Nozzle state real-time monitoring and closed-loop control system based on image processing

Abstract

The invention relates to a nozzle state real-time monitoring and closed-loop control system based on image processing, and the system comprises an original printing device which comprises a rack and a printing trolley. The control device is in communication connection with the printing trolley, the camera and the light supplementing lamp and used for controlling the printing trolley, the camera and the light supplementing lamp to operate, obtaining the detection pattern and conducting image processing on the detection pattern to obtain an image recognition result. And when the image recognition result is normal, the printing trolley is controlled to complete a printing task on the printing medium, and when the image recognition result is abnormal, the printing trolley is controlled again to print a detection pattern until the image recognition result of the detection pattern is normal after a spray head cleaning task is executed. The nozzle state real-time monitoring and closed-loop control system based on image processing provided by the invention remarkably reduces the material waste rate, improves the effective operation time of a printer, reduces the manual maintenance frequency, and further reduces the maintenance cost.

Description

Technical Field

[0001] This invention relates to the field of image processing and printing technology, and in particular to a real-time monitoring and closed-loop control system for printhead status based on image processing during image processing on fabric surfaces. Background Technology

[0002] An inkjet printer is a printing device that sprays liquid ink into tiny droplets through nozzles, which then adhere to a substrate. In the field of fabric surface image processing, common malfunctions of inkjet printers include minor clogging, complete clogging, printhead misalignment, and uneven ink distribution. These conditions differ from those encountered in paper surface pattern processing. Due to the nature of inkjet printing on flexible materials such as clothing fabrics, home textiles, and industrial textiles, issues like ink droplet diffusion caused by fabric texture, printhead misalignment caused by tension changes, and uneven ink absorption due to differences in fiber structure arise. These issues differ from the common problems encountered in paper printing for office or packaging applications, and are specific to smooth surfaces. Printhead clogging and uneven ink distribution are more frequent in fabric printers due to the porous nature of fabric fibers and poor surface smoothness. The detection accuracy of existing general printers is insufficient to meet the requirements of this field, resulting in low reliability.

[0003] Therefore, current market monitoring of these types of malfunctions remains at the manual monitoring stage, relying on on-duty personnel to monitor and address problems as they arise. However, due to inherent individual differences and limited attention spans among manual personnel, manual monitoring and handling often suffer from delays, low efficiency, and poor accuracy, leading to high material waste, short effective printer uptime, and overall high maintenance costs. Therefore, the market needs a machine vision-based automated monitoring and handling system that can not only improve the efficiency of inkjet printer malfunction handling but also reduce maintenance costs. Summary of the Invention

[0004] The purpose of this invention is to at least address one of the shortcomings of the prior art and provide a real-time monitoring and closed-loop control system for nozzle status based on image processing.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: Specifically, a real-time monitoring and closed-loop control system for nozzle status based on image processing is proposed, including the following: The original printing unit includes a frame and a printing carriage; The frame is provided with a printing media placement area and a detection area. The printing media placement area is used to place the printing media. The printing carriage is used to print a preset detection pattern in the detection area through the printhead, and complete the printing task on the printing media when the image recognition result of the detection pattern is normal. A camera, mounted on the frame, is used to capture the detection pattern in the detection area; A supplementary light, mounted on the frame, is used to provide a light source for the camera, enabling the camera to clearly capture the detection pattern in the detection area. The control device is communicatively connected to the printing carriage, camera, and supplementary light. It controls the operation of the printing carriage, camera, and supplementary light, acquires the detection pattern, and performs image processing on the detection pattern to obtain an image recognition result. When the image recognition result is normal, it controls the printing carriage to complete the printing task on the printing medium. When the image recognition result is abnormal, it performs a printhead cleaning task and then re-controls the printing carriage to print the detection pattern until the image recognition result of the detection pattern is normal.

[0006] Furthermore, specifically, the preset detection pattern is a continuous long strip of colored bars, which is composed of continuous ink dots.

