On the application of machine vision system in pri

2022-08-14
  • Detail

Talking about the application of machine vision system in printing and packaging

1. Automatic printing quality detection

the detection system used by automatic printing quality detection equipment mostly uses high-definition, high-speed camera lens to shoot standard images, and sets certain standards on this basis; Then take the detected image, and then compare the two. The CCD linear sensor converts the light change of each pixel into an electronic signal. After comparison, as long as the detected image is different from the standard image, the system considers the detected image as unqualified. All kinds of errors in the printing process are only the difference between the standard image and the detected image for the computer, such as stains, ink dot color difference and other defects

the first technology used to detect the quality of printed matter is the gray comparison between the standard image and the detected image. Now the more advanced technology is the comparison based on RGB three primary colors. What is the difference between automatic machine detection and human eye detection? Take human vision as an example. When we focus on a printed matter, if the contrast color of the printed matter is relatively strong, the smallest defect that can be found by human eyes is the defect with obvious contrast color and no less than 0.3mm; However, it is difficult to maintain a sustained and stable visual effect by relying on human ability. However, in another case, if you are looking for defects in prints of the same color system, especially quality defects in a light color system, the defects that human eyes can find need to have at least 20 gray level differences. The automatic machine can easily find the defect with the size of 0.10mm, even if the defect is only one gray level different from the standard image

but in practical use, even the same panchromatic contrast system has different ability to distinguish color differences. Some systems can find defects with large changes in contour and color difference, while others can identify extremely small defects. For white cardboard and some simple style printed materials, such as the Kent cigarette label in Japan and the Marlboro cigarette label in the United States, simple detection may be enough, while most domestic printed materials, especially the speed labels of the oil volume control hydraulic material testing machine, have many characteristics, with too many flash elements guided by the major national strategies, such as gold and silver cardboard, hot stamping, embossing or polishing printed materials, This requires that the quality inspection equipment must have enough ability to find a very small gray level difference, perhaps five gray level differences, or a more stringent one. This is crucial to the domestic label market

the accuracy of the comparison between the standard image and the image of the inspected printed matter is the key problem of the detection equipment. Usually, the detection equipment collects the image through the lens. In the middle part of the lens range, the image is very clear, but the image of the edge part may produce phantom, and the detection result of the phantom part will directly affect the accuracy of the whole detection. From this point of view, if only the comparison of the whole area is not suitable for some fine prints. If the obtained image can be subdivided again, for example, the image can be divided into 1024dpi x 4096dpi or 2048dpi x 4096dpi, the detection accuracy will be greatly improved, and at the same time, because the phantom of the edge part is avoided, the detection result will be more stable

using testing equipment for quality testing can provide real-time reports and detailed and perfect analysis reports of the whole process of testing. The on-site operator can rely on the timely alarm of the full-automatic detection equipment and adjust the problems in the work in time according to the real-time analysis report. Perhaps the reduction will not only be one percentage point of the scrap rate. The manager can track the production process according to the analysis report of the detection results, which is more conducive to the management of production technology. Because the high-quality testing equipment required by customers is not only to detect the good and bad of printed matter, but also to have the ability of post analysis. What some quality testing equipment can do can not only improve the qualified rate of finished products, but also help manufacturers improve their process flow, establish a quality management system, and achieve a long-term stable quality standard

2. Position control and product detection of gravure printing machine

the video image of the printed product is continuously captured by the camera set on the production line, and the camera speed is less than 30 frames/s and adjustable. The image collected by the camera is first quantified, and the analog signal is converted into digital signal, from which a key frame effectively representing the content of the lens is extracted and displayed on the display. For a frame of image, the analysis method of still image can be used to process. Through size measurement and multispectral analysis, each color code on the video image can be identified, and the color code spacing and color parameters of the color code and some other related information can be obtained. Due to various factors, there will be a variety of noise, such as Gaussian noise, salt and pepper noise and random noise

noise brings many difficulties to image processing. It has a direct impact on image segmentation, feature extraction and image recognition. Therefore, the images collected in real time need to be filtered. Image filtering requires that the noise outside the image can be removed, while maintaining the details of the image. When the noise is Gaussian noise, the most commonly used is linear filter, which is easy to analyze and implement; However, the filtering effect of linear filter on salt and pepper noise is very poor. The traditional median filter can reduce the salt and pepper noise in the image, but the effect is not ideal, that is, the fully dispersed noise is removed, and the noise close to each other will be retained. Therefore, when the salt and pepper noise is serious, its filtering effect is obviously worse. This system is an improved median filtering method. This method first obtains the median value after removing the maximum and minimum gray value pixels in the noisy image window, then calculates the difference between the median value and the corresponding pixel gray value, and then compares it with the threshold value to determine whether to replace the gray value of the pixel with the obtained value

