Machine vision in production. Machine vision. What is it and how to use it? Optical source image processing. Advantages and disadvantages of machine vision systems

Computer vision at its core involves analyzing visual information to further decide what action to take in relation to an object that is in focus. The simplest example of using technology: checking the condition of a product on a conveyor belt or before sending a parcel by mail. It is also common for machine vision to be used to evaluate the quality of printed circuit boards, instantly comparing each new product to a reference board before automatically moving it to the next assembly stage. These technologies provide an invaluable resource for assessing quality and reducing defect rates where the human eye and brain simply cannot. objective assessment due to the need to view the same items hundreds or thousands of times a day.

The computing need of “machine vision”

As resolution increases optical systems The potential of machine vision is also increasing, as the number of details to evaluate increases along with the resolution. More and more small objects can be processed according to a template principle, which leads to an increase in the load on the processor, which must analyze a significant amount of data and quickly make decisions about next step(match/fail, hold, return to start, etc.). For example, when sorting vegetables simple solutions about suitability and size are no longer suitable, because the standards different countries vary and product quality varies depending on the season. To minimize the amount of defects for the manufacturer and at the same time ensure required quality for the consumer, more detailed algorithms and categories are needed - and this is an almost impossible task for human eyes and brain.

Let's say there is a Danish company called Qtechnology, which supplies “smart cameras” for sorting vegetables. They are capable of processing up to 25 tons of products per hour without human intervention. Such volumes are achieved by analyzing over 250,000 individual products based on more than 500,000 images. And since each image takes up approximately 6.2 MB, it turns out that in the end it requires analysis of over 2.5 terabytes of graphic data per hour - a colossal amount of information! Only to transfer such an array will require more than 6 hours using a gigabit Ethernet connection.

To solve this problem using simpler algorithms, you need to break it down into stages and install multiple cameras, increase lighting areas, allocate more space in factories, and so on. As an alternative, more productive computing systems can be used: with centralized power and a faster connection or distributed information processing with “smart cameras” that will record data in real time at each stage, supplying only ready-made parameters to the final decision-making mechanism.

In standard visual inspection systems, the quality and safety of products are most often determined by external physical signs such as texture and color. Hyperspectral imaging gives the food industry the ability to evaluate products using additional chemical and biological parameters to determine the sugar, fat, liquid and bacteria levels in each product.
In hyperspectral imaging, three-dimensional sets of spatial and spectral information are obtained from each pixel. Additional spectral characteristics give more detailed description parameters, allowing for their classification. 3D data sets include the intensity (reflected or transmitted light) of each pixel, which is calculated by measuring the length of all visible light waves, resulting in each data set containing a wealth of information. This amount of information reflects an exponential growth in the computational task for conducting qualitative and quantitative analysis product status in real time.

Application of heterogeneous computing

To meet the demands, as well as meet future challenges, high-performance and scalable computing systems are needed.

The mentioned Qtechnology uses AMD's APU hybrid processors in smart camera platforms. These processors combine a GPU (graphics processing unit) and a CPU (central processing unit) on a single chip. As a result, the system has the ability to send arrays of graphics data directly to the GPU for processing without any transmission delay between components. And the CPU is able to process other tasks without delay, increasing the performance of the entire system in real time and providing required power For modern requirements systems with machine vision.
Combining different computing modules on one chip or in one system allows each element to be assigned its corresponding load - and this is the basis of heterogeneous computing. The Heterogeneous System Architecture (HSA) Foundation consortium was founded in 2012 to formulate open industry specifications for processors and systems that leverage all available computing elements to improve end-use efficiency. AMD is promoting the concept of heterogeneous computing, the essence of which is sharing all computing resources of the system: both central and graphic processors.

Specifically, the GPU is a parallel computing engine that can easily apply the same instructions to large sets of data (pixels in our case) simultaneously; and this is exactly what companies need to operate machine vision installations. In addition, system performance can be increased by combining the capabilities of the APU with an external discrete graphics card. This approach allows companies to add GPU computing resources as needed to support even more complex tasks machine vision.