[0007] Furthermore, specifically, the process of performing image processing on the detected pattern to obtain the image recognition result includes, The detected pattern is preprocessed to obtain a first image with geometric transformation, noise reduction, and contrast enhancement; The first image is binarized using a local adaptive thresholding method to obtain the second image; The second image is smoothed at the edges using morphological operations to obtain the third image; Edge detection processing is performed on the third image to obtain the edge to be monitored; The edge to be monitored is compared with the template lines. When the diameter deviation of an ink dot exceeds the first threshold range of the template lines, the ink dot is judged to be missing, and the number of missing ink dots is counted. If the number of missing ink dots reaches the first threshold and the number of consecutive missing ink dots is lower than the second threshold, it is judged as an ink dot defect abnormality. If the number of consecutive missing ink dots is not lower than the second threshold, it is judged as an ink dot defect abnormality. When there is an ink dot defect abnormality or an ink dot defect abnormality, the image recognition result is judged as abnormal; otherwise, it is normal.

[0008] Furthermore, specifically, the first image is binarized using a local adaptive thresholding method to obtain the second image, including: After the first image is converted to grayscale, it is evenly divided into M blocks at preset intervals, that is, by default, each block contains the same number of ink dots. Calculate the average gray value of each patch, set the gray value of pixels with a gray value lower than the average gray value to 0, and set the gray value of other pixels with a gray value not lower than the average gray value to 255. The second image is obtained after all tiles have been processed.

[0009] Furthermore, the process of performing image processing on the detected pattern to obtain the image recognition result also includes, Let P_i be the number of pixels with a gray value of 255 in the i-th patch, where i∈[1,M]. Calculate the average number P_avg of pixels with a grayscale value of 255 in the detected pattern. Calculate the overall deviation value Deviat_i for the i-th patch. The number of patches with an overall deviation value greater than the third threshold, Q, is counted. If the value of Q is greater than the fourth threshold, it indicates that there is an abnormality in the amount of ink. When there is an abnormality in the amount of ink, the image recognition result is also considered abnormal.

[0010] Furthermore, the process of performing image processing on the detected pattern to obtain the image recognition result also includes, If the number of consecutive missing ink dots is greater than α times the second threshold, it is judged as a serious abnormality of missing ink dots. At this time, the control device will stop printing and issue an alarm to notify the staff to handle the situation. α is an adjustment coefficient, which is greater than 1.

[0011] Furthermore, if the image recognition result of the detection pattern is still abnormal after the fifth consecutive spray head cleaning task is performed, the control device will issue an alarm to notify the staff to handle the situation.

[0012] Furthermore, specifically, the fill light is constructed using a coaxial light source.

[0013] Furthermore, specifically, the camera is positioned at the top of the frame, ensuring that its captured content precisely covers the detection area, and its shooting angle is set perpendicular to the center of the detection area.

[0014] The beneficial effects of this invention are as follows: This invention proposes a real-time printhead status monitoring and closed-loop control system based on image processing. It uses a preset detection pattern and adds a camera and associated supplementary lighting to the original printing device to acquire images of the preset detection area. Based on the degree of error between the preset detection pattern and the template image, it automatically determines the type of inkjet printer malfunction and performs closed-loop control based on the judgment result. This greatly ensures the normal operation of the inkjet printer, only notifying personnel to handle serious problems or when automatic control fails to resolve the issue multiple times. This significantly reduces material waste, increases effective printer uptime, and reduces the frequency of manual maintenance, thereby lowering maintenance costs. This invention is specifically designed for printhead status monitoring in fabric printing scenarios, adapting to changes in ink droplet morphology caused by fabric fiber structure and deformation of the detection pattern caused by tension, unlike the stable planar detection environment in paper printing. Attached Figure Description

[0015] The above and other features of this disclosure will become more apparent from the detailed description of the embodiments illustrated in conjunction with the accompanying drawings. In the accompanying drawings, the same reference numerals denote the same or similar elements. Obviously, the drawings described below are merely some embodiments of this disclosure. For those skilled in the art, other drawings can be obtained from these drawings without any creative effort. In the drawings: Figure 1 The diagram shown is a schematic diagram of the real-time monitoring and closed-loop control system for nozzle status based on image processing according to the present invention. Figure 2 The diagram shown is an example of an anomaly involved in this invention. Detailed Implementation

[0016] The following will provide a clear and complete description of the concept, specific structure, and technical effects of the present invention in conjunction with embodiments and accompanying drawings, so as to fully understand the purpose, solution, and effects of the present invention. It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. The same reference numerals used throughout the accompanying drawings indicate the same or similar parts.

