Analysis of automatic printed online detection system
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
The automation of the modern printing industry is getting higher and higher, the printing speed and printing quality are greatly improved, and the online control of the quality of the printing products is crucial for the management of the printing process, the elimination of waste and the improvement of the quality of the printed products. At present, most of the domestic print quality is detected by the human eye offline. The minimum defect that can be found by the human eye requires at least 20 gray level differences, and the size is not less than 0.3 mm, but it is difficult to maintain continuous and stable, and it is prone to fatigue. If problems are found in offline inspection, high-speed printing presses have already produced a large amount of waste products, resulting in waste of printing materials and working hours, and low production efficiency. Therefore, online quality inspection of printed materials came into being.
First, online print detection technology
On-line inspection technology of printed matter is to detect the printed matter online by means of online testing instruments in the printing process, and send the information to the central console through feedback loop in real time, thereby automatically adjusting the corresponding printing machine components to realize online control of print quality. , automatic detection of printed products. The principle is shown in Figure 1.
Second, the advantages of printed online detection technology
On-line inspection of printed matter is a fully automatic detection method suitable for high-speed printing by combining image processing technology with printing principle. Its advantages are mainly reflected in the following aspects:
1 can detect extremely fine defects. On-line inspection makes it easy to find defects of 0.1mm size, and only one gray level difference is needed.
2 will not damage the printed matter. On-line monitoring uses the camera as a sensor and does not come into contact with the printed matter, so it does not damage the surface quality of the printed product.
3 Improve production efficiency. Online testing can be continuously detected for a long time without false detection due to fatigue. At the same time, in the event of quality defects, the operator can timely solve problems in the work according to the real-time report in the field, reduce the scrap rate, and the manager can also track the production process according to the analysis report of the test results. Management benefits.
4 full process: adapt to the gravure printing, laser, bronzing, holography, fluorescence, watermark and other printing processes.
5 The application of online detection system can also reduce the labor intensity of workers, improve working conditions and so on.
Third, print online detection system
1. Defect types of printed products
Print defects mainly include shape defects and color defects. Shape defects mainly refer to defects, foreign objects, stains, knives, etc., which are expressed as the difference between the gray value of the defect image and the standard image. The gray value is compared with the standard image to determine whether the difference has exceeded a preset threshold range, and whether the image has a shape defect can be determined.
The color defect mainly refers to the deviation between the color of the printed product and the standard printed product. The color value conversion model can be used to obtain the chromaticity value CIE L*a*b* of the printed image, and the color difference is used to evaluate the color cast of the printed product.
2. The working principle of printed online detection system
The working principle of the online detection system: through the image acquisition unit (such as: CCD camera lens), first through the image acquisition of a certain number of qualified products, remove the influence of random factors, obtain the standard image of the printing layout as a template, and then on the printing production line The image to be inspected is collected, and each image to be inspected is transmitted to the scene for comparison and analysis with the standard image. According to the comparison result, it is determined whether the product on the production line meets the quality requirement, whether there is a defect and the position of the defect is determined. Find the image with the quality problem, find the quality of the print corresponding to the image, and finally adjust the corresponding printed parts to achieve online control of the quality of the print.
The specific implementation steps of printed online testing are divided into two steps: first, preparation, that is, through the image acquisition of qualified products, the standard printed image, also called template. The sampling process should pay attention to the scanning resolution, the spatial resolution of the image, and the image. The effects of spatial coordinate correction, color image hue, CCD speed, illumination conditions, and mechanical vibration on the sampled image. Secondly, the image of the printed matter to be inspected is compared with the standard template, thereby determining the presence or absence of the defect and the location of the defect according to the comparison result, and recording the defect information. The workflow is shown in FIG.
Various errors in the printing process are only different for the computer after the standard image is compared with the detected image, such as stains, ink dot chromatic aberrations and the like.
3. Online detection system
A typical online inspection system for printed products, including light source and optical system, CCD camera, image acquisition card, image processing software, computer and its peripherals, control execution module and so on. The schematic structure diagram is shown in Figure 3.
The working principle of the printed online detection system is: the object to be detected is collected into the computer through the CCD camera and the image acquisition card, and the image processing software processes, analyzes and interprets the obtained image, and then makes some decisions, and finally passes peripherals. Output test results or directly use the results to control industrial processes.
The following is a brief introduction to the role of each part, and has an overall understanding of the printed online detection system.
3.1 CCD camera.
Usually it is one or more cameras and lenses (optical components). The function of the camera is to take the printed matter to be tested, complete the acquisition of the target image, and convert the optical image received by the image sensor into an electrical signal that can be processed by the computer. High-quality image information is the original basis for the correct judgment and decision-making of the system, and is another key to the success of the entire system. The camera selection is mainly based on the camera's color, scanning system, resolution, scanning speed or frame rate, synchronization mode, communication interface and other characteristics.
3.2 light source.
It is used to illuminate the printed matter to be tested in order to capture a better image from the camera to ensure the consistency of the imaging environment. In the online inspection system of printed products, good light source and lighting solutions are often the key to the success of the whole system, playing a very important role, it is not simply to illuminate objects. The cooperation between the light source and the illumination scheme should emphasize the feature quantity of the object as much as possible. The difference between the part to be detected and the unimportant part should be as obvious as possible to increase the contrast; at the same time, sufficient overall brightness should be ensured. Changes in position should not affect the quality of the image. The choice of source must be in accordance with the required geometry, illumination brightness, uniformity, spectral characteristics of the illumination, etc., while also considering the luminous efficiency and lifetime of the source. Generally, high-frequency fluorescent lamps, LEDs, incandescent lamps, and quartz halogen lamps can be used.
3.3 image acquisition card.
