A method and system for detecting ink droplets ejected from a printhead array

Abstract

This invention belongs to the field of novel displays, specifically relating to a method and system for detecting ink droplets ejected from a printhead array. The method includes: controlling multiple vision systems to simultaneously acquire images of ink droplets ejected from the nozzles of multiple printheads; calculating the flight state of the ink droplets ejected from each nozzle based on the image; adjusting the position of the motion frame and / or printhead array as a whole, combining the coordinates of the currently detected nozzles and the next nozzles to be detected by each vision system, and fine-tuning the position of vision systems that have not reached the image acquisition position of the nozzles to be detected after the overall adjustment; and repeating the image acquisition process. The coordinates of each nozzle are obtained by multiplying its position in the corresponding printhead Cartesian coordinate system by the corresponding printhead offset angle and adding the coordinates of the origin of the corresponding printhead Cartesian coordinate system in the global coordinate system. This invention can simultaneously scan and detect multiple printheads, effectively improving detection efficiency.

Description

Technical Field

[0001] This invention belongs to the field of novel displays, and more specifically, relates to a method and system for detecting ink droplets ejected from a printhead array. Background Technology

[0002] New display technologies are one of the key development directions of my country's strategic emerging industries in the information and intelligent era. They are developing towards ultra-high resolution, large size, thinness, flexibility, and low cost. Traditional technologies such as vacuum evaporation are facing problems such as high energy consumption, serious material waste, and complex processes. Inkjet printing technology, on the other hand, is an emerging display manufacturing technology that uses functional raw materials to form ink and then uses inkjet printing technology to fabricate organic or inorganic electronic devices on rigid or flexible substrates. It has advantages such as adaptability to large-area preparation, high material adaptability and utilization, low cost, and green manufacturing, and is one of the important development directions of the future new display industry.

[0003] During inkjet printing, abnormal conditions such as nozzle conditions in the printhead array, ink properties, and printing atmosphere can occur due to the influence of process parameters. These abnormalities include nozzle clogging, droplet flight deviation, satellite droplets, trailing, and abnormal droplet size and velocity. These abnormalities can lead to defects such as missed prints, confluence, scattered dots, and uneven ink distribution upon landing on the substrate, ultimately resulting in display screen defects. Large-size substrate inkjet printing typically requires multiple printheads in the printhead array to operate simultaneously, resulting in a large number of nozzles to be inspected. Therefore, it is essential to efficiently detect parameters such as the volume, velocity, and angle of the ejected ink droplets to ensure printing quality.

[0004] Existing inkjet printing droplet measurement methods can be broadly categorized into offline and online methods. Offline measurement primarily uses the gravimetric method, while online measurement includes inductive methods, laser interferometry, and visual methods, all of which have been applied in experimental research and industrial production. The gravimetric method works by using a precision measuring device to collect a large number of ink droplets at a specific frequency over a certain period, measuring the total mass of the droplets, calculating the average mass of each droplet, and then calculating the droplet volume based on the density. This method offers high measurement accuracy but is only applicable to deposited droplets and is susceptible to ink solvent evaporation. It cannot detect droplet velocity and angle, making it suitable as a comparative verification method for other measurement approaches. The inductive method primarily measures the self-induced voltage during nozzle ejection to determine if the ejection is abnormal, but this method is only qualitative and not suitable for precise droplet measurement. The laser interferometry method utilizes the phase Doppler principle, using two or more laser beams to interfere in the droplet measurement area. A photodetector is placed at a certain angle to the emitted light, and the droplet volume, velocity, and angle are calculated through interference signal analysis. However, its high cost limits its application to some extent.

[0005] Furthermore, some literature discloses visual methods for detecting flying ink droplets using stroboscopic imaging; however, these methods typically only allow for the detection of a single nozzle at a time, resulting in low detection efficiency. Considering the simultaneous use of multiple nozzles in a printhead array and the large number of nozzles to be detected, designing a new detection system and scheme to achieve efficient and accurate online measurement of printhead arrays constitutes a key technological need that urgently needs to be addressed in this field. Summary of the Invention

[0006] To address the shortcomings and improvement needs of existing technologies, this invention provides a method and system for detecting ink droplets ejected from a printhead array. The purpose is to improve the efficiency of detecting ink droplets ejected from multiple printheads in a printhead array and to achieve efficient and accurate online measurement of the printhead array.

