A method for overall inspection of a digital printing apparatus

Abstract

This invention discloses a whole-machine inspection method for digital printing equipment, including steps a) printhead spacing detection; b) drying status detection; c) printing effect detection; d) gap effect detection; and e) product inspection. Step a) Printhead spacing detection: The position image of the printheads is captured using an industrial camera. This invention uses a computer to determine color. Computer color recognition technology is unaffected by external factors such as light, noise, and environment, thus enabling more accurate color identification and classification. Furthermore, it does not require pixel-level image processing, allowing for faster processing and identification of color information. In contrast, the human visual system has limitations in color recognition, such as weak ability to distinguish certain colors or difficulty differentiating similar colors. This method offers advantages such as high accuracy, fast processing speed, wide applicability, and strong customizability in color differentiation.

Description

Technical Field

[0001] This invention relates to the field of digital printing inspection technology, specifically to a method for inspecting the entire digital printing equipment. Background Technology

[0002] Digital printing inspection is a crucial step in ensuring the quality of printed materials. Appropriate inspection standards are developed based on the requirements and specifications of the printed materials. These standards may include specific indicators related to color, contrast, printing precision, and image sharpness. The color difference values ​​of each color in the printed material are compared with the prescribed standards to determine whether they meet the requirements.

[0003] According to the patent number 201910686332.2 published on the China Patent Network, the patent title is: A method for visually inspecting printing quality, which includes the following steps: First, designing a grid-like screen printing pattern with a spacing of 0.05-0.3mm between adjacent grids; second, placing the designed pattern on the area of ​​the screen printing plate that needs to be monitored; and then, copying the pattern onto the coated silicon wafer through printing. This method for visually inspecting printing quality allows for the judgment of the printing quality of fine grid lines by visual inspection.

[0004] However, existing digital printing inspection methods mainly rely on visual inspection, which is prone to errors during the inspection process. Furthermore, subtle differences cannot be effectively judged by the human eye, and manual inspection is slow, resulting in extended digital printing preparation time and affecting processing efficiency.

[0005] Therefore, it is necessary to redesign and modify the overall inspection method for digital printing equipment. Summary of the Invention

[0006] To address the problems mentioned in the background art, the present invention aims to provide a whole-machine inspection method for digital printing equipment, which has the advantage of improving the efficiency of whole-machine inspection. It solves the problems of existing digital printing inspection methods that mainly rely on visual inspection, which is prone to errors during the inspection process. At the same time, subtle differences cannot be effectively judged by the human eye, and manual inspection is slow, resulting in extended digital printing preparation time and affecting processing efficiency.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a method for inspecting a digital printing equipment, comprising the steps of: a) printhead spacing detection; b) drying status detection; c) printing effect detection; d) gap effect detection; and e) product inspection.

[0008] Step a) Nozzle spacing detection: The position image of the nozzle is captured by an industrial camera. The captured image is transmitted to a computer running a convolutional neural network for dimensionality reduction processing. The convolutional neural network fully learns the data features and also reduces the dimensionality of the feature vector output by the OHE encoding. Finally, the data is mapped to a one-dimensional convolutional neural network for display. The one-dimensional convolutional neural network image data is compared with the pre-stored correct nozzle positioning data. The image recognition system judges the changes that appear after the comparison processing, thereby achieving the effect of rapid detection of nozzle positioning accuracy.

[0009] Step b) Drying status detection: Turn on the infrared lamps and hot air drying module inside the printing equipment, check whether the lamps are fully lit, and test the blowing and exhaust directions of the hot air drying module to check whether the temperature and air volume have reached the set values.

[0010] Step c) Printing effect inspection: The test sample that meets the printing standards is placed into the printing machine that has passed the inspection in steps a) and b) for continuous printing. After printing is completed, the texture, image and text details of the printed matter are checked and the durability of the ink is tested using a friction tester.

[0011] Step d) Gap effect detection: Use an industrial camera to take pictures of the gaps between the splicing positions of the printed material and transmit the images to the computer mentioned in step a). The computer performs image scaling, color space conversion and data augmentation operations on the images to improve the accuracy of the model's color recognition. Then, the trained color recognition model is used to recognize the colors in the images and the color recognition results are displayed to the user in a graphical way.

