The main research content of computer vision

Computer vision is an interdisciplinary field of science that focuses on how computers can obtain a high level of understanding from digital images or videos. From an engineering perspective, it attempts to understand and automate tasks that can be performed by the human visual system. Computer vision tasks include methods for acquiring, processing, analyzing, and understanding digital images, as well as methods for extracting data from the real world to generate numerical or symbolic information.

Computer vision key technology

Feature detection

In computer vision technology, feature detection is a very basic and important role. A variety of tasks in computer vision, such as target recognition, image classification, image segmentation, stereo vision, three-dimensional reconstruction and other tasks are based on feature detection, through the detection and extraction of features to complete the subsequent tasks. Features in feature detection include feature points, contours, edges, etc., and locations with obvious recognizable differences from the surrounding environment are features.

Image Segmentation

Image segmentation as the name suggests image segmentation that is to want to recognize the target from the image segmentation, image segmentation is a very important task in computer vision, it has a wide range of applications in real life, and plays a central role, for example, in the pedestrian detection, video surveillance, automated driving, medical image analysis and so on, image segmentation plays an indispensable role. The following figure shows an example of image segmentation.

Stereo Vision

Stereo vision refers to the technique of using two or more cameras to acquire visual information about depth.

Binocular vision solving for depth is introduced first. Binocular vision solving depth is based on the principle of triangulation of perspective geometry, through the left side of the image captured above any point, in the right side of the image captured to find the corresponding matching point, you can determine the three-dimensional coordinates of the point. The following figure shows the process of solving depth by binocular vision.

Your task

Give examples of typical algorithms for deep learning.

Share your thoughts and ideas in the comments below.