[0007] To achieve the above objectives, according to one aspect of the present invention, a method for detecting ink droplets ejected from a printhead array is provided, comprising:

[0008] Control multiple vision systems to simultaneously acquire images of ink droplets ejected from the nozzles of multiple printheads; calculate the flight state of the ink droplets ejected from each nozzle based on the image of the ink droplets ejected from that nozzle.

[0009] By combining the coordinates of the nozzles currently detected by each vision system with the coordinates of the nozzles to be detected next, the position of the moving frame and / or printhead array is adjusted as a whole, and the position of the vision system that has not reached the image acquisition position of the ink droplets of the nozzles to be detected after the overall adjustment is fine-tuned; image acquisition is repeated until the ink droplet detection of all nozzles to be detected in the printhead array is completed.

[0010] The multiple vision systems are mounted on the motion frame. The coordinates of each nozzle are determined as follows: the angle between a certain axis vector of each nozzle's Cartesian coordinate system and the corresponding axis vector of the global coordinate system of the motion frame is obtained as the nozzle offset angle of that nozzle. The coordinate position of each nozzle in the corresponding nozzle's Cartesian coordinate system is multiplied by the corresponding nozzle offset angle, and the coordinates of the origin of the corresponding nozzle's Cartesian coordinate system in the global coordinate system are added to obtain the coordinates of each nozzle in the global coordinate system. The two coordinate axes of the global coordinate system correspond to the same two mutually perpendicular degrees of freedom directions of the motion frame and / or the nozzle array.

[0011] The beneficial effects of this invention are as follows: This invention proposes a detection method for synchronous scanning of ink droplets ejected from a printhead array. By using parameters such as the printhead array splicing offset angle (theoretically, there should be no included angle between adjacent printheads on a plane, but in reality, there will be splicing misalignment on the plane, with a certain included angle between adjacent printheads; the printhead offset angle indicates the degree of misalignment) and the coordinates of the nozzle position distribution, the system calculates the motion step length of each degree of freedom during the scanning and detection of all nozzles in the printhead array. This allows multiple vision inspection systems to be aligned with the ink droplets ejected from their respective nozzles, synchronously performing the detection of multiple printhead nozzles in the printhead array. Compared to conventional flying droplet detection schemes, this overcomes the limitation of only being able to detect a single nozzle at a time, effectively improving observation efficiency. In addition, the synchronous scanning and detection method for ink droplets ejected from a printhead array designed in this invention does not limit the number of printheads to be detected simultaneously. Only the addition of corresponding vision systems and motion mechanisms is needed to increase the number of printheads to be detected simultaneously, enabling the detection needs of larger-scale printhead arrays and achieving high-efficiency detection.

[0012] Furthermore, the method for fine-tuning the position of the overall adjustment motion frame and / or printhead array, and for fine-tuning the position of the vision system that has not reached the image acquisition position of the ejected ink droplets at the nozzle to be detected after the overall adjustment, is as follows:

[0013] The spacing between two adjacent vision systems is set to an integer multiple of the spacing between adjacent nozzles in the nozzle array to be inspected; and a position fine-tuning mechanism is configured for each vision system.

[0014] A vision system is selected as the reference vision system. Based on the coordinates of its pre-detection nozzle, the motion frame and / or the printhead array to be detected are controlled to move in the two degrees of freedom, so that the reference vision system is located at the image acquisition position of the ejected ink droplets of its pre-detection nozzle. Based on the coordinates of the pre-detection nozzles of each other vision system, the position of each vision system is adjusted in the two degrees of freedom to the image acquisition position of the ejected ink droplets of its pre-detection nozzle through the position fine-tuning mechanism of each vision system.

[0015] A further beneficial effect of this invention is that by setting the spacing between two adjacent vision systems to an integer multiple of the spacing between adjacent printheads within the printhead array to be inspected, simultaneous detection of multiple printhead nozzles can be achieved during each inspection. Furthermore, each vision system is equipped with a corresponding position fine-tuning mechanism. One vision system is selected as a reference vision system, and this reference vision system is adjusted to the position where it acquires the image of ink droplets ejected from the nozzle to be inspected. Other vision systems achieve position fine-tuning through their respective fine-tuning mechanisms, thus achieving efficient position adjustment of the vision systems for simultaneous detection.