[0012] Step e) Product inspection: The curing quality is judged by observing the appearance characteristics of the film layer formed by the cured ink. The curing is judged by lightly pressing the cured surface to determine whether the curing is complete. A sharp tool is used to scratch the cured surface and the scratches are observed to judge the curing status.

[0013] As a preferred embodiment of the present invention, step a) nozzle spacing detection: after the nozzle spacing image is acquired and photographed, it needs to be processed by a calculator to make the nozzle placement angle aligned with the horizontal plane of the photo, while eliminating excess image parts.

[0014] As a preferred embodiment of the present invention, step b) drying status detection: the infrared lamp tube and hot air drying module are run empty for thirty minutes, and the presence of water droplets on the inner wall of the machine is observed and the humidity inside the machine is determined using a humidity detector.

[0015] As a preferred embodiment of the present invention, step c) printing effect detection: before printing, the length, width, and thickness of the printed material are measured using precise measuring tools, such as a tape measure or calipers.

[0016] As a preferred embodiment of the present invention, step d) gap effect detection: the color recognition model uses the cross-entropy loss function to measure the difference between the model's prediction results and the real labels and uses the Adam optimization algorithm to update the model's parameters.

[0017] As a preferred embodiment of the present invention, step e) product testing: using acetone or ethanol chemical reagents to drop onto the surface of the film layer, and observing whether the film layer dissolves rapidly to determine the curing quality.

[0018] As a preferred embodiment of the present invention, step a) nozzle spacing detection: the industrial camera is an industrial camera based on a CCD or CMOS chip.

[0019] As a preferred embodiment of the present invention, step d) gap effect detection: the identification and display method is to mark the identified color labels on the image, or to use a color histogram to display the distribution of colors in the image.

[0020] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0021] 1. This invention uses a computer to determine color. Computer color recognition technology is not affected by light, noise, or external environmental factors, thus it can more accurately identify and classify colors. At the same time, it does not require pixel-level image processing, so it can process and identify color information more quickly. In contrast, the human visual system has certain limitations in color recognition, such as a weak ability to distinguish certain colors or difficulty in distinguishing similar colors. This method has the advantages of high accuracy, fast processing speed, wide applicability, and strong customizability in color discrimination.

[0022] 2. This invention uses a computer to calculate the nozzle spacing, which can improve calculation efficiency, avoid calculation errors, and significantly improve user detection efficiency, saving users the steps of manual measurement using tools.

[0023] 3. By inspecting the drying components of the printing equipment, this invention can ensure the accuracy of temperature and airflow, thereby guaranteeing the drying efficiency of printed materials and preventing cracking or undried conditions.

[0024] 4. This invention, by processing the final product, enables the judgment of the final printing effect. Detailed Implementation

[0025] The technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0026] This invention provides a method for inspecting a digital printing machine, comprising the following steps: a) printhead spacing detection; b) drying status detection; c) printing effect detection; d) gap effect detection; and e) product inspection.

[0027] Step a) Nozzle spacing detection: The position image of the nozzle is captured by an industrial camera. The captured image is transmitted to a computer running a convolutional neural network for dimensionality reduction processing. The convolutional neural network fully learns the data features and also reduces the dimensionality of the feature vector output by the OHE encoding. Finally, the data is mapped to a one-dimensional convolutional neural network for display. The one-dimensional convolutional neural network image data is compared with the pre-stored correct nozzle positioning data. The image recognition system judges the changes that appear after the comparison processing, thereby achieving the effect of rapid detection of nozzle positioning accuracy.

[0028] Step b) Drying status detection: Turn on the infrared lamps and hot air drying module inside the printing equipment, check whether the lamps are fully lit, and test the blowing and exhaust directions of the hot air drying module to check whether the temperature and air volume have reached the set values.

[0029] Step c) Printing effect inspection: The test sample that meets the printing standards is placed into the printing machine that has passed the inspection in steps a) and b) for continuous printing. After printing is completed, the texture, image and text details of the printed matter are checked and the durability of the ink is tested using a friction tester.