[0017] Example 1, referring to Figure 1 This invention proposes a real-time monitoring and closed-loop control system for nozzle status based on image processing, comprising the following: The original printing unit includes a frame and a printing carriage; The frame is provided with a printing media placement area and a detection area. The printing media placement area is used to place the printing media. The printing carriage is used to print a preset detection pattern in the detection area through the printhead, and complete the printing task on the printing media when the image recognition result of the detection pattern is normal. A camera, mounted on the frame, is used to capture the detection pattern in the detection area; A supplementary light, mounted on the frame, is used to provide a light source for the camera, enabling the camera to clearly capture the detection pattern in the detection area. The control device is communicatively connected to the printing carriage, camera, and supplementary light. It controls the operation of the printing carriage, camera, and supplementary light, acquires the detection pattern, and performs image processing on the detection pattern to obtain an image recognition result. When the image recognition result is normal, it controls the printing carriage to complete the printing task on the printing medium. When the image recognition result is abnormal, it performs a printhead cleaning task and then re-controls the printing carriage to print the detection pattern until the image recognition result of the detection pattern is normal.

[0018] In this embodiment 1, a preset detection pattern is used, and a camera and its associated supplementary light are added to the original printing device to collect images of the preset detection area. The inkjet printer fault type is automatically judged based on the degree of error between the preset detection pattern and the template image, and closed-loop control is automatically performed based on the judgment result. This greatly ensures the normal operation of the inkjet printer. Only when there is a serious problem or the automatic control fails to solve the problem multiple times will the staff be notified for handling. This significantly reduces the material waste rate, increases the effective running time of the printer, reduces the frequency of manual maintenance, and thus reduces maintenance costs.

[0019] As a preferred option, a 20-megapixel area-scan camera can be used, with a resolution of 5472×3648, a pixel size of 2.4μm×2.4μm, a dynamic range of 65.5dB, and a signal-to-noise ratio of 41.5dB. At full resolution of 5472×3648, the maximum frame rate is 5.9fps. The high resolution of 20 megapixels (5472×3648) can clearly capture the minute details printed by the inkjet printer, making it highly effective for detecting anomalies that may occur during the printing process, such as broken lines caused by nozzle clogging, uneven ink droplets, and blurred patterns. While the maximum frame rate of 5.9fps at full resolution is not particularly high, it is sufficient for real-time monitoring of the inkjet printer printing process. It can capture images of each frame in real time during the printing process to analyze the specific circumstances of printing anomalies. The exposure time ranges from 46μs to 2.5sec, the gain range is from 0dB to 24dB, and it supports automatic or manual adjustment. This allows the camera to flexibly adjust exposure and gain under different lighting conditions, based on changes in printing materials and printing environment, to obtain optimal image quality and thus accurately detect printing abnormalities.

[0020] The camera uses a GigE interface, which can directly transmit image data to devices such as computers via network cable. The computer needs to install a corresponding image acquisition card to receive and process this data.

[0021] The type, power, and illumination angle of the light source are selected in the following ways to ensure image quality: Light source type: For specular / transparent media with surface defects, choose a coaxial light source. Power selection: Refer to the camera's illumination requirements (≥500 lux). For matte media, select 15-30W; for glossy / transparent media, select 5-15W. Match the exposure time, prioritizing adjustable power options to avoid overexposure / underexposure. Angle selection: Strong defect contrast, interference reduction. High-gloss medium + minor defects use 0°-15° (coaxial / low angle); matte medium + edge defects use 30°-60° (medium angle); transparent medium use 90° (backlight transmission); high-gloss anti-reflective use 15°-30° low angle + ring light source.

[0022] The nozzle cleaning process is as follows: Negative pressure ink extraction and solvent immersion are employed.

[0023] Negative pressure ink extraction: Pressure: -5kPa ~ -15kPa (negative pressure value; the smaller the absolute value, the weaker the suction).

[0024] Minor clogging: -5~-8 kPa (avoid excessive suction to prevent damage to the nozzle); Medium clogging: -10~-15kPa (requires short-term ink extraction).

[0025] Duration: 1-3 seconds per cycle, no more than 3 cycles in a row (10-second interval) to prevent negative pressure vacuum from forming inside the nozzle cavity.

[0026] Solvent / Ink: Fresh ink is usually drawn directly (it has a self-cleaning effect). In special cases, a special cleaning solution can be used (such as deionized water for water-based inks). Draw ink and clear blockages at the pressure above.

[0027] Solvent type: Water-based inks: deionized water, special water-based cleaning solution (containing surfactants); Solvent-based inks: Special organic solvents (such as cyclohexanone diluent, which must be matched with the printhead material); UV ink: Special UV cleaning solution (can dissolve and cure the surface layer of the ink).

