In the era of intelligent manufacturing, more and more factories have begun to introduce automated production line technology, and through the empowerment of industrial robots or robotic arms, the production line can improve production efficiency.
In artificial intelligence-led robots, if you want your industrial robots to make accurate movements, then machine vision will be a very important core function. Of course, things cannot be absolute. For some robots, this Maybe not be a problem, but for some applications machine vision is essential.
What is the difference between robot vision, computer vision, image processing, machine vision, and graphic recognition? Let’s take a look at what these terms mean exactly, and how they relate to robotics.
What is machine vision?
“Machine vision” refers to the use of machines instead of human eyes for measurement and judgment. The computer is mainly used to simulate the human visual function, but it is not just a simple extension of the human eye, more importantly, it has a part of the function of the human brain – extracting information from the image of objective things, processing, and understanding, and finally using for practical detection, measurement, and control. Therefore, machine vision technology is an interdisciplinary subject involving artificial intelligence, neurobiology, psychophysics, computer science, image processing, pattern recognition, and many other fields.
Difference between machine vision and computer vision
Machine vision is different from computer vision, it involves image processing, artificial intelligence, and pattern recognition
Machine vision is the engineering that focuses on integrating mechanics, optics, electronics, and software systems to inspect natural objects and materials, artificial defects, and manufacturing processes in order to detect defects and improve quality, operational efficiency, and product and process safety. It is also used to control machines.
Machine vision is the application of computer vision to industrial automation.
Therefore, the difference between machine vision and computer vision is mainly reflected in the application field. Machine vision is more used in product appearance inspection, dimension measurement, barcode/QR code recognition, visual positioning, and so on in manufacturing enterprises. It is more focused on a specific application rather than just focusing on the technical parts. Technologies such as image processing, pattern recognition, etc. are scientific fields, while machine vision is an engineering field. It refers to the vision for industrial use for automatic inspection, process control, and robot guidance.
What is machine vision used for?
Because the machine vision system can quickly obtain a large amount of information, it is easy to automatically process and integrate with design information and processing control information. Therefore, in the modern automated production process, people use the machine vision system widely for working condition monitoring and finished product inspection. and quality control areas.
The machine vision system is characterized by improving the flexibility and automation of production. In some dangerous working environments that are not suitable for manual work or where artificial vision is difficult to meet the requirements, machine vision is often used to replace artificial vision; at the same time, in the process of mass industrial production, the efficiency and accuracy of using artificial vision to check product quality is low. , the use of machine vision inspection methods can greatly improve production efficiency and production automation. Moreover, machine vision is easy to realize information integration, and it is the basic technology to realize computer-integrated manufacturing.
The introduction of machine vision, instead of traditional manual inspection methods, has greatly improved the quality of products put on the market and increased production efficiency.
In a way, you can think of machine vision as the child of computer vision because it uses techniques and algorithms from computer vision and image processing. But while it can be used to guide robots, it’s not exactly robot vision.
Machine vision and applications
Machine vision systems have been widely used in various aspects of quality inspection, such as large-scale workpiece parallelism and perpendicularity measuring instruments using laser scanning and CCD detection systems, which use stable collimated laser beams as the measurement rotates the shaft system, rotate the pentagonal standard prism to scan out the reference planes parallel or perpendicular to each other, and compare them with the sides of the large workpiece to be measured. When processing or installing large workpieces, the detector can be used to measure the parallelism and perpendicularity between surfaces.
The strobe light is used as the illumination source, and the area array and linear array CCD are used as the detection device of the outline dimension of the rebar to realize the dynamic detection system for online measurement of the geometric parameters of the hot-rolled rebar.
Vision technology monitors the load and temperature changes of the bearings in real-time, eliminating the danger of overloading and overheating. Change the traditional passive measurement by measuring the surface of the ball to ensure processing quality and safe operation into active monitoring.
A microwave is used as the signal source, and square waves with different baud rates are sent out according to the microwave generator to measure the cracks on the metal surface. The higher the frequency of microwave waves, the narrower the cracks that can be measured.
With machine vision as the core, DBM provides vision-based robot workstations in metallurgical and other industrial processes, unmanned factories, visual inspection, intelligent inspection, unmanned vehicles, and other equipment research and development, manufacturing, and process improvement services. If you have related needs, please feel free to contact us!
