Professional image technology and
machine vision development and manufacturing

24-hour service hotline:
Current Lcation:Home / News / Industry information
Machine vision? What is machine vision in the end?
The author:SIPO  Time:2018-04-02  Read: 151

Robot vision, computer vision, image processing, machine vision, and pattern recognition. What distinguishes these in the end?

It's sometimes confusing to figure out which one of them is. Let's take a look at the specific meaning of these terms and how they relate to robotics. After reading this article, you will no longer be confused by these concepts!

When people sometimes talk about robot vision, they get confused. When they say that they are using "computer vision" or "image processing," they actually mean that they are using 'machine vision'. This is a completely understandable mistake. Because the boundaries between all the different terms are sometimes somewhat ambiguous.

In this article, we have decomposed the “family spectrum” of robot vision to show where the wider signal processing area is.

What is Robot Vision?


In basic terms, robot vision involves the use of a combination of camera hardware and computer algorithms that allow the robot to process visual data from the real world. For example, your system can make a two-dimensional camera that detects an object that the machine will pick up. A more complicated example might be to use a 3D stereo camera to guide the robot to mount the wheel to a moving vehicle.

Without machine vision, your robot is basically a die. This may not be a problem for some robotic tasks. But for some applications, robot vision is helpful, even essential.

Robot Vision's "Genealogy"

Robot vision is closely related to machine vision. We will introduce machine vision later. Both of them are closely related to computer vision. If they are talking about a “family tree,” computer vision can be seen as their “parents.” However, in order to understand their position in the entire system in detail, we will further introduce their "grandparents" - signal processing.



Signal processing includes processing electronic signals, or cleaning (eg, denoising), extracting information, preprocessing for output to the display, or preprocessing for their further processing. Anything can be a signal, more or less. There are various types of signals that can be processed, such as: analog signals, digital signals, frequency signals, and the like. The image is basically just two-dimensional (or more dimensional) signals. For robot vision, we are interested in image processing. So, we are discussing image processing, right? wrong.

ImageProcessing vs Computer Vision


Computer vision and image processing are like cousins, but they have very different goals. Image processing technology is mainly used to improve the quality of an image, convert it to another format (such as a histogram) or change it for further processing. Computer vision, on the other hand, is more focused on extracting information from images to perceive them. Therefore, you may use image processing to convert a color image to a grayscale image, and then use computer vision to detect the object in the image. If we look further up at this “family genealogy,” we see that both fields are greatly affected by the physical field, especially optics.

Pattern Recognition and Machine Learning

So far, the situation is still so simple. When we added pattern recognition or broader machine learning to the “family tree,” things started to get a little more complicated. This branch focuses on identifying the graphics in the data, which is important for the more advanced features of Asada that require robot vision. For example, in order to be able to identify an object from its image, the software must be able to monitor whether it sees objects that were previously seen. Therefore, machine learning is another aspect of computer vision than signal processing.

However, not all computer vision technologies require machine learning. You can also use the signal instead of the image for machine learning and then input it as an Input to the machine learning algorithm. E.g. Computer vision detects the size and color of the parts on the conveyor belt, and the machine learns to determine if these parts are defective or not, based on knowledge it learned from what normal good products should look like.



Now that we talk about machine vision, everything will change. This is because machine vision is completely different from the previously mentioned terms. It is more focused on specific applications than just on technology. Machine vision refers to the visual inspection of industrial applications for automatic detection, process control, and robot guidance. The rest of the "Genealogy" is in the field of science, and machine vision is an engineering field.

In a way, you can think of machine vision as a child of computer vision because it uses computer vision and image processing techniques and algorithms. However, although it can be used to guide robots, he is not exactly robot vision.



Finally, finally reached the robot vision. If you have been following this article until now, you will realize that robot vision uses all the previous techniques. In many cases, robot vision and machine vision are used alternately. However, there are still some subtle differences. Some machine vision applications, such as part monitoring, have nothing to do with the robot. The workpiece is simply placed in front of a vision sensor that is used to detect defects.

In addition, robot vision is not only an engineering field. It is also a science with its own specific research field. Unlike spring computer vision research, robot vision must incorporate robotics into its technologies and algorithms such as kinematics, reference frame alignment, and the ability of lover physics to influence the environment. Visual servoing is a perfect example of what smart intelligence is called robot vision instead of computer vision. It involves the robot's motion control, using the visual sensor to detect the feedback of the robot's position.

【Return to the front page】
Prev:【Focus】Artificial Intelligence New Air Vent “Industrial Eye” Machine Vision
Next:Machine Vision in the Age of Industry 4.0
Relevant articles
Recommended Products
Contact Us

Address: 11th Floor, Kongza Industrial Park, No. 132, Gangzha Street, Fuyong Industrial Zone, Xinqiao, Shajing Town, Bao'an District, Shenzhen City, Guangdong Province, China.

Phone: 86 755 29927385

Fax: 86 755 81466627


Sweep more exciting
Copyright © 2019. Shenzhen Sipotek Technology Co., Ltd. .All rights reserved.粤ICP备05044648号-2