The ecosystem's extensive support for the x86 architecture allows companies to use open source image processing libraries or plug in third-party solutions such as OpenCV, Mathworks Matlab, and Halcon. Debugging tools, latency analyzers and profilers (perf, ftrace) are also widely available today. Machine vision represents the latest example of harnessing the computing power of semiconductors to reduce costs, speed up production, improve quality and provide a range of other useful benefits across many applications and industries. Thus, thanks to innovation and successful ideas engineers for embedded solutions, there is a positive effect in general for the economy, culture and each of us in particular.

Technology Market Overview computer vision

The modern world of computer systems is difficult to imagine without machine or computer vision technologies. In the article “Why does a computer need vision?” (ComputerPress No. 5’2002) the history of the formation of this technology was reviewed and an overview of a number of its applications was given. Of course, the article describes only small part applications from wide range applied computer vision systems, and in future issues we will return to consider this very interesting and rapidly developing area of knowledge. Yes, it is rapidly developing. After all, this technology is only about 50 years old, which by the standards of many exact sciences does not go beyond its infancy. Increasing its scientific and practical potential in parallel with the improvement of computing and recording technology, computer vision is gradually conquering new technological frontiers. High-performance computers of the latest generation (including modern personal computers) already allow solving many problems of processing digital video information streams and making decisions in real time. And today, sometimes unnoticed by most of us, computer vision is quite firmly established in many areas of human life, helping him, and sometimes replacing him, relieving him of monotonous, routine or, often, life-threatening work.

It's no secret that computer vision as a technology has received the most widespread, complete and comprehensive development in the West, especially in the USA, in South Korea and in Japan. This is primarily due to the strong financial support for this area from the government and investors, who predict a great future for it. Moreover, the government mainly supports the development of technology in educational centers, and investors provide support to private, highly promising companies. Most striking examples such well-funded scientific centers can serve as a laboratory Artificial Intelligence Massachusetts Institute of Technology (MIT Artificial Intelligence Laboratory), UC Berkeley Computer Vision Group, Vision and Autonomous Systems Center at Cornegie Mellon University, Stanford Vision Laboratory and several others. Examples of supported private companies include companies such as Visionics, Eyematic, etc. In total, the Internet site uniting developers in the field of computer vision is Computer Vision Home Page (http://www.2.cs.cmu.edu/afs /cs/project/cil/ftp/html/txtvision.html) - about 200 groups and scientific laboratories working on this issue are registered. It should be noted that this does not exhaust the range of organizations involved in computer vision, since there are a large number of commercial firms specializing in the field of computer vision and image processing. Information about them can be found on specialized thematic Internet sites dedicated to individual areas of this technology. In other words, developers of various technologies within computer vision technology itself seem to unite into clubs of interest. For example, those interested in advances in gesture recognition can find plenty detailed information about research, research groups, commercial applications, patents on the corresponding specialized Internet site - Gesture Recognition Home Page (http://www.cybernet.com/~ccohen/gesture.html). There you can download some demo applications and get acquainted with the latest scientific publications. If the reader prefers to engage in technologies related to facial recognition, then he has a direct route to the virtual club on another Internet site - Face Detection and Recognition Home Page (http://home.t-online.de/home/Robert.Frischholz/ face.htm).

It should be noted that all of the above leads to the rapid growth and improvement of computer vision technologies. Currently, foreign research and commercial centers attract a large number of scientists and highly qualified programmers, conduct parallel research in various fields of computer vision, achieving quite significant results.

Russia, as a full member of the world economic community, has not remained aloof from this process. For several years now, the Russian technology market has also seen a trend of increasing interest in computer vision problems, both from the heads of a number of IT companies and companies operating in the security market, and from consumers (users) and students who want to specialize in this areas. In response to this interest, laboratories, groups and commercial structures emerged that set themselves the task of developing various types of technologies and applications for solving computer vision problems. And if a decade ago we were in the role of catching up, today many companies - leaders in the field of advanced technologies are seeking to enter the Russian market in order to acquire relevant computer vision technologies or place orders for advanced research and development in this area.