[0028] Soaking time: 30 minutes to 2 hours for slight drying; 4 to 8 hours for severe clumping (avoid exceeding 24 hours to prevent solvent penetration and damage to the internal structure).

[0029] In a preferred embodiment of the present invention, the preset detection pattern is a continuous elongated colored strip, which is composed of continuous ink dots.

[0030] The overall workflow is as follows: When the printing carriage moves the printhead to print a pattern on the printing medium, a printhead detection pattern is simultaneously printed on the edge of the rightmost detection area. A supplementary light is installed next to the camera, enabling the camera to clearly capture the printhead detection pattern in the detection area and perform real-time detection. If an anomaly is detected, the printer pauses and performs a cleaning operation. If the anomaly is resolved after cleaning, printing resumes.

[0031] When an abnormal printhead condition is detected, exceeding the set range, the printer pauses, performs a cleaning operation, and resumes printing only after the printhead condition has been cleaned and restored.

[0032] Table 1 (Response Time of the Invention in Application), Table 2 (Accuracy of the Invention in Identifying Different Fault Types in Application), and Table 3 (Comparison Before and After Applying the Invention) below show the performance data of the system proposed by the present invention during testing: Table 1 Table 2 Table 3 In addition, the following steps were performed: Three inkjet printers of the same model were selected from a packaging and printing factory, and the following methods were used respectively: Control group A: Traditional manual inspection (inspection every 2 hours) Control group B: Existing detection system based on a line scan camera (12 megapixels resolution) Experimental Group C: The system of this invention (MV-CS200-10GC + adaptive algorithm) The same food packaging printing task was performed (8 hours per day for 10 consecutive days), and the results are shown in Table 4 below: Table 4 Data Description All tests were conducted in a real industrial environment (temperature 25±5℃, humidity 50±10% RH), with food-grade composite film as the printing medium and solvent-based black ink as the ink.

[0033] Accuracy calculation uses the results of manual review as the "true value". Missed judgment refers to real anomalies that the system failed to identify, and false judgment refers to non-abnormal events that the system falsely reported.

[0034] Material waste rate = (weight of printing material scrapped due to defects) / (total weight of material consumed) × 100%.

[0035] The quantitative data above clearly demonstrates the significant advantages of this system in terms of response speed, recognition accuracy, and economic benefits, far exceeding manual detection and existing technological levels.

[0036] In this preferred embodiment, by setting the detection group as a long strip of colored bar, it is easier to find anomalies during the subsequent image recognition process.

[0037] In a preferred embodiment of the present invention, specifically, the process of performing image processing on the detection pattern to obtain an image recognition result includes: The detected pattern is preprocessed to obtain a first image with geometric transformation, noise reduction, and contrast enhancement; The first image is binarized using a local adaptive thresholding method to obtain the second image; The second image is smoothed at the edges using morphological operations to obtain the third image; Edge detection processing is performed on the third image to obtain the edge to be monitored; The edge to be monitored is compared with the template lines. When the diameter deviation of an ink dot exceeds the first threshold range of the template lines, the ink dot is judged to be missing, and the number of missing ink dots is counted. If the number of missing ink dots reaches the first threshold and the number of consecutive missing ink dots is lower than the second threshold, it is judged as an ink dot defect abnormality. If the number of consecutive missing ink dots is not lower than the second threshold, it is judged as an ink dot defect abnormality. When there is an ink dot defect abnormality or an ink dot defect abnormality, the image recognition result is judged as abnormal; otherwise, it is normal.

[0038] Reference Figure 2 In this preferred embodiment, by comparing the acquired detection pattern with the template pattern, if the diameter deviation of the ink dot exceeds the first threshold range of the template lines, it is determined that the ink dot is missing. If the number of missing dots is small and discontinuous, we consider this to be a fault type of ink dot incompleteness. If the number is indeed continuous and reaches a certain level, we consider this to be a fault type of ink dot missingness. The above method can better detect the above fault types. Considering ink dot displacement (in image processing, it will be directly judged as ink dot incompleteness and ink dot missingness, because after displacement, the pixel corresponding to the position will be directly recorded as ink dot missingness through image processing, and edge detection will automatically filter out this type of edge, so this situation is not considered).