[0016] Furthermore, the motion step size matrix of the two-degree-of-freedom fine-tuning mechanism corresponding to the two degrees of freedom directions of each position fine-tuning mechanism is set as follows:

[0017]

[0018] In the formula, u l，n This indicates that when the position fine-tuning mechanism of the l-th vision system observes the n-th nozzle of the H(l)-th nozzle, it has a position in degree of freedom U. l The direction of motion, v l，n This indicates that when the position fine-tuning mechanism of the l-th vision system observes the n-th nozzle of the H(l)-th nozzle, it has a position in degree of freedom V. l The direction of motion, l∈{1, 2, ..., L}, n∈{1, 2, ..., N}, where L is the total number of vision systems and N is the total number of nozzles in each nozzle; A′ H(l) A represents the nozzle coordinate array matrix A of nozzle number H(l), consisting of the coordinates of all its nozzles in the Cartesian coordinate system for the nozzle. H(l) Rotation β i The matrix after that, A′ H(k) A represents the nozzle coordinate array matrix A, which is the coordinates of all nozzles of nozzle H(k) observed by the k-th vision system as a reference, in the Cartesian coordinate system for nozzles. H(k) Rotation β k The matrix after that.

[0019] Furthermore, each vision system employs stroboscopic exposure and performs double flashes on each ejected ink droplet within an exposure time period to acquire images, resulting in the acquired images containing projected images of the same ink droplet at two different times.

[0020] A further beneficial effect of the present invention is that the printhead array jet ink droplet synchronous scanning detection method designed in this invention adopts a design of two flashes of the light source within a single exposure time of the camera. It can obtain the projected images of the flying ink droplets at two moments before and after in a single frame image, and the time interval between the two flashes is adjustable. This image acquisition method of the vision system can calculate the flight speed and angle of the ink droplets using a single frame image. Compared with the conventional detection scheme that acquires two frames before and after and then compares them to calculate the flight speed and angle of the ink droplets, it significantly reduces the measurement time and saves detection steps.

[0021] Furthermore, each vision system, when acquiring images of the ejected ink droplets from the corresponding nozzle, specifically does the following:

[0022] During an exposure period, multiple ink droplets before and after the ejection are subjected to double flashes to obtain an image of the ejected ink droplets from the nozzle. The projection image on top of the ejected ink droplet image is a superposition of the projection images captured at the previous moment when each ejected ink droplet is subjected to double flashes, and the projection image at the back of the ejected ink droplet image is a superposition of the projection images captured at the later moment when each ejected ink droplet is subjected to double flashes.

[0023] A further beneficial effect of the present invention is that by extending the exposure time of the camera, it can contain multiple sets of double flashes within one exposure time period, enabling the camera to capture superimposed ink droplet images, thereby improving image contrast.

[0024] Furthermore, based on the ejected ink droplet image corresponding to each nozzle, the actual volume of the ejected ink droplet for that nozzle is calculated as follows:

[0025] Edge detection is performed on the upper and lower projected image regions of the ink droplet image corresponding to each nozzle, and the actual height of the ink droplet corresponding to each projected image region after edge detection is calculated based on the camera calibration value; the ink droplet corresponding to each projected image region is discretized into n slices with a thickness of Δh in the height direction;

[0026] The Laplace transform of Gaussian is used to process each projected image region after edge detection, and the gray-level variance σ of the projected image region after the Laplace transform of Gaussian is calculated. 2 As a sharpness evaluation metric, the weighting coefficient λ corresponding to each projected image region is determined based on the following expression:

[0027]

[0028] In the formula, t is the gray-level variance σ of the projected image region. 2 The value after normalization to the interval [-10, 10];

[0029] The actual volume V of the ejected ink droplets from this nozzle is calculated based on the following expression:

[0030]

[0031] Where λ1 and λ2 represent the weighting coefficients of the upper and lower projected image regions, respectively, and d i Let d represent the diameter of the cross-section of the i-th circular slice layer of an ink droplet in one of the projected image regions, i∈{1,2,…,n}. j Let represent the diameter of the cross-section of the j-th circular slice layer of the ink droplet in another projected image region, where j∈{1,2,…,n}.

[0032] A further beneficial effect of the present invention is that by increasing the volume calculation weight of the clearer projected image region and nonlinearly reducing the volume calculation weight of the blurrier projected image region, the weighted averaging method fully utilizes the effective information in the image, thereby improving the sampling rate and reliability of the volume measurement results.