This article is devoted to this topic, the purpose of which is not only to demonstrate the interest in this topic on the part of Russian and foreign commodity producers, but also to talk about a number of Russian companies developing software For various systems image processing and analysis.

Who's who in the Russian computer vision market

research Russian market developers of computer vision technology shows that the number of companies involved in computer vision is relatively small. Let's look at the most notable of these companies and give a brief description of some interesting computer vision technologies that they supply to the domestic and world markets.

SPIRIT Company

The most famous photogrammetric systems in the world include such hardware and software systems as Leica and Intergraph, supplied with powerful workstations. These are very expensive systems, and few companies can afford them. With the development of computer technology, less expensive systems that allow image processing on personal computers are becoming increasingly popular. Russian digital photogrammetric systems "Talka" (http://www.talka-tdv.ru/), Photomod (company "Rakurs" (http://www.racurs.ru/)), Z-Space (GosNIIAS), TsFS TsNIIGAiK (Roscartography) or “Photoplan” (29th Institute of the Ministry of Defense), not inferior, and sometimes superior, in the quality of digital video signal processing foreign analogues, while being tens of times cheaper than similar foreign developments. Consideration of the characteristics and capabilities of such systems is the subject of a separate article.

Another direction in the field of computer vision is the construction of character recognition systems. In this article, we only indirectly mentioned this area in which computer vision technologies can be considered mature. In particular, we considered only highly specialized problems solved by companies as part of commercial projects. If we talk about established commercial products and technologies for character recognition systems, then we cannot fail to mention the largest Russian and global suppliers of this technology - ABBYY with a series of FineReader programs and Cognitive Technologies with a series of CuneiForm programs. More than one article on the pages of ComputerPress is devoted to a review of the technologies supplied by these companies. Information about the achievements of these companies can be found in this issue of the magazine. Therefore, while paying tribute to these companies and their technologies, we only briefly mention them in this article.

To summarize, we can confidently say that Russian computer vision technologies are not inferior to, and in many ways superior to, foreign analogues. Often, companies developing these technologies lack worldwide famous name. Therefore, investments in them are, as a rule, reluctant. However, there is no doubt that high level technology and high qualifications Russian specialists in the near future will lead to Russian computer vision technologies dominating the world market.

ComputerPress 7"2002

Master's student. Mukhamediarov R.M.

Kazakh National Technical University K.I.Satpayeva, Almaty, Kazakhstan

Machine vision: concepts, tasks and applications

1. Basic definitions and concepts of computer vision

Machine vision is a scientific direction in the field of artificial intelligence, in particular robotics, and related technologies for obtaining images of objects real world, their processing and use of the obtained data to solve various kinds applied tasks without (full or partial) human participation.

Machine vision closely interacts with areas such as Computer vision, Image processing , Image Analysis , Pattern recognition, etc.There is also no standard formulation of how the problem in this area should be solved and howIt is often difficult to unambiguously attribute emerging problems and applied solution methods to one of these areas.If you review techniques,algorithms, image processing techniques that are used and ,developed in these fields can be seen to be more or less ,identical.

Machine vision focuses on mainly industrial applications, such as autonomous robots and visual inspection and measurement systems. This means that image sensor technology and control theory are related to video data processing to control the robot, and real-time processing of the resulting data is carried out in software or hardware.

Image processing And Image Analysis mainly focused on working with 2D images, i.e. how to convert one image to another. For example, pixel-by-pixel operations to increase contrast, operations to highlight edges, remove noise, or geometric transformations such as image rotation. These operations assume that image processing/analysis operates independently of the content of the images themselves.

Computer vision focuses on processing three-dimensional scenes projected onto one or more images. For example, restoring the structure or other information about 3 D scene from one or more images. Computer vision often depends on more or less complex assumptions about what is represented in images.

There is also an area called Visualization , which was originally concerned with the process of creating images, but sometimes dealt with processing and analysis. For example, radiography works with video data analysis for medical applications.

Finally, Pattern recognition is an area that uses various methods to obtain information from video data, mainly based on a statistical approach. Much of this area is devoted to practical application these methods.

Thus, we can conclude that the concept of “computer vision” today includes: computer vision, visual pattern recognition, image analysis and processing, etc.