The image acquisition card is an important device for controlling the camera to take pictures, complete image acquisition and digitization, and coordinate the whole system. Its main function is to collect the video data output by the camera in real time, establish the connection between the camera and the computer, and obtain the data (analog signal or digital signal) from the camera, and then convert it into information that the computer can process, and at the same time provide Camera control parameters. When selecting the image acquisition card in the online detection system, it is necessary to focus on the matching camera type, camera control function, data processing capability, and supporting software. There are many models of image capture cards that support different types of cameras and different computer buses.
3.4 computer platform
Computers are a key component of machine vision, and in testing applications, Pentium IV or higher CPUs are often used. In general, the faster the computer is, the shorter the time it takes for the vision system to process each picture. Industrial computers are generally required because of the frequent vibrations, dust, heat radiation, and the like in the manufacturing site.
Fourth, factors affecting online detection capabilities
1. Printing substrate
A non-textured substrate such as film or paper can be regarded as a defect by detecting an area different from the background. Some simple algorithms such as thresholding, zero-sequence background tracking, or delta trackers can be used to detect defects; for textures, the more complex the pattern, the more difficult it is to detect defects, while the unordered texture is more difficult to detect than repeated textures.
The defect detection of the roll material generally adopts the method of template matching, and the image to be tested is aligned with the defect-free template, and the difference is compared, but the difficulty of this method is accurate alignment and acquisition of the template. This is because the influence of tension and the reproducibility of the production process can not really obtain the ideal template, and the image to be tested can only match the template from the image contour, defect details and material deformation regardless of the image restoration algorithm or alignment algorithm. The details still cannot be separated.
For soft material stretchable prints, under the action of tension, a large degree of tensile deformation occurs along the walking direction, and the deformation is random, it is difficult to find an effective model to restore the image. The direct effect of this deformation on the image to be measured is that after the image to be detected is compared with the standard template, obvious misdetection occurs in the contour of the texture. Even if the image is of good quality, spots will appear. Many researchers adopt neighborhood comparison. Algorithm to solve the error problem,
2. Image processing networking
With the increase of observation area and the increasing complexity of detection tasks, the amount of data processing has increased dramatically. The single-machine system cannot meet the requirements of image display, data transmission, image processing and real-time control. Network-centric multi-view detection and distributed computing It has become the mainstream demand for modern automated production line quantity and quality inspection.
3. Information collection equipment
The cameras used for printing inspection are divided into two categories according to the actual situation, line array cameras and area array cameras. The linear array CCD is used to image one line at a time, and the CCD camera and the target image are in relative motion. One image needs to be combined with multiple exposures. The main advantage is that there is no deformation in the direction of motion, which is suitable for shooting wide-format targets. It is necessary to keep the light source and speed stable during the imaging process. Therefore, it is very suitable for image detection of objects (printed products) moving at a constant speed at a constant speed. When using the area array CCD, an image can be imaged by one exposure. There is no need for relative motion between the CCD camera and the target image during single imaging. The main advantage is that the exposure is completed once and the anti-interference is strong. The disadvantage is that the imaging exists. The spherical effect is not suitable for shooting wide format targets, like the digital cameras we use every day. Therefore, the surface scan camera can achieve the same vertical and horizontal resolution, but the observation range is small; the line scan camera uses line integration to sample, the observation range can be infinite, but it is not suitable for high-speed production environment.
Linear array CCD imaging method
4. Design of the light source
The basic requirements of image data acquisition are good linearity, low noise, high resolution and fast conversion speed. In the image data acquisition system, since the CCD image sensor is a photosensitive device, the intensity of the illumination directly affects the quality of the scanned image, and therefore the requirements for the illumination device are generally high. In different occasions, at different times or in different weather conditions, the natural light intensity fluctuates within a wide range, which greatly reduces the consistency of imaging, and may directly lead to a decrease in the recognition rate. Solving this problem is mainly to ensure the consistency of the imaging environment. In a sealed environment, a fixed light source can be used to ensure a stable light intensity. In the case of no sealing, a high-intensity strobe device can be used to shield natural light.
5. Image signal processing speed
Image signal processing is the core of online detection system. With the rapid development of computer technology, microelectronic technology and large-scale integrated circuit technology, in order to improve the real-time performance of the system, many tasks of image processing can be completed by hardware, such as DSP. Etc., the software mainly completes the part of the algorithm that is very complicated, not mature enough, and needs to be continuously explored and changed. At the time of image signal processing, it must be noted that in order to meet the real-time requirements of the system for continuous processing of images, it is necessary to make the processing time of one frame of image less than or equal to the acquisition time of one frame of image, that is, the image processing speed is greater than or equal to The speed of image acquisition.
PCI bus technology + MMX / SSE technology has become a key technology of the new generation of image processing systems, can use powerful computer resources to achieve fast, low-cost computing processing, but to achieve real-time real-time processing, you also need to provide dedicated acquisition hardware .
6. Equipment detection accuracy
For white cardboard and some minimalist prints, simple contrast testing is enough. Most domestic prints have many flashing elements, such as gold and silver cardboard, hot stamping, embossing or glazing, etc., and quality inspection equipment must have the ability to find gray level differences within five.
V. Conclusion
Enterprises with good and sustained development in the long run must have a reasonable and complete quality management system. Online control of print quality is critical to print process management, waste rejection and improved print quality. Since the on-line inspection technology of printed materials greatly reduces the human factors of print quality control, it greatly improves the production efficiency and reduces the requirements for the skills of operators. Its application provides the feasibility of realizing the automation and intelligence of printing machinery. Therefore, the printed online inspection system has broad development and application prospects, allowing us to focus more on the development of automatic print quality inspection equipment and technology.