[0030] Step d) Gap effect detection: Use an industrial camera to take pictures of the gaps between the splicing positions of the printed material and transmit the images to the computer mentioned in step a). The computer performs image scaling, color space conversion and data augmentation operations on the images to improve the accuracy of the model's color recognition. Then, the trained color recognition model is used to recognize the colors in the images and the color recognition results are displayed to the user in a graphical way.

[0031] Step e) Product inspection: The curing quality is judged by observing the appearance characteristics of the film layer formed by the cured ink. The curing is judged by lightly pressing the cured surface to determine whether the curing is complete. A sharp tool is used to scratch the cured surface and the scratches are observed to judge the curing status.

[0032] As a technical optimization of the present invention, step a) nozzle spacing detection: after the nozzle spacing image is acquired and photographed, it needs to be processed by a calculator to make the nozzle placement angle aligned with the horizontal plane of the photo, and at the same time, the excess image parts are eliminated.

[0033] As a technical optimization of the present invention, step b) Drying status detection: the infrared lamp tube and hot air drying module run empty for thirty minutes, observe whether there are water droplets on the inner wall of the machine, and use a humidity detector to determine the internal humidity of the machine.

[0034] As a technical optimization of the present invention, step c) printing effect detection: before printing, the length, width and thickness of the printed matter are measured by a precise measuring tool, such as a tape measure or caliper.

[0035] As a technical optimization scheme of the present invention, step d) gap effect detection: the color recognition model uses the cross-entropy loss function to measure the difference between the model's prediction results and the true labels, and uses the Adam optimization algorithm to update the model's parameters, wherein the algorithm of the cross-entropy loss function is as follows: ;

[0036] (N) is the number of colors.

[0037] (y_i) is the actual label (0 or 1) of the sampled color (i).

[0038] (p_i) is the probability that the color sample (i) is a positive class (usually a class with label 1).

[0039] As a technical optimization of the present invention, step e) product testing: use acetone or ethanol chemical reagent to drop onto the surface of the film layer, and observe whether the film layer dissolves rapidly to determine the curing quality.

[0040] As a technical optimization of the present invention, step a) nozzle spacing detection: the industrial camera is an industrial camera based on CCD or CMOS chip.

[0041] As a technical optimization of the present invention, step d) gap effect detection: the identification and display method is to mark the identified color labels on the image, or to use a color histogram to display the distribution of colors in the image.

[0042] The working principle and usage process of this invention are as follows: An industrial camera, based on a CCD or CMOS chip, is used to capture images of the nozzle positions. After acquisition, the nozzle spacing image needs to be processed by a computer to ensure the nozzle placement angle is aligned with the horizontal plane of the image, while removing excess image data. The captured image is then transmitted to a computer running a convolutional neural network for dimensionality reduction. The convolutional neural network fully learns the data features and also reduces the dimensionality of the feature vector output by the OHE encoding. Finally, the data is mapped onto a one-dimensional convolutional neural network for display. The printhead registration accuracy is quickly assessed by comparing the data with pre-stored printhead registration data and using an image recognition system to identify any changes after comparison. The internal infrared lamps and hot air drying module of the printing equipment are then activated to check if all lamps are lit. The blowing and exhaust directions of the hot air drying module are tested to ensure the temperature and airflow meet the set values. The infrared lamps and hot air drying module are run idle for 30 minutes, and the presence of water droplets on the machine's inner walls is observed. A humidity detector is used to determine the internal humidity. Before printing, the length and width of the printed material are measured using precise measuring tools such as a tape measure or calipers. For thickness and dimension parameters, test samples meeting printing standards are continuously printed into the printing machine that passed the tests in steps a) and b). After printing, the texture, image, and text details of the printed matter are inspected, and the durability of the ink is tested using a friction tester. An industrial camera is used to photograph the gaps between the seams of the printed material, and the images are transmitted to the computer mentioned in step a). The computer performs image scaling, color space conversion, and data augmentation operations on the images to improve the accuracy of the model's color recognition. Then, the trained color recognition model is used to identify the colors in the images, and the color recognition results are displayed graphically. The color recognition model uses a cross-entropy loss function to measure the difference between the model's predictions and the actual labels, and uses the Adam optimization algorithm to update the model's parameters. The recognition is displayed by marking the identified color labels on the image, or by using a color histogram to show the distribution of colors in the image. The curing quality is judged by observing the appearance of the film formed by the cured ink. The curing is judged by lightly pressing the cured surface to determine whether the curing is complete. The cured surface is scratched with a sharp tool, and the scratches are observed to determine the curing status. Acetone or ethanol is dropped onto the film surface, and the film is observed to dissolve quickly to determine the curing quality.