Main elements modern systems Computer vision can be called a camera with which the image is obtained, an input board that digitizes the image, and a motion control board. Machine vision technology has several stages of system operation. The first step is to obtain an image of the controlled object. Next, the resulting image must be entered into an industrial controller or another computer, where computer processing, analysis of the received data and decision-making takes place in accordance with the embedded control program. The final stage is the output of control actions to actuators.

In general, the tasks of computer vision systems include obtaining a digital image, processing the image in order to highlight significant information in the image, and mathematical analysis of the obtained data to solve the assigned problems.

Literature:

1. Computer Vision: A Modern Approach by D. A. Forsyth and J. Ponce, Prentice Hall, Upper Saddle River, N.J., 2002

2. Computer Vision. L. Shapiro and G. Stockman, Prentice-Hall, Upper Saddle River, N.J., 2000

3. K. Ugh . Structural Methods in Pattern Recognition. Publishing house "Mir". Moscow, 1977.

4. Edward A. Patrick. Basic theory pattern recognition. Moscow "Soviet Radio", 1980.

5. Artificial intelligence. Modern approach. Stuart Russell, Peter Norvig. Moscow/St. Petersburg/Kyiv, 2006.

UDC 004.93"1

Machine vision

Tatyana Vadimovna Petrova, group 4241/3

Machine vision is the application of computer vision to industry and manufacturing. Areas of interest in machine vision are digital devices input/output and computer networks designed to control production equipment. Machine vision has some advantages over human vision. Accordingly, it is important to develop this area of science. This review describes the history of the development of computer vision, the components of a computer vision system, the application of computer vision and the future of this field of science.

Introduction

computer machine vision production

A person receives the bulk of information about the outside world through the visual channel and then very effectively processes the received information using the apparatus of analysis and interpretation of visual information. Therefore, the question arises about the possibility of machine implementation of this process.

Due to the increasing complexity of scientific and technical problems being solved, automatic processing and analysis of visual information are becoming more and more topical issues. These technologies are used in highly sought-after areas of science and technology, such as process automation, increasing productivity, improving the quality of manufactured products, control of production equipment, intelligent robotic systems, control systems for moving vehicles, biomedical research and many others. Moreover, it can be said that success modern business based mainly on the quality of the products offered. And to ensure this, if we talk about the production of material things, visual control is required.

Further we will use the term “machine vision” as a concept that most fully encompasses the range of engineering technologies, methods and algorithms associated with the task of interpreting visual information, as well as the practical use of the results of this interpretation.

1. History of the development of machine vision

Computer vision emerged as an independent discipline by the end of the 60s. This direction arose within the framework of artificial intelligence at a time when there were still heated debates about the possibility of creating a thinking machine. It emerged from work on pattern recognition. [Zueva, 2008]

A brief history of the development of machine vision is presented in Figure 1.

Rice. 1. History of machine vision

In the history of the development of machine vision, the following stages can be distinguished:

· 1955 - Massachusetts Institute of Technology (MIT) professor Oliver Selfridge published the article “Eyes and Ears for the Computer.” In it, the author put forward the theoretical idea of equipping a computer with sound and image recognition tools.

· 1958 - psychologist Frank Rosenblatt from Cornell University created a computer implementation of the perceptron (from perception - perception) - a device that simulates a pattern recognition circuit human brain. The perceptron was first modeled in 1958, and its training required about half an hour of computer time on an IBM-704 computer. The hardware version - Mark I Perceptron - was built in 1960 and was intended for visual image recognition [Computer Vision, 2010] .

However, the consideration of computer vision problems was rather speculative, since neither the technology nor the mathematical support for solving such complex problems was yet available.

· 1960s - the emergence of the first image processing software systems (mainly for removing noise from photographs taken from aircraft and satellites) began to develop applied research in the field of printed character recognition. However, there were still limitations in the development of this field of science, such as the lack of cheap optical data input systems, limited and rather narrow specialization computing systems. The rapid development of computer vision systems throughout the 60s can be explained by the expanding use of computers and the obvious need for faster and more efficient human-computer communication. By the early 60s, computer vision problems mainly covered the area of space research, which required processing a large amount of digital information.