[0033] Furthermore, when each nozzle's corresponding droplet image contains only the upper and lower projection images of a single droplet, the method for calculating the flight velocity of the droplet from that nozzle based on its corresponding droplet image is as follows:

[0034] Determine the time interval Δt between applying two consecutive flash signals to the same ink droplet when acquiring the image of the ejected ink droplet corresponding to each nozzle;

[0035] Edge detection is performed on the upper and lower projected image regions of the ink droplet image corresponding to each nozzle. Based on the two contour lines corresponding to the upper and lower projected image regions after edge detection, the coordinates of the two contours in the camera image coordinate system R are determined. Cam The coordinates of the two centroids (x1, z1) and (x2, z2) are given; the instantaneous velocity of the ink droplet in the camera's observation direction is calculated based on the following expression.

[0036] Alternatively, if each nozzle's corresponding droplet image contains two projected images of Q droplets, then the flight velocity of the droplets ejected from that nozzle is calculated based on the droplet image corresponding to that nozzle as follows:

[0037] Determine the time interval Δt between applying two flash signals to the same ink droplet when acquiring the image of the ejected ink droplet corresponding to the nozzle;

[0038] Taking the two projected image regions above and below the same ink droplet as a group, the ink droplet contour lines are segmented from all projected image regions corresponding to Q ink droplets in the image of the ejected ink droplets corresponding to the nozzle. The coordinates in the camera image system R are determined based on the contour lines. Cam Multiple sets of centroid coordinates (x) 1，1 , z 1，1 (x) 1，2 , z 1，2 ), (x 2，1 , z 2，1 (x) 2，2 , z 2，2 ), ..., (x Q，1 , z Q，1 (x) Q，2 , z Q，2 );

[0039] Calculate the flight velocity of the ejected ink droplets from the nozzle. In the formula, λ i The weight of the projection image region corresponding to the i-th ink droplet is used to calculate the flight speed of the ejected ink droplets in the nozzle. The value is determined according to the position of the projection image region corresponding to the i-th ink droplet in the ejected ink droplet image of the nozzle. The weight is greater when the region is located in the middle of the ejected ink droplet image of the nozzle than when it is located at the edge of the ejected ink droplet image of the nozzle.

[0040] A further beneficial effect of the present invention is that, given the weak correlation between angle measurement accuracy and projection sharpness, this method fully utilizes the information in the image rather than selecting only a local image of the clear part to calculate the ink droplet velocity, thereby improving the sampling rate and reliability of velocity measurement.

[0041] Furthermore, each nozzle's corresponding image contains two projected images of the ink droplet, one above the other. Therefore, based on the image of the ink droplet corresponding to each nozzle, the flight angle of the ink droplet at that nozzle is calculated as follows:

[0042] Based on the centroid coordinates (x1, z1) and (x2, z2) of the same ink droplet's projected images at two consecutive moments within a single camera frame, the flight angle of the flying ink droplet in the camera's observation direction is calculated using the following expression.

[0043] Furthermore, if each nozzle corresponds to an ink droplet image containing W projected images, the method for calculating the flight angle of the ink droplets from that nozzle based on the ink droplet image is as follows:

[0044] In the image of the ejected ink droplets corresponding to the nozzle, the ink droplet contour lines are segmented from W projected images, and the coordinates in the camera image system R are determined based on the corresponding ink droplet contour lines. Cam Multiple centroid coordinates (x1, z1), (x2, z2), ..., (x w ,z w Based on W centroid coordinates, a linear regression is used to fit a linear function f(x) = kz + b. The following expression is used to measure the flight angle γ of the flying ink droplet in the camera observation direction: γ = tan -1 k.

[0045] A further beneficial effect of the present invention is that, compared with using only the coordinates of two projected images at two different times, this method utilizes all the coordinate data of the projected images in the image, thereby improving the sampling rate and accuracy of angle detection.

[0046] This invention also provides a printhead array ink droplet detection system for performing the printhead array ink droplet detection method described above. The detection system includes: a control module, a data processing module, a vision inspection module, and a mechanical auxiliary module. The vision inspection module includes multiple vision systems arranged in an array, with the spacing between two adjacent vision systems being an integer multiple of the spacing between adjacent printheads in the printhead array. The mechanical auxiliary module includes a motion frame and multiple position fine-tuning mechanisms configured on the motion frame for each vision system.

