Overview of Automatic Printing Quality Inspection Technology and 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 printing process is often interfered by various factors such as temperature, humidity, machine precision, equipment operation, etc. The printing quality does not meet the established requirements, which requires the detection and control of the entire printing process. Prints often have defects of one kind or another. Common print defects mainly include: missing marks, flying ink, color cast, black spots, scratches, and overprinting. However, since people are limited by their own conditions and cannot complete real-time monitoring, it is very important to establish effective automatic printing quality inspection technology. This paper takes this as the research object, studies the development status of automatic detection systems at home and abroad and the basic components of automatic detection systems.
1. Two core problems affecting printing quality and the necessity and feasibility of automatic detection technology Summarizing the most common printing quality problems in printing can be divided into two major categories: printing color problems and printing defects. Printing color problems are mainly color cast, and printing defects mainly represent shape defects.
Offset presses used in color printing typically have a monochrome offset press, a two-color offset press, and a four-color offset press. There are three main reasons for the color cast: First, the general principle of printing a color cover on a two-color offset press. When a two-color offset press is used to print a color cover, the two colors printed at the same time are wet and wet, and the two colors are printed. The printing method is wet-stacked relative to the first two colors. Second, the general principle of printing a color cover on a four-color offset press. When printing a color cover with a four-color press, it is a wet-wet printing method, which adds difficulty to the printing process and requires four colors of ink. The density value after overprinting should be small, and the thickness of the ink layer of each color should be thin, otherwise there will be a back dirty or other printing accident. Third, the relationship between the viscosity of the ink and the color. The viscosity of the ink refers to the size of the cohesive force of the ink or the resistance generated when the ink layer breaks. In multi-color printing, we should follow the printing order of the ink viscosity from large to small. Printing, that is, the viscosity of the ink of the previous color is greater than the viscosity of the ink of the latter color, so as to ensure the normal overprinting of the ink.
According to the analysis of the factors affecting the quality of printed matter, it is known that the defects are generally shape defects. It focuses on shape features such as line shape defects. Among the most available image quality tests, some are metrics suitable for evaluating image quality. These tests include: point quality, line quality, text quality, over-range spray and spatial resolution. Common print defects mainly include: flying ink, pinholes, missing marks, black spots, scratches, inaccurate overprinting, etc. At present, the detection of these defects is generally a manual visual measurement method, which is labor intensive, time-consuming and laborious, and the detection standards are not uniform. Especially with the increase of printing speed, it has gradually failed to meet the demand of production. Therefore, the automatic detection of print defects has gradually become a trend in the industry.
In order to improve production efficiency, the online intelligent detection technology introduces the printing process. The basic principle is as follows: The system collects the defect-free printed matter as a standard image through the CCD camera lens, and then collects the image to be inspected on the printing production line, and each frame will be collected. The image to be inspected is transmitted to the scene for comparison and analysis with the standard image to find the image with the quality problem, thereby discovering the quality problem of the printed matter corresponding to the frame image, and finally automatically adjusting the corresponding online quality control of the printing department. In the online implementation of printed products, it is generally divided into two steps: the first is the preparation of the test, that is, the image of the qualified product is obtained through image acquisition of the qualified product. Followed by the actual test, 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 based on the comparison result, and recording the defect information. The following is an analysis of the status quo of national research and its system composition.
2. Current status of automatic detection technology for printed matter at home and abroad The automatic detection technology for printed image quality started in the 1980s and 1990s. In 1990, Katsuyuki Tanimizu of Tokyo, Japan conducted an automatic quality inspection study in the printing industry, and proposed an Index Space Method for automatically detecting surface defects of printed matter. X and Y indicate the position coordinates of each pixel. The image point gray value is represented by the Z axis, and the spatial coordinate system is established, so that each map point can find its corresponding position in the coordinate system, by comparing the template image in the coordinate system with the gray position of the image to be inspected. The degree value determines whether there is a defect point in the image to be inspected. The image processing process and the detection process of the method are independent of each other, and can detect a more complicated picture. However, the algorithm is more complicated and there are many inconveniences in the application.
In 1993, France’s B. Mehenni also carried out research work on this subject. He proposed a method combining the n-tupe method and the pixel-by-pixel comparison method. This method has the characteristics of fast speed and multiple parameter output, but it requires special hardware equipment. At the same time, the task of automatic quality inspection can be completed through teaching.
In 1998, some scholars introduced the Gabor filter method into the quality inspection of printed images. This method can detect a variety of image defects and has certain adaptability. It is suitable for the detection of big data systems, but the abor method. There is a big flaw, the recognition speed is slow, because its good recognition is based on exact matching, which greatly increases the complexity of the operation and reduces its practicability.
In 2003, J. Luo and Z. Zhang of the University of Exeter in the United Kingdom proposed a color print detection algorithm based on image processing technology. The algorithm first performs illumination correction, then gives the color three-dimensional histogram for feature extraction. Finally, the neural network is used to classify the image and identify the qualified image.
At present, many foreign manufacturers have developed a variety of automated quality inspection systems for printed matter. For example, the printing quality inspection system developed by Vision Experts of Germany can detect printing errors on the surface of various papers and materials. The GED NOTA-SAVE series of high-precision printing quality inspection systems produced by ELTROMAT in Germany have been used for quality inspection of banknote printing. Practice has proved that they greatly shorten the detection time, improve the detection speed, and achieve the purpose of monitoring the production process of high-precision prints. Israel's AVT has also produced PrintVision products for detecting print quality, which can detect printing errors such as color differences, missing marks, streaks and spots.
