Computer Vision Introduction Concept _ Computer Vision

Vision plays an important role in human's observation and understanding of the world, and it is a means to observe the world and understand the world.

The ultimate goal of vision: in the narrow sense of the objective scene to make a meaningful interpretation and description of the observer;

It also includes, in broad terms, the development of behavioral planning based on these interpretations and descriptions and the desires of the surrounding environment and the observer, and for the world around it.

Computer vision is the use of computer to achieve visual function.

The goal of computer vision is summed up in two ways: on the one hand, the computer vision system is established to accomplish various visual tasks, that is, the computer obtains the scene image with various visual sensors, from the kind of perception and the response environment the object's geometrical nature, the attitude structure, the movement situation, the mutual position and so on, and the objective scene identification, description, Explain, and then make judgments and decisions. On the other hand, explore the human visual mechanism. For those who do information engineering, the first problem is to establish a computer vision system to accomplish various visual tasks.

Computer Vision related science, beginners often confused about these science, not the concept of various disciplines, so that do not know how to learn.

(related subject one)--image engineering: Image engineering content rich including three levels of image processing, image analysis and image understanding

(1) Image processing (can be said to be the process of image to image): Image processing emphasizes the transformation between images (image input and output), chivalrous image processing is mainly concerned with the visual observation of output images (such as scaling, improve contrast, etc.). The image output after the specific image processing technology can facilitate the subsequent image analysis and so on.

(2) image analysis (can be said to be the result of image to target feature measurement): To detect and measure the parts of interest in the image, so as to obtain the description of the object in the image (that is, to obtain the target data from the image), to describe the characteristics and properties of the objects of interest

(3) Image understanding (interpretation of the results of the target feature measurement): On the basis of image analysis (target description), the paper studies the characteristics of objects in the image and the target image in the objective world of the meaning and interpretation of the objective world, so as to make judgments, guidance and planning behavior.

(Related Disciplines II)-machine vision: Most of the time Machine vision and computer vision are the same, specifically computer vision is focused on scene analysis and image interpretation of the theory and methods, while machine vision is focused on the visual sensor to obtain environmental images, A system with visual perception is constructed to realize the algorithm of detecting and identifying objects.

(three related disciplines)-pattern recognition: a kind of image pattern recognition, which realizes the classification, analysis and description of the interested part.

(Related disciplines IV)-Artificial Intelligence and machine learning: Artificial intelligence is one of the ways to simulate, execute or regenerate certain human beings to understand the world, to judge things, to learn the environment, to plan behavior, to think logically, to solve problems, to realize visual function. Machine learning is the core of artificial intelligence, the study of computer simulation and the realization of human learning ability, so as to acquire new knowledge or skills.

(related disciplines five)--computer graphics: Also known as the inverse problem of computers, generating lifelike images from the data.

It can be seen that the computer vision and these disciplines are inextricably linked to each other, these disciplines intersect each other, so to learn computer vision should be the basic knowledge of the relevant disciplines to understand.

The understanding of these concepts will also help beginners to grasp the direction of the various disciplines, can be clearer choice of the direction of their specialization.