· 1970s - Lawrence Roberts, a graduate student at MIT, put forward the concept of machine construction of three-dimensional images of objects based on the analysis of their two-dimensional images. At this stage, a more in-depth analysis of the data began. Various approaches to recognizing objects in an image have begun to develop, such as structural, feature, and texture.

· 1979 - Professor Hans-Helmut Nagel from the University of Hamburg laid the foundations of the theory of dynamic scene analysis, which makes it possible to recognize moving objects in a video stream.

· At the end of the 1980s, robots were created that were capable of more or less satisfactorily assessing the world around them and independently performing actions in natural environment

· The 80s and 90s were marked by the emergence of a new generation of sensors for two-dimensional digital information fields of various physical natures. The development of new measuring systems and methods for recording two-dimensional digital information fields in real time has made it possible to obtain time-stable images generated by these sensors for analysis. Improving the production technologies of these sensors has made it possible to significantly reduce their cost, and therefore significantly expand the scope of their application.

· Since the beginning of the 90s, in the algorithmic aspect, the sequence of actions for image processing has been considered in accordance with the so-called modular paradigm. This paradigm, proposed by D. Marr based on a long study of the mechanisms visual perception human, argues that image processing should be based on several successive levels of ascending information line: from the “iconic” representation of objects (raster image, unstructured information) to their symbolic representation (vector and attribute data in structured form, relational structures, etc. .). [Visilter et al., 2007]

· In the mid-90s, the first commercial automatic vehicle navigation systems appeared. Effective means computer analysis of movements was developed at the end of the 20th century

· 2003 - the first fairly reliable corporate systems face recognition.

2. Problems of computer vision and areas of its application

2.1 Definition of “machine vision”

Machine vision is the application of computer vision to industry and manufacturing. The area of interest of machine vision as an engineering field is digital input/output devices and computer networks designed to monitor production equipment such as robotic arms or defective product retrieval devices.

Machine vision is the study of methods and techniques whereby artificial vision systems can be constructed and usefully employed in practical applications. As such, it embraces both the science and engineering of vision .

Its study includes not only the software but also the hardware environment and image acquisition techniques needed to apply it. As such, it differs from computer vision, which appears from most books on the subject to be the realm of the possible design of the software, without too much attention on what goes into an integrated vision system (though modern books on computer vision usually say a fair amount about the "nasty realities" of vision, such as noise elimination and occlusion analysis).

2.2 Machine vision today.

Currently, there is a clear boundary between the so-called monocular and binocular computer vision. The first area includes research and development in the field of computer vision related to information coming from a single camera or from each camera separately. The second area includes research and development that deals with information simultaneously received from two or more cameras. Multiple cameras in such systems are used to measure the depth of observation. These systems are called stereo systems.

To date, the theory of computer vision has fully developed as an independent branch of cybernetics, based on a scientific and practical knowledge base. Every year hundreds of books and monographs are published on this topic, dozens of conferences and symposia are held, and various software and hardware are produced. There are a number of scientific and public organizations that support and cover research in the field modern technologies, including computer vision technologies.

2.3. Main tasks of computer vision

However, computer vision allows you to solve many problems, which can be divided into four groups (Fig. 2) [Lysenko, 2007] :

Fig.2. Computer vision tasks

· Position recognition

The purpose of machine vision is this application- determining the spatial location (the location of the object relative to the external coordinate system) or the static position of the object (in what position the object is located relative to the coordinate system with the origin within the object itself) and transmitting information about the position and orientation of the object to the control system or controller.
An example of such an application would be a loading and unloading robot, which is tasked with moving objects various shapes from the bunker. The intelligent task of machine vision is, for example, to determine the optimal basic system coordinates and its center to localize the center of gravity of the part. The information obtained allows the robot to properly grasp the part and move it to the proper location.

