Application of machine vision in printing quality inspection
We are a big printing company in Shenzhen China . We offer all book publications, hardcover book printing, papercover book printing, hardcover notebook, sprial book printing, saddle stiching book printing, booklet printing,packaging box, calendars, all kinds of PVC, product brochures, notes, Children's book, stickers, all kinds of special paper color printing products, game cardand so on.
For more information please visit
http://www.joyful-printing.com. ENG only
http://www.joyful-printing.net
http://www.joyful-printing.org
email: info@joyful-printing.net
First, the working principle of machine vision inspection system
1. The meaning of machine vision
Machine Vision, as its name implies, combines the power of an automatic control machine with visual sensing. The image says that people perceive changes in the outside world through their eyes, then analyze and process them through the brain, make judgments, and finally complete the movements by hands and feet. Machine vision is the process of passing this process through an automated control device with computing power. Specifically, machine vision replaces the human eye with a camera and uses a computer instead of the human brain to make measurements and judgments. Machine Vision System refers to the conversion of an object being captured into an image signal by a machine vision product (ie, an image capture device, divided into CMOS and CCD), and transmitted to a dedicated image processing system, according to pixel distribution and brightness, Information such as color is converted into a digitized signal; the image system performs various operations on these signals to extract features of the target, such as area, length, number, position, etc., and then outputs the results according to preset tolerances and other conditions, such as: Size, angle, offset, number, pass/fail, presence/absence, etc. Finally, the device action at the site is controlled based on the result of the discrimination.
Machine vision technology is an interdisciplinary subject involving many fields such as artificial intelligence, neurobiology, psychophysics, computer science, image processing, and pattern recognition. It is characterized by high speed, high precision, high degree of automation and intelligence, non-contact, objectivity, high reliability, large amount of information, many functions, and high production efficiency. Because machine vision has these characteristics, in modern automated production processes, machine vision systems are widely used in the fields of condition monitoring, finished product inspection and quality control, and are widely used in food and beverage, cosmetics, pharmaceuticals, building materials. And in the chemical, metal processing, electronics manufacturing, packaging and printing, automotive manufacturing and other industries.
Machine vision is used for quality inspection in the printing and packaging industries. The basic working principle is to capture (acquire) the image on the printed matter with a camera and compare it with the standard image (template) of the printed product in the computer. If the difference is found, And beyond the set tolerance range, that is, judged to be a non-conforming product. The actual machine vision system is much more complicated.
2. Structure and working principle of visual inspection system
The visual inspection system is mainly composed of four parts: illumination light source, optical lens, camera, image information processing and actuator.
2.1 light source and lighting system
Illumination light source and lighting system is one of the key components of the entire visual inspection system, playing a very important role, it is not simply to illuminate objects. First of all, the illumination source should be able to truly reproduce the color of the print, and it can not cause color distortion. This requires that the light spectrum of the light source should be close to daylight, and at the same time ensure sufficient brightness and brightness in the entire illumination area. Uniformity; secondly, the design of the light source and illumination system should avoid some special areas with high reflectivity on the printed products, such as bronzing, laser anti-counterfeiting marks, etc., which cause the collected images to overflow, resulting in reduced detection accuracy or no detection at all; In some applications, it should also be considered that the cooperation of the light source and the illumination scheme should highlight the feature quantity of the object as much as possible, and the difference between the part to be detected and the unimportant part should be made as much as possible to increase the contrast. At the same time, sufficient overall brightness should be ensured, and changes in the position of the object should not affect the quality of the image. Transmitted and reflected light is commonly used in machine vision applications. For the reflected light, the relative position of the light source and the optical lens, the texture of the surface of the object, the geometry of the object, the background, etc. should be fully considered. In short, when selecting the light source and designing the lighting system, the geometry, illumination brightness, uniformity, spectral characteristics, luminous efficiency and service life of the required light source should be determined according to the specific conditions of the printed product.
2.2 optical lens
The optical lens is equivalent to the lens of the human eye. The quality of the optical lens affects the ability of the visual inspection system to resolve objects, affecting the contrast of the image and the extent of clear imaging. The quality of a lens's imaging quality, that is, its excellent or bad for aberration correction, can be measured by the aberration size. Common aberrations include spherical aberration, coma, astigmatism, field curvature, distortion, and chromatic aberration. .
2.3 camera
We know that the optical lens passes the optical information of the object being photographed, and in the computer is the electrical information of the object being photographed, then the camera is the device that completes the conversion from optical information to electrical information.
A. Working principle
In cameras, the key device is a chip called a vision sensor. The cells like small pools are neatly arranged on the chip, but they are not water, but electric charges. Light that carries the optical information of the subject passes through the lens and reaches the surface of the chip, and the optical information becomes a charge signal, that is, exposure. There is a lot of electric charge in the place where the light is strong, and there is less electric charge in the weak place, which corresponds to the brightness of different areas on the printed matter. Different colors are decomposed into three primary colors of red, green and blue, which are stored in different units. The charge signal generated by the exposure is processed by amplification, analog-to-digital conversion, etc. in the scanning order, and finally output from the camera.
B. Classification
The cameras used for print quality inspection can be divided into two categories according to the actual situation, line array cameras and area array cameras. A line-array camera can only obtain one line of information at a time, and the object to be photographed must pass back and forth from the front of the camera in a straight line to obtain a complete image, so it is very suitable for image detection of an object moving at a constant speed at a constant speed. The area camera can get the information of the whole image at a time, so the camera speed is faster. For example, we use digital cameras for everyday use.