At the practical application level, Japan and Germany are currently the main researches in this field in the world. Companies such as FUTEC in Japan and TOKIMEC in Japan have EasyMax series products and Print-Pac systems, and Print-Expert 4000 OCV/2 systems developed by Vision-Experts in Germany. In addition, we can provide automatic print quality inspection equipment, such as BOBST in Switzerland and PROIMAGE in the United States.
3. Online detection system composition In order to solve the problem of automatic defect detection in the printing process, the online detection system can be designed according to the characteristics of printing defects, which mainly consists of four parts: image acquisition, positioning, detection, and result output. The image is collected by CCD, lens, light source and video image. The acquisition card and the computer are composed, and the positioning is mainly through software programming to complete the image noise removal, geometric transformation and positioning determination. The image detection system mainly performs the automatic detection process on the printed product through the binarized image, and the output of the result is mainly calculated by the data conversion part. The obtained data is output, and display of the printed feature amount, such as the ink amount display. When the full-screen printing quality inspection system is established, the CCD camera is used to continuously take photos of the printed matter, and each frame of the captured image is transmitted to the on-site computer. The image processing software is used to analyze and process the image information to find out the quality. The image of the problem gives the quality of the print corresponding to the image, and then the information is fed back to the operator via the transmission line or directly fed back to the printer for adjustment. This will not only reduce the labor intensity of workers, but also reduce defective products and increase production efficiency.
The hardware part of the online detection system for print defects is shown in Figure 1. It is mainly composed of CCD, lens, image acquisition card, computer and its display system.
Figure 1 Print defect online system
First, the CCD, lens, light source and image acquisition card work together to capture and digitize the image. High-quality image information is the original basis for the correct judgment and decision-making of the system, and is the key to the success of the entire system. CCD devices can be divided into two types: line array and area array. The line CCD can only obtain one line of information at a time, and the object to be photographed must be moved forward from the camera in a straight line to obtain a complete image, so it is very suitable for image detection of objects moving at a constant speed at a constant speed. The area CCD can get the information of the whole image at one time. In the full-screen inspection system, Sony's Bayer-converted area array CCD is used.
After the experiment, for the small-format printed matter (200mm × 200mm), when the object distance between the printed matter and the lens is 15mm, the illumination of the light source is the most uniform and the imaging quality is good. Therefore, the focal length of the lens is 3.5-8mm, the imaging size is 1/3 inch, and the aperture is F1.4. The light source adopts a method of forward vertical illumination.
Again, the task of the image acquisition module is mainly to complete the control of acquiring images to the real-time display of the screen. The collection mode is divided into single frame acquisition and real-time acquisition and display. In the full-screen printing quality detection system, it is important to analyze the image data after binarizing the acquired image, which is a complicated process. Therefore, after storing the image captured in a single frame to a specified position, The image is called using the pointer of the image for processing.
Finally, the output system primarily performs the output of the results and adjusts the printing process based on the output data, including pauses, reprints, overlays, and the like.
4. Main hardware features
4.1 CCD camera. CCD (Change Coupled Device) is a new type of semiconductor optoelectronic imaging device developed in the 1970s. It is a special-purpose chip that uses the photoelectric effect principle to achieve image ingestion. There are two types of CCD wired and surface types, both of which require an optical imaging system to image the scene image on the CCD. An electrical imaging device is a specialized application chip that utilizes the principle of photoelectric effect to achieve image pickup. Here we choose the area array CCD, which arranges the photosensitive cells and displacement integrator of the one-dimensional linear array CCD into a two-dimensional array in a certain way. For the selected CCD camera, it is necessary to focus on six parameters: color, resolution, minimum illumination, CCD chip size, exposure mode (Exposure), shutter speed (Shutter).
4.2 optical lens. The lens is equivalent to the lens of the human eye. If there is no lens, the human eye cannot see anything. If there is no lens, then the image output by the camera is a piece of white, and there is no clear image output. When the camera is shooting an image, if the image becomes unclear, the back focus of the camera can be adjusted to change the distance between the CCD chip and the lens reference surface, so that the blurred image becomes clear.
4.3 image acquisition card. The image acquisition card is responsible for converting the analog video signal captured by the camera into a digital image signal for processing by the computer. Usually, the image capture card takes up one slot of the PC bus, and has an external CCD camera, image monitor, and video signal interface. The image acquisition card together with the camera, monitor, and PC constitutes the basic hardware environment of a typical microcomputer image processing system. After the signal enters the image acquisition card, it is divided into two paths. One channel is separated by the synchronous separator and the field synchronization signal is sent to the phase detector to keep the same relationship with the line and field synchronization signals generated by the card timing generator. The control circuit synchronizes the units on the card in accordance with the requirements of the video signal line and field TV system. The other video signal is subjected to a pre-processing circuit to amplify the gradation signal of the video from a standard television signal having a peak value of 1V to the amplitude required by the A/D converter, and adjust the level and contrast. The signal output from the pre-processing circuit is sent to the A/D converter to be converted into a digital signal. The timing controller stores the digital signal in a frame memory. At the same time, the card is provided with a full TV signal output unit for the analog monitor. It consists of a lookup table, a D/A converter, and a synchronous synthesis circuit. The lookup table puts the address of the same gray value in the digital image output by the A/D converter into a specified space under the control of the microcomputer interface circuit. These data are converted to analog voltage values by D/A, so that the output of the D/A converter can quickly restore the image to the video monitor by looking at the gray scale of the row and column specified by the table. Under the action of software, the image card can conveniently store, detect, add, subtract, etc. digital images. There are many types of image acquisition cards. According to different classification methods, there are black and white images and color image acquisition cards, analog signals and digital signal acquisition cards, and composite signals and RGB component signals input capture cards. When selecting an image acquisition card, factors such as the system's functional requirements, image acquisition accuracy, and matching with the camera output signal should be considered.