[0039] In a preferred embodiment of the present invention, specifically, the first image is binarized using a local adaptive thresholding method to obtain the second image, including: After the first image is converted to grayscale, it is evenly divided into M blocks at preset intervals, that is, by default, each block contains the same number of ink dots. Calculate the average gray value of each patch, set the gray value of pixels with a gray value lower than the average gray value to 0, and set the gray value of other pixels with a gray value not lower than the average gray value to 255. The second image is obtained after all tiles have been processed.

[0040] In a preferred embodiment of the present invention, the process of performing image processing on the detection pattern to obtain the image recognition result further includes... Let P_i be the number of pixels with a gray value of 255 in the i-th patch, where i∈[1,M]. Calculate the average number P_avg of pixels with a grayscale value of 255 in the detected pattern. Calculate the overall deviation value Deviat_i for the i-th patch. The number of patches with an overall deviation value greater than the third threshold, Q, is counted. If the value of Q is greater than the fourth threshold, it indicates that there is an abnormality in the amount of ink. When there is an abnormality in the amount of ink, the image recognition result is also considered abnormal.

[0041] In this preferred embodiment, considering that the diameter and area of ​​the ink dots in the overall detection pattern should theoretically be consistent, if the ink volume is insufficient, the ink dots will have low gray values ​​or even be the background color, which will be set to 0 by the adaptive binarization process. Therefore, the number of pixels that can be counted will be small. If there are too many such color blocks, it indicates that there is an uneven ink volume. Therefore, the above method can accurately identify this type of anomaly.

[0042] In a preferred embodiment of the present invention, the process of performing image processing on the detection pattern to obtain the image recognition result further includes... If the number of consecutive missing ink dots is greater than α times the second threshold, it is judged as a serious abnormality of missing ink dots. At this time, the control device will stop printing and issue an alarm to notify the staff to handle the situation. α is an adjustment coefficient, which is greater than 1.

[0043] In this preferred embodiment, considering that too many missing ink dots would be a very serious situation, the machine is stopped immediately and staff are notified to take targeted measures.

[0044] In a preferred embodiment of the present invention, if the image recognition result of the detected pattern is still abnormal after the nozzle cleaning task has been performed continuously for a fifth threshold number of times, the control device will issue an alarm to notify the staff to handle the situation. Preferably, the fifth threshold value can be set to 3, which yields better results.

[0045] In a preferred embodiment of the present invention, the supplementary light is constructed using a coaxial light source.

[0046] In this preferred embodiment, a coaxial light source is used because a coaxial light source is suitable for detecting the flatness, gloss, and presence of scratches or other abnormalities on the surface of printed materials. It can provide uniform illumination and reduce the effects of reflections and shadows.

[0047] In a preferred embodiment of the present invention, the camera is specifically positioned at the upper end of the frame, ensuring that its captured content precisely covers the detection area, and its shooting angle is set perpendicular to the center of the detection area.

[0048] In this preferred embodiment, the camera is positioned downstream of the printhead relative to the paper's direction of travel. This allows for timely inspection of the printed content after printing, ensuring that the complete image printed by the printhead is captured. The camera lens is perpendicular to the printing plane to ensure a clear view of the printed content.

[0049] The current camera target size is approximately 14.6mm × 10.2mm. Based on the relationship between object distance, field of view, target size, and focal length: WD = (f × FOV) / target size, the vertical distance from the camera lens to the printed surface is approximately 960mm.

[0050] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0051] If the integrated module is implemented as a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or system capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc.

[0052] Although the description of the invention has been quite detailed and particularly of several described embodiments, it is not intended to limit it to any of these details or embodiments or any particular embodiment, but should be considered as providing a broad possible interpretation of the claims by referring to the appended claims and taking into account the prior art, thereby effectively covering the intended scope of the invention. Furthermore, the invention has been described above with respect to embodiments foreseeable by the inventors in order to provide a useful description, and non-substantial modifications to the invention that have not yet been foreseen may still represent equivalent modifications.

[0053] The above description is merely a preferred embodiment of the present invention. The present invention is not limited to the above-described embodiments. Any embodiment that achieves the technical effects of the present invention using the same means should fall within the protection scope of the present invention. Within the protection scope of the present invention, various modifications and variations can be made to the technical solutions and / or implementation methods.