image segmentation in this stage, each color mark is detected and separated from the background. The edge of the object is reflected by the gray discontinuity. The type of L edge can be divided into two kinds. One is the step edge, and the gray values of pixels on both sides are significantly different; The second is the roof edge, which is located at the turning point l of the gray value from increase to decrease. For the step edge, the second-order directional derivative is zero crossing at the edge, so the differential operator can be used as the edge detection operator. The differential operator type edge detection method is similar to the high pass filter in the high space domain, which has the effect of increasing high-frequency components. This kind of operator is quite sensitive to noise. For the step edge, the commonly available operators are gradient operator, Sobel operator and Kirsh operator. Laplace transform and Kirsh operator can be used for roof edge. Because the color code is rectangular and the gray level difference between adjacent edges is large, edge detection is used to segment the image. Here, the Sobert edge sub is used for edge detection. It uses the local difference operator to find the edge, which can better separate the color code. In the actual detection process, the color image edge detection method is used, and the appropriate color base (such as intensity, chroma, saturation, etc.) is selected for detection. According to the type characteristics of the printing machine, that is, the color and layout characteristics of each color of the printing machine, multi threshold processing is carried out to obtain the binary map of each color

measure the segmented image and identify the object through the measured value. Since the color code is a rectangle with regular shape, the following features can be extracted: (1) calculate the rectangular area from the pixels, (2) rectangularity, (3) chromaticity (H) and saturation (s), and then obtain the spacing between the color codes according to the number of pixel points at the interval of each color code. Compare with the set value to obtain the difference between the two. A total of M measurements are made and the average difference is taken, Provide the corresponding adjustment signal to the digital AC servo adjustment part. To adjust the relative position of the color roller, so as to eliminate or reduce the printing dislocation. In feature extraction, multispectral image analysis of the image can quantitatively represent the color code, such as the color of pixels in the color number image. Two parameters of the color information of each color code, chromaticity and saturation, are obtained in his format to detect the quality of the ink. Then carry out statistical calculation on the binary images of various colors or match the template with the standard graphics, and measure the parameters such as ink chips in the printing process

the printing machine passes through each printing unit in turn by unwinding operation of the uncoiler for printing and drying of various colors, and is rewound by the rewinder. Each color printing will be printed with a color mark for chromaticity on the edge of the printing material. The horizontal length of the color mark line is 10 mm, and the width is 1 mm. The mark lines of each adjacent color should be parallel to each other when the overprinting is accurate, and the vertical (longitudinal) distance is 20 mm, The camera set on the production line continuously captures the video image of the tubing printing products that need to be replaced with higher intensity. Through dimensional measurement and multispectral analysis, each color mark on the video image can be identified, and the color mark spacing and color parameter l of the color mark can be obtained. If the spacing between the adjacent two color marks is greater than or less than 20 mm, it indicates that there is a deviation in overprint. The deviation signal is sent to the servo variable-frequency drive unit to drive the AC servo motor to make the corresponding chromatic correction roller ml move up and down to extend or shorten the stroke of the printing material from the previous unit printing plate roller to the unit printing plate roller for dynamic correction

3. Application in modern packaging industry

in the automatic production of modern packaging industry, various inspections and measurements are involved, such as the printing quality inspection of beverage bottle caps, bar codes and character recognition on product packaging, etc. The common characteristics of these applications are continuous mass production and high requirements for appearance quality. Usually, this highly repetitive and intelligent work can only be completed by manual inspection. We often see hundreds or even more than thousands of inspection workers performing this process behind the modern assembly line of some factories. While adding huge labor and management costs to the factories, it still cannot guarantee a 100 inspection qualification rate (i.e. zero defects). Today's competition among enterprises, Even 0.1 defects are not allowed. Sometimes, such as accurate and rapid measurement of small size, shape matching, color recognition, etc., it is simply impossible to carry out continuous and stable with the human eye, and other physical quantity sensors are also difficult to be used. At this time, people began to consider the rapidity, reliability and repeatability of the computer, thus introducing the robot vision technology

generally speaking, first, CCD camera is used to convert the captured target into image signal, which is transmitted to a special image processing system, and then converted into digital signal according to pixel distribution, brightness, color and other information; The image system performs various operations on these signals to extract the characteristics of the target, such as: area, length, quantity, position, etc; Finally, output the results according to the preset tolerance and other conditions, such as size, angle, offset, number, qualified/unqualified, yes/no, etc. Machine vision is characterized by automation, objectivity, non-contact and high precision. Compared with the general image processing system, machine vision emphasizes precision and speed, as well as reliability in the industrial field environment

machine vision is very suitable for measurement, inspection and identification in the process of mass production, such as the identification of characters printed on the IC surface, the identification of production date on food packaging, and the inspection of label placement position.

Copyright © 2011 JIN SHI