[0043] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus.

[0044] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their likenesses.

Claims

1. A method for inspecting the entire digital printing equipment, characterized in that: This includes steps a) printhead spacing inspection; b) drying status inspection; c) printing effect inspection; d) gap effect inspection; and e) product inspection. Step a) Nozzle spacing detection: The position image of the nozzle is captured by an industrial camera. The captured image is transmitted to a computer running a convolutional neural network for dimensionality reduction processing. The convolutional neural network fully learns the data features and also reduces the dimensionality of the feature vector output by the OHE encoding. Finally, the data is mapped to a one-dimensional convolutional neural network for display. The one-dimensional convolutional neural network image data is compared with the pre-stored correct nozzle positioning data. The image recognition system judges the changes that appear after the comparison processing, thereby achieving the effect of rapid detection of nozzle positioning accuracy. Step b) Drying status detection: Turn on the infrared lamps and hot air drying module inside the printing equipment, check whether the lamps are fully lit, and test the blowing and exhaust directions of the hot air drying module to check whether the temperature and air volume have reached the set values. Step c) Printing effect inspection: The test sample that meets the printing standards is placed into the printing machine that has passed the inspection in steps a) and b) for continuous printing. After printing is completed, the texture, image and text details of the printed matter are checked and the durability of the ink is tested using a friction tester. Step d) Gap effect detection: Use an industrial camera to take pictures of the gaps between the splicing positions of the printed material and transmit the images to the computer mentioned in step a). The computer performs image scaling, color space conversion and data augmentation operations on the images to improve the accuracy of the model's color recognition. Then, the trained color recognition model is used to recognize the colors in the images and the color recognition results are displayed to the user in a graphical way. Step e) Product inspection: The curing quality is judged by observing the appearance characteristics of the film layer formed by the cured ink. The curing is judged by lightly pressing the cured surface to determine whether the curing is complete. A sharp tool is used to scratch the cured surface and the scratches are observed to judge the curing status.

2. The whole machine inspection method for a digital printing equipment according to claim 1, characterized in that: Step a) Nozzle spacing detection: After the nozzle spacing image is acquired and captured, it needs to be processed by a calculator to make the nozzle placement angle aligned with the horizontal plane of the photo, and at the same time, remove excess image parts.

3. The whole machine inspection method for a digital printing equipment according to claim 1, characterized in that: Step b) Drying status detection: Run the infrared lamp and hot air drying module empty for thirty minutes, observe whether there are water droplets on the inner wall of the machine, and use a humidity detector to determine the internal humidity of the machine.

4. The whole machine inspection method for a digital printing equipment according to claim 1, characterized in that: Step c) Printing effect inspection: Before printing, measure the length, width, and thickness of the printed material using precise measuring tools such as tape measures or calipers.

5. The whole machine inspection method for a digital printing equipment according to claim 1, characterized in that: Step d) Gap effect detection: The color recognition model uses the cross-entropy loss function to measure the difference between the model's prediction results and the real labels, and uses the Adam optimization algorithm to update the model's parameters.

6. The whole machine inspection method for a digital printing equipment according to claim 1, characterized in that: Step e) Product testing: Acetone or ethanol chemical reagents are dropped onto the surface of the film layer, and the film layer is observed to determine whether it dissolves rapidly in order to judge the curing quality.

7. The whole machine inspection method for a digital printing equipment according to claim 1, characterized in that: Step a) Nozzle spacing detection: The industrial camera is an industrial camera based on CCD or CMOS chips.

8. The whole machine inspection method for a digital printing equipment according to claim 1, characterized in that: Step d) Gap effect detection: The identification and display method is to mark the identified color labels on the image, or to use a color histogram to display the distribution of colors in the image.