Claims

1. A real-time monitoring and closed-loop control system for nozzle status based on image processing, characterized in that, Including the following: The original printing unit includes a frame and a printing carriage; The frame is provided with a printing media placement area and a detection area. The printing media placement area is used to place the printing media. The printing carriage is used to print a preset detection pattern in the detection area through the printhead, and complete the printing task on the printing media when the image recognition result of the detection pattern is normal. A camera, mounted on the frame, is used to capture the detection pattern in the detection area; A supplementary light, mounted on the frame, is used to provide a light source for the camera, enabling the camera to clearly capture the detection pattern in the detection area. The control device is communicatively connected to the printing carriage, camera, and supplementary light. It controls the operation of the printing carriage, camera, and supplementary light, acquires the detection pattern, and performs image processing on the detection pattern to obtain an image recognition result. When the image recognition result is normal, it controls the printing carriage to complete the printing task on the printing medium. When the image recognition result is abnormal, it performs a printhead cleaning task and then re-controls the printing carriage to print the detection pattern until the image recognition result of the detection pattern is normal.

2. The real-time monitoring and closed-loop control system for nozzle status based on image processing according to claim 1, characterized in that, Specifically, the preset detection pattern is a continuous long strip of colored bars, which is composed of continuous ink dots.

3. The real-time monitoring and closed-loop control system for nozzle status based on image processing according to claim 2, characterized in that, Specifically, the process of performing image processing on the detected pattern to obtain the image recognition result includes, The detected pattern is preprocessed to obtain a first image with geometric transformation, noise reduction, and contrast enhancement; The first image is binarized using a local adaptive thresholding method to obtain the second image; The second image is smoothed at the edges using morphological operations to obtain the third image; Edge detection processing is performed on the third image to obtain the edge to be monitored; The edge to be monitored is compared with the template lines. When the diameter deviation of an ink dot exceeds the first threshold range of the template lines, the ink dot is judged to be missing, and the number of missing ink dots is counted. If the number of missing ink dots reaches the first threshold and the number of consecutive missing ink dots is lower than the second threshold, it is judged as an ink dot defect abnormality. If the number of consecutive missing ink dots is not lower than the second threshold, it is judged as an ink dot defect abnormality. When there is an ink dot defect abnormality or an ink dot defect abnormality, the image recognition result is judged as abnormal; otherwise, it is normal.

4. The real-time monitoring and closed-loop control system for nozzle status based on image processing according to claim 3, characterized in that, Specifically, the first image is binarized using a local adaptive thresholding method to obtain the second image, including: After the first image is converted to grayscale, it is evenly divided into M blocks at preset intervals, that is, by default, each block contains the same number of ink dots. Calculate the average gray value of each patch, set the gray value of pixels with a gray value lower than the average gray value to 0, and set the gray value of other pixels with a gray value not lower than the average gray value to 255. The second image is obtained after all tiles have been processed.

5. The real-time monitoring and closed-loop control system for nozzle status based on image processing according to claim 4, characterized in that, The process of performing image processing on the detected pattern to obtain the image recognition result also includes, Let P_i be the number of pixels with a gray value of 255 in the i-th patch, where i∈[1,M]. Calculate the average number P_avg of pixels with a grayscale value of 255 in the detected pattern. Calculate the overall deviation value Deviat_i for the i-th patch. The number of patches with an overall deviation value greater than the third threshold, Q, is counted. If the value of Q is greater than the fourth threshold, it indicates that there is an abnormality in the amount of ink. When there is an abnormality in the amount of ink, the image recognition result is also considered abnormal.

6. The real-time monitoring and closed-loop control system for nozzle status based on image processing according to claim 4, characterized in that, The process of performing image processing on the detected pattern to obtain the image recognition result also includes, If the number of consecutive missing ink dots is greater than α times the second threshold, it is judged as a serious abnormality of missing ink dots. At this time, the control device will stop printing and issue an alarm to inform the staff to handle the situation. α is an adjustment coefficient, which is greater than 1.

7. The real-time monitoring and closed-loop control system for nozzle status based on image processing according to claim 1, characterized in that, If the image recognition result of the detection pattern is still abnormal after the fifth consecutive spray head cleaning task is performed, the control device will issue an alarm and notify the staff to handle the situation.

8. The real-time monitoring and closed-loop control system for nozzle status based on image processing according to claim 1, characterized in that, Specifically, the fill light is constructed using a coaxial light source.

9. The real-time monitoring and closed-loop control system for nozzle status based on image processing according to claim 1, characterized in that, Specifically, the camera is positioned at the top of the frame, ensuring that its captured content precisely covers the detection area, and its shooting angle is set perpendicular to the center of the detection area.

Images