Overview of the computer vision technology market

The modern world of computer systems is difficult to imagine without machine or computer vision technologies. In the article “Why does a computer need vision?” (ComputerPress No. 5’2002) the history of the formation of this technology was reviewed and an overview of a number of its applications was given. Of course, the article describes only a small part of the applications from a wide range of used computer vision systems, and in future issues we will return to consider this very interesting and rapidly developing area of knowledge. Yes, it is rapidly developing. After all, this technology is only about 50 years old, which by the standards of many exact sciences does not go beyond its infancy. Increasing its scientific and practical potential in parallel with the improvement of computing and recording technology, computer vision is gradually conquering new technological frontiers. High-performance computers of the latest generation (including modern personal computers) already make it possible to solve many problems of processing streams of digital video information and making decisions in real time. And today, sometimes unnoticed by most of us, computer vision is quite firmly established in many areas of human life, helping him, and sometimes replacing him, relieving him of monotonous, routine or, often, life-threatening work.

It's no secret that computer vision as a technology has received the widest, most complete and comprehensive development in the West, especially in the USA, South Korea and Japan. This is primarily due to the strong financial support for this area from the government and investors, who predict a great future for it. Moreover, the government mainly supports the development of technology in educational centers, and investors provide support to private, highly promising companies. The most striking examples of such well-funded research centers are the MIT Artificial Intelligence Laboratory, UC Berkeley Computer Vision Group, Vision and Autonomous Systems Center at Cornegie Mellon University, Stanford Vision Laboratory and a number of others. Examples of supported private companies include companies such as Visionics, Eyematic, etc. In total, the Internet site uniting developers in the field of computer vision is Computer Vision Home Page (http://www.2.cs.cmu.edu/afs /cs/project/cil/ftp/html/txtvision.html) - about 200 groups and scientific laboratories working on this issue are registered. It should be noted that this does not exhaust the range of organizations involved in computer vision, since there are a large number of commercial firms specializing in the field of computer vision and image processing. Information about them can be found on specialized thematic Internet sites dedicated to individual areas of this technology. In other words, developers of various technologies within computer vision technology itself seem to unite into clubs of interest. For example, those interested in advances in the field of gesture recognition can find quite detailed information about studies, research groups, commercial applications, patents on the corresponding specialized Internet site - Gesture Recognition Home Page (http://www.cybernet.com/~ccohen/gesture. html). There you can also download some demo applications and check out the latest scientific publications. If the reader prefers to engage in technologies related to facial recognition, then he has a direct route to the virtual club on another Internet site - Face Detection and Recognition Home Page (http://home.t-online.de/home/Robert.Frischholz/ face.htm).

This article is devoted to this topic, the purpose of which is not only to demonstrate the interest in this topic on the part of Russian and foreign commodity producers, but also to talk about a number of Russian companies that develop software for various image processing and analysis systems.

Who's who in the Russian computer vision market

A study of the Russian market for computer vision technology developers shows that the number of companies involved in computer vision is relatively small. Let's look at the most notable of these companies and give a brief description of some interesting computer vision technologies that they supply to the domestic and world markets.

SPIRIT Company

The most famous photogrammetric systems in the world include such hardware and software systems as Leica and Intergraph, supplied with powerful workstations. These are very expensive systems, and few companies can afford them. With the development of computer technology, less expensive systems that allow image processing on personal computers are becoming increasingly popular. Russian digital photogrammetric systems "Talka" (http://www.talka-tdv.ru/), Photomod (company "Rakurs" (http://www.racurs.ru/)), Z-Space (GosNIIAS), TsFS TsNIIGAiK (Roscartography) or “Photoplan” (29th Institute of the Ministry of Defense), not inferior to, and sometimes superior to, foreign analogues in the quality of digital video signal processing, while being tens of times cheaper than similar foreign developments. Consideration of the characteristics and capabilities of such systems is the subject of a separate article.

To summarize, we can confidently say that Russian computer vision technologies are not inferior to, and in many ways superior to, foreign analogues. Often, companies developing these technologies lack a world-famous name. Therefore, investments in them are, as a rule, reluctant. However, there is no doubt that the high level of technology and high qualifications of Russian specialists will in the near future lead to dominance of Russian computer vision technologies in the world market.

ComputerPress 7"2002