In modern industrial automation production, a variety of inspection, production monitoring and parts identification applications are involved, such as dimensional inspection of batch processing of spare parts, integrity inspection of automatic assembly, and automatic positioning of components in electronic assembly lines. Character recognition on 1C, etc. Usually the human eye cannot complete these highly repetitive and intelligent tasks continuously and stably, and other physical sensors are also difficult to use.

Therefore, people began to consider using the photoelectric imaging system to collect the image of the controlled target, and then digitally process it through a computer or a dedicated image processing module. Wait half way. In this way, the rapidity and repeatability of computers are combined with the highly intelligent and abstract capabilities of human vision, and the concept of machine vision is born.

A successful machine vision system is one that has been meticulously engineered to meet a defined set of requirements. When these requirements are fully determined, the system is designed and built to meet these precise requirements.

Machine vision concerns include the following:

High precision

As an accurate measuring instrument, a well-designed vision system is capable of spatially measuring one of a thousand or more parts. Because no contact is required for this measurement, there is no wear and no danger to weak components.

continuity

The visual system can save people from the drudgery of domestic work. Because there is no human operator, there is no human-induced operational changes, and multiple systems can be set to run independently.

cost-effective

As the price of computer evaluation devices has just dropped, the rate of machine learning and connection costs has also become more and more high. Each tester needs a quarterly salary of $20,000 per year, and the operation and maintenance of the video competition system is very low.

flexibility

The vision system can perform a variety of different measurements. When the application changes, it is only necessary to change or upgrade the software to adapt to the new requirements.

Many companies that apply flow process control (PC) are considering the application of machine vision speed to deliver accurate, hot inquiry and precise measurement 5PC commands. The control of the entire discussion is to ensure that this is within a certain range. This enables manufacturers to adjust process parameters if the production process gets out of control or a bad part occurs.

The matrilineal connection of the machine agency has a better communication response than the semi-intensive machine transmission agency. They make the machine of the company have multiple parts, flexible and reproducible parts. When the production process needs to be changed, the machine will be accepted. To bring "tools to hold" is only a change of software, and there is no replacement for expensive hardware. Vision systems can often be reused when production lines are restructured.

The composition of the machine vision system

Machine standardization technology uses a computer to analyze an image, and draws a conclusion based on the analysis. It is understood that the machine has two uses. The machine has to be controlled by the system to extract and measure parts. Targeting and serving parts can make effective decisions by knowing when to discard them; machine vision can also be used to create a part that uses a combination of complex optics and software to directly guide the manufacturing process.

Although machine vision applications vary, they all include the following processes;

Image Acquisition

The optical system captures the image, which is converted to analog format and transferred to computer memory.

Image Processing

The processor applies different algorithms to improve image elements that are important to the conclusion.

Feature extraction

The processor identifies and quantifies key characteristics of the image, such as the location of holes on a printed circuit board or the number of pins on a connector. These data are then passed to the control program.

Judgment and Control

The processor's control program draws conclusions based on the received data. For example; these data include whether holes in a printed circuit board are within required specifications or how an automated machine must move to pick up a part.

Machine Vision System Analysis

A typical vision system generally includes: light source, optical system, camera, image processing unit (or image acquisition card), image analysis and processing software, monitor, communication/input and output unit, etc.

Image Acquisition

The acquisition of the image is actually to convert the visual image and intrinsic characteristics of the measured object into data that can be processed by the computer, which directly affects the stability and reliability of the system. Generally, a light source, an optical system, a camera, and an image processing unit (or an image capture card) are used to obtain the image of the measured object.

light source

The light source and an important factor affecting the input of the machine vision system, because it directly affects the quality of the input data and at least 30% of the application effect. Since there is no universal machine vision lighting equipment, for each specific application instance, the corresponding lighting equipment should be selected.

setting for best results. Many industrial machine vision systems use visible light as the light source, mainly because visible light is readily available, inexpensive, and easy to operate. Several commonly used visible light sources are incandescent lamps, fluorescent lamps, mercury lamps and sodium lamps

However, one of the biggest drawbacks of these light sources is that the light energy cannot be kept stable. Taking fluorescent lamps as an example, the light energy will drop by 15% in the first 100 hours of use, and the light energy will continue to decline as the use time increases. Therefore, how to make light energy in a certain degree

It is an urgent problem that needs to be solved in the practical process. On the other hand, ambient light will change the total light energy irradiated by these light sources on the object, resulting in noise in the output image data. Generally, the method of adding a protective screen is used to reduce the influence of ambient light. by dry

The above problems exist. In today's industrial applications, invisible light such as X-rays and ultrasonic waves are often used as light sources for some demanding detection tasks.

