are created in terms of physiological processes. Computer vision, on the other hand, studies and describes computer vision systems that are implemented in hardware or software. The interdisciplinary exchange between biological and computer vision turned out to be very productive for both scientific fields. Sub-sections of computer vision include reproducing actions, detecting events, tracking, pattern recognition, image restoration.

The field of computer vision can be characterized as young and diverse. Even though there are earlier works, it can be said that it was not until the late 1970s that an intensive study of this problem began, when computers were able to manage the processing of large data sets, such as images. However, these studies usually began with other different areas, and, therefore, there is no standard formulation of the computer vision problem. Also, and even more importantly, there is no standard formulation of how the problem of computer vision should be solved. Instead, there are many methods for solving various well-defined tasks of computer vision, where methods often depend on tasks and rarely can be generalized for a wide range of applications. Many of the methods and applications are still in basic research, but an increasing number of methods are used in commercial products, where they often form part of a larger system that can solve complex problems (for example, in the field of medical imaging or quality control). in manufacturing processes). In most practical applications of computer vision, computers are pre-programmed to solve individual problems, but knowledge-based methods are becoming more common.

An important part in the field of artificial intelligence is automatic planning or decision making in systems that can perform mechanical actions, such as moving a robot through a certain environment. This type of processing usually requires input data provided by computer vision systems that act as a video sensor and provide high-level information about the environment and the robot. Other areas that are sometimes described as belonging to artificial intelligence and that are used relative to computer vision are pattern recognition and teaching methods. As a result, computer vision is sometimes seen as part of the field of artificial intelligence or the field of computer science in general.

Physics is another science that is closely related to computer vision. Much of computer vision deals with methods that require a thorough understanding of the process in which electromagnetic radiation, usually in the visible or infrared range, is reflected by the surface of objects and measured by an image sensor to get video data. This process is based on optics and solid state physics. More sophisticated image sensors even require knowledge of quantum mechanics to fully understand the process of image formation. Also, various measurement problems in physics can be solved using computer vision (for example, related to motion in liquids). Therefore, computer vision can be considered as an extension of physics.

The third area of ​​science that plays an important role is neurobiology, especially the study of biological vision systems. Over the past century, extensive studies have been conducted on the eyes, neurons and brain structures related to the processing of visual stimuli in both humans and various animals. This led to a rough, and more complicated, description of how the “real” vision systems work, which helped to solve some problems. The results of these studies led to the creation of artificial systems that simulate the operation and functioning of similar biological systems at various levels of complexity. Also, some study methods developed in the field of computer vision owe their origin to biology.

Another area related to computer vision is signal processing. Many methods for processing one-dimensional signals, usually time signals, can be naturally extended to process two-dimensional or multi-dimensional signals in computer vision. However, due to the peculiar nature of images, there are many methods developed in the field of computer vision that have no analogues in the field of processing one-dimensional signals. A special property of these methods is their non-linearity, which, together with the multidimensionality of the signal, makes the corresponding subdomain in signal processing part of the field of computer vision.
created: 2014-09-22
updated: 2021-03-13
132562



Rating 9 of 10. count vote: 2
Are you satisfied?:



Comments


To leave a comment
If you have any suggestion, idea, thanks or comment, feel free to write. We really value feedback and are glad to hear your opinion.
To reply

Natural Language Modeling of Thought Processes and Character Modeling

Terms: Natural Language Modeling of Thought Processes and Character Modeling