What is robot vision?
Robot vision refers to the system that makes the robot have a visual perception function and is one of the important parts of the robot system.
In practical applications, this module mainly integrates industrial cameras or smart cameras and vision software. Taking image information is converted into electronic signals, and finally through algorithm analysis to achieve the purpose of “seeing” by the robot, so as to process production Visual data in action.
In basic terms, robot vision involves using a combination of camera hardware and computer algorithms to let robots process visual data from the real world. For example, your system could use a 2D camera to detect an object that the machine will pick up, a more complex example might use a 3D stereo camera to guide the robot to attach wheels to a moving vehicle.
Without machine vision, your robot is basically blind. For some robotic tasks, this may not be a problem, but for some applications, robot vision is helpful, even essential.
Robot vision adopts all previous technologies. In many cases, robot vision and machine vision are used interchangeably. However, there are some subtle differences. Some machine vision applications, such as part inspection, are robot-independent, and the workpiece is simply placed in front of a vision sensor used to detect defects.
In addition, robot vision is not only an engineering field but also a science with its own specific research field. Different from pure computer vision research, robot vision must incorporate robotics into its technology and algorithms. Visual servoing is a perfect example of an intelligent technology called a robotic vision rather than computer vision. It involves the motion control of the robot by using vision sensors to detect the feedback of the robot’s position.
Judging from the current general environment, the emergence of industrial robots is destined to adapt to a more harsh production environment to replace manual labor to complete repetitive and mechanical work. In this process, in order to make robots more artificial intelligence, we Give the robot vision function through the method of machine vision.
Related Technology of Robot Vision
Robot vision is closely related to machine vision, and both of them are closely related to computer vision. However, in order to understand their position in the overall system in detail, we will go further and introduce another technology-signal processing.
Signal Processing
Signal processing includes manipulating electronic signals, either cleaning (eg: noise removal), extracting information, preprocessing for output to a display, or preparing them for further processing. Anything can be a signal, more or less, there are various types of signals that can be processed, eg: analog, digital, frequency, etc. Images are basically just two-dimensional (or more) signals, for robot vision, we are interested in image processing.
Image Processing vs Computer Vision
Computer vision and image processing are like cousins, but they have very different goals. Image processing techniques are mainly used to improve the quality of an image, convert it to another format (eg histogram) or alter it for further processing. Computer vision, on the other hand, focuses more on extracting information from images in order to perceive them. So you might use image processing to convert a color image to grayscale, and then use computer vision to detect objects in the image. Both fields are heavily influenced by the field of physics, especially optics.
Pattern Recognition and Machine Learning
So far, the situation has been as simple as that. Things start to get a bit complicated when we add pattern recognition or machine learning more broadly, this branch focuses on recognizing patterns in data, which is quite important for functions that require relatively more advanced robotic vision. For example, in order to be able to recognize an object from its image, the software must be able to detect whether the object it sees is one it has seen before. Therefore, machine learning is another parent of computer vision besides signal processing.
However, not all computer vision techniques require machine learning, you can also use only signals instead of images for machine learning, and then use it as input to the machine learning algorithm. For example. Computer vision detects the size and colour of parts on the conveyor belt, and machine learning decides whether those parts are defective based on what it has learned from what normal good parts should look like.
In actual technical applications, it is usually realized by binocular, multi-eye technology, laser camera, 3D vision, etc., which improves the autonomy of the equipment and further improves production efficiency.
From the perspective of future trends, robot vision will develop in the direction of AI+AR. Artificial intelligence can improve the accuracy of recognition, while AR technology can better realize the function of human-computer interaction.
The application of robot vision:
①Provide visual feedback for the motion control of the robot. Its function is to identify the workpiece, determine the position and orientation of the workpiece, and provide visual feedback for the adaptive control of the robot’s trajectory. Operations that require the application of robot vision include picking workpieces from conveyor belts or feeding bins, management and control of workpieces or tools during manufacturing, e.g. Assembly operations for feedback.
②Visual navigation of mobile robots. At this time, the function of robot vision is to use visual information to track the path, detect obstacles and identify road signs or the environment to determine the orientation of the robot.
③Replacing or assisting manual visual inspection for quality control and safety inspection.
DBM vision inspection project >>>