The lighting system composed of light sources can be divided into: back lighting, forward lighting, structured light and stroboscopic lighting according to their illumination methods. Among them, back lighting is that the object to be measured is placed between the light source and the camera, and its advantage is that it can obtain high-contrast images; forward lighting is

The light source and the camera are located on the same side of the object to be measured, which is easy to install; structured light illumination is to project a grating or a line light source onto the object to be measured, and demodulate the three-dimensional image of the object to be measured according to the distortion produced by them. Information, stroboscopic lighting is the irradiation of high-frequency light pulses to objects

Physically, the scanning speed of the camera is required to be synchronized with the stroboscopic speed of the light source.

Optical system

For machine vision systems, the image is the only source of information, and the quality of the image is determined by the proper choice of optics. Usually, errors due to poor image quality cannot be corrected with software. Machine vision technology combines optical components with imaging electronics and uses a computer-controlled system to distinguish, measure, classify, and detect components that are passing through an automated processing system. Machine vision systems can often detect processed products up to 100% faster without slowing down the production line. This capability is important as more and more manufacturers are requiring "6-sigma" (less than three parts per million effective units) results in order to be more competitive in today's quality-conscious market . Additionally, these systems are ideally suited for Satisfactory Process Control (SPC). The main parameters of the optical system are related to the format of the photosensitive surface of the image sensor, generally including: aperture, field of view, focal length, F number, etc.

camera

The camera is actually an optical converter that converts the optical image received by the image transmitter into a computer signal that can be processed by the computer. Light is the core device of the camera. At present, the light of the warped torture is converted into a Baikonti image by switching the device

tube, CCD, CMOS image sensor, etc.

The direct-air TV camera tube is composed of two parts: the camera switch enclosed in the glass tube, etc., and the electron gun. "Camera" converts the light distribution of the input optical image into the one-dimensional spatial distribution of the corresponding pixel charges on the target surface, and mainly completes the photoelectric conversion and the task of applying and storing leaves; the electron gun completes the scanning and picking up process of the image signal. TV camera tube imaging The system has the characteristics of high definition, high sensitivity, wide spectrum and high frame rate imaging, but the dry TV camera tube is a dry vacuum tube device, and its weight, volume and power consumption are relatively large.

CCD is currently the most commonly used image sensor in machine vision. It integrates photoelectric conversion, charge storage, charge transfer, and signal reading, and is a typical solid-state imaging device. The outstanding feature of CCD is to use electric charge as a signal, while different devices use current or voltage as a signal. Such imaging devices form charge packets through photoelectric conversion, which are then transferred under the action of driving pulses to amplify and output image signals. A typical CCD camera consists of an optical lens, a timing and synchronization signal generator, a vertical driver, and an analog/digital signal processing circuit.

As a functional device, CCD has the advantages of no burns, no lag, low voltage operation, and low power consumption compared with vacuum tubes.

The development of CMOS (Complementary Metal Oxide Semiconductor) image sensors first appeared in the early 1970s. In the early 1990s, with the development of very large scale integration (VLSI) manufacturing process technology, CMOS image sensors

to rapid development. CMOS image sensor integrates photosensitive element array, image signal amplifier, signal reading circuit, analog-to-digital conversion circuit, image signal processor and controller on one chip, and also has the advantage of programming random access of local pixels. At present, CMOS image sensors are widely used because of their good integration, low power consumption, wide dynamic range and almost no smear in the output image.

Image processing and analysis

In the machine vision system, the main function of the camera is to convert the light signal received by the photosensitive element into a voltage amplitude signal output. In order to obtain the digital signal processed and recognized by the computer, the video information needs to be quantized. Frame grabber is an important tool for quantifying video information.

Image acquisition/processing card

The image acquisition card mainly completes the digitization process of the analog video signal. The video signal is first filtered by a low-pass filter and converted into an analog signal that is continuous in time; according to the requirements of the application system for image resolution, the sampling/holding circuit must be used to measure the time interval of the frontier video signal. The video letter is an analog letter that has been converted to a pro-dissertation. Then it is converted into a digital signal output by the A/returner. While the image acquisition/external processing card is in the right mode, the new conversion function can also be used to analyze the video image, and the external processing function can effectively control the camera at the same time.

image processing software

In the machine vision system, the processing technology of visual information mainly relies on dry image processing methods, which include image enhancement, data encoding and transmission, smoothing, edge sharpening, segmentation, feature extraction, image recognition and understanding. After these processing, the quality of the output image is improved to a considerable extent, which not only improves the visual effect of the image, but also facilitates the computer to analyze, process and recognize the image.

Application of Machine Vision System

The machine vision system is an effective way to achieve precise control, intelligence and automation of instruments and equipment, and can be called the "machine eyes" of modern industrial production. Its maximum advantage is

(1) Realize non-contact measurement. It will not cause any damage to the observer and the observed person, thus improving the reliability of the system;

(2) It has a relatively narrow spectral response range, and the machine vision can use special photosensitive elements, which can observe the world that cannot be seen by human meters, thus expanding the scope of human meters.

(3) Working long hours. It is difficult for humans to observe the same object for a long time. Machine vision systems can perform observation, analysis and identification tasks for a long time, and can be applied to harsh working environments.

The article is excerpted from: China Machine Vision Network