The photoelectric conversion device is the core device that constitutes the camera. At present, typical photoelectric conversion devices mainly include CCD and CMOS image sensors. CCD is currently the most commonly used image sensor for machine vision. It integrates photoelectric conversion and charge storage, charge transfer, and signal reading. It is a typical solid-state imaging device. The outstanding feature of CCD is that it is a signal of charge, and its device is a signal of current or voltage. Such an imaging device forms a charge packet by photoelectric conversion, and then transfers and amplifies the output image signal under the action of a driving pulse. A typical CCD camera consists of an optical lens, timing and sync signal generator, vertical driver, and analog/digital signal processing circuitry. The development of CMOS (Complementary Metal Oxide Semiconductor) image sensor first appeared in the early 1970s. In the early 1990s, with the development of very large scale integrated circuit (VLSI) manufacturing process technology, CMOS image sensors were rapidly developed. The CMOS image sensor integrates a photosensitive element array, an image signal amplifier, a signal reading circuit, an analog-to-digital conversion circuit, an image signal processor and a controller on one chip, and has the advantage of program random access of local pixels. At present, CMOS image sensors have been widely used due to their good integration, low power consumption, wide dynamic range and almost no smear of output images.
C. Printed product requirements for cameras
It can be seen from the working principle of the above cameras that if the system is to be able to make judgments and decisions correctly, it must be required: a. First, the quality of the original image information is high; b, the camera must have sufficient dynamic range, that is, on the printed matter. The very bright and very dark parts can be accurately reflected; c, the camera should have sufficient resolution, that is, the tiny details on the printed product can be clearly reproduced; d, the camera should be able to truly reproduce the color of the printed product; e. The read speed of the camera should be consistent with the forward speed of the print to ensure sufficient accuracy and 100% acquisition.
2.4 image information processing and execution mechanism
The processing and execution of image information, the computer, is the core of the machine vision system, which is equivalent to the human brain. The processing of the entire image information usually includes:
A. Pretreatment. Compared with the template image, the high-speed moving printing often produces positional movement, rotation, and sometimes even fluttering, which causes the captured image to be distorted; the instability of the light source and the influence of the external environment make The acquired images are uneven in brightness and the color changes; other external devices, power supplies, and acquisition devices themselves introduce certain interference and noise. For this reason, in the process of installing the vision system, on the one hand, these unfavorable factors are eliminated, and on the other hand, image pre-processing is used to eliminate and dilute their influence. Common image preprocessing operations include image enhancement, data encoding and transmission, smoothing, edge sharpening, segmentation, and more. After these treatments, the quality of the image has been 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.
B. Feature extraction and recognition;
C. Measurement. In a concise sentence, it is "to find out the difference compared to the template."
In summary, machine vision is a technology that combines light, machine, electricity and computer. It is very suitable for repeatability measurement, inspection and identification in mass production.
Second, the application of machine vision in the printing and packaging industry
1. Online detection and offline detection
The visual inspection system can be divided into an online detection system and an offline detection system according to the different carriers installed therein.
The on-line inspection system is installed on printing equipment such as offset presses, gravure presses, flexo presses, and printers to detect print quality in real time. The print can be a single sheet of paper or a roll of paper. The visual inspection part of the online detection system and the offline detection system are basically the same in function, and the difference mainly lies in the treatment method of the non-conforming product. When a defect is found, the on-line inspection system usually records the location of the defect (web: length, sheet: number of sheets), or marks the edge of the paper using a printer, labeler, or marking machine. , the number detection records the error number or the number of the adjacent sheet.
Off-line inspection systems are usually installed on inspection machines or winders. In addition to the position recording and marking function of the online inspection system, the inspection machine usually also has a binning function, that is, automatically and badly distribute the bad and bad products to different bins according to the detection result.
2. Quality analysis, tracking and management
The quality detection by the detection system can provide real-time alarms and detailed and complete analysis reports throughout the whole process. The field operator can also perform the corresponding problems in the work according to the real-time alarm and analysis report of the automatic detection system. Adjustment. And the manager can also follow the analysis report of the test results to track the production process, which is more conducive to the management of production technology. That is to say, the quality inspection equipment can not only improve the qualified rate of the finished product, but also assist the manufacturer to improve the process flow, establish a quality management system, and achieve a long-term stable quality standard.
Third, the promotion and popularization of machine vision inspection in the printing and packaging industry
At present, although China is a printing country, it is not a printing power. To go to the world, we must use advanced printing equipment and testing equipment. The installation of an automated visual inspection system, although a certain amount of investment in the early stage, but its advantages in quality testing, will bring long-term benefits to the quality and labor costs of enterprises.
Visual quality inspection can be applied to prepress, printing and postpress processing, but most applications today focus on post-print quality inspections. There are tens of thousands of printing companies in China, but there are very few real-use detection systems. In addition to special industries such as printing money and tickets, it is mainly in the packaging industry. The packaging industry has higher requirements for printing, and the requirements for quality and color are special. This requires packaging and printing plants to have unique capacity advantages, technical advantages and quality advantages in order to meet the customer's comprehensive requirements for printing, cycle and quality.
To promote machine vision inspection, we must first make printing and packaging companies more aware of machine vision. Where to install, how to install, how to achieve the best results, and what specific benefits can be brought about, all of which need to be continuously popularized and publicized. Practitioners in the printing and packaging industry should propose their own specific needs, learn and master the concepts and operations of machine vision inspection; practitioners in the machine vision industry must understand the printing process, production process, operation methods and quality standards, only the two are closely integrated. In order to produce an automated visual quality inspection system that truly conforms to China's national conditions, the fundamental purpose of improving quality, improving efficiency and reducing costs can be achieved.