[0047] The control module is used to control the movement of the motion frame and / or the printhead array to be detected according to the coordinates of the multiple nozzles to be detected synchronously, and to control the position fine-tuning structure of each position to fine-tune the position of its corresponding vision system so that each vision system is located at the image acquisition position of the ejected ink droplets of its nozzle to be detected, and to control the multiple vision systems to synchronously acquire images of ejected ink droplets from the nozzles of multiple printheads; the data processing module is used to calculate the flight state of ejected ink droplets of each nozzle based on the ejected ink droplet image corresponding to each nozzle.

[0048] The multiple vision systems are mounted on the motion frame. The coordinates of each nozzle are determined as follows: the angle between a certain axis vector of each nozzle's Cartesian coordinate system and the corresponding axis vector of the global coordinate system of the motion frame is obtained as the nozzle offset angle of that nozzle; the coordinate position of each nozzle in the corresponding nozzle's Cartesian coordinate system is multiplied by the corresponding nozzle offset angle, and the coordinates of the origin of the corresponding nozzle's Cartesian coordinate system in the global coordinate system are added to obtain the coordinates of each nozzle in the global coordinate system. The two coordinate axes of the global coordinate system correspond to the same two mutually perpendicular degrees of freedom directions of the motion frame and / or the nozzle array.

[0049] The beneficial effects of this invention are: This invention designs a printhead array ink droplet detection system, which, based on the corresponding observation position calibration and detection method, can execute the above detection process and effectively improve detection efficiency while ensuring detection accuracy.

[0050] In summary, the above-described technical solutions conceived in this invention can achieve the following beneficial effects:

[0051] This invention, through the design of multiple vision inspection systems and corresponding observation position calibration and inspection methods, enables the simultaneous scanning and inspection of ink droplets ejected from multiple nozzles within a printhead array. Compared to the limitation of existing inspection schemes that can only inspect a single nozzle at a time, this invention effectively improves inspection efficiency. Furthermore, the inspection method provided by this invention can compensate for printhead splicing errors in the printhead array, keeping the ink droplets at the observation position, which helps reduce defocusing and improves inspection accuracy. Additionally, the inspection method provided by this invention can measure multiple parameters such as the volume, velocity, and ejection angle of the flying ink droplets. Attached Figure Description

[0052] Figure 1 A flowchart illustrating a method for detecting ink droplets ejected from a printhead array, provided in an embodiment of the present invention.

[0053] Figure 2 This is a schematic diagram of the overall structure of the printhead array ink droplet synchronous scanning and detection system provided in an embodiment of the present invention;

[0054] Figure 3 This is a timing diagram of the vision system control signals provided in an embodiment of the present invention;

[0055] Figure 4 A schematic diagram illustrating the method for measuring the volume, ejection angle, and velocity of flying ink droplets provided in an embodiment of the present invention;

[0056] Figure 5 This is a three-dimensional structural diagram of the printhead array ink droplet synchronous scanning and detection system provided in an embodiment of the present invention;

[0057] Figure 6 This is a flowchart illustrating the overall method for synchronous scanning and detection of ink droplets ejected from a printhead array, as provided in an embodiment of the present invention.

[0058] In all the accompanying drawings, the same reference numerals are used to denote the same elements or structures, wherein:

[0059] 10 is the printhead, 20 is the strobe light source, 21 is the camera, 22 is the lens, 30 is the position fine-tuning mechanism, 31 is the motion frame, 40 is the control module, and 50 is the ink collection device. Detailed Implementation

[0060] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0061] Example 1

[0062] A method for detecting ink droplets ejected from a printhead array, such as Figure 1 As shown, it includes:

[0063] Control multiple vision systems to simultaneously acquire images of ink droplets ejected from the nozzles of multiple printheads; calculate the flight state of the ink droplets ejected from each nozzle based on the image of the ink droplets ejected from that nozzle.

[0064] By combining the coordinates of the nozzles currently detected by each vision system with the coordinates of the nozzles to be detected next, the position of the moving frame and / or printhead array is adjusted as a whole, and the position of the vision system that has not reached the image acquisition position of the ink droplets of the nozzles to be detected after the overall adjustment is fine-tuned; image acquisition is repeated until the ink droplet detection of all nozzles to be detected in the printhead array is completed.

[0065] The aforementioned multiple vision systems are mounted on a motion frame. The coordinates of each nozzle are located in the global coordinate system of the motion frame. The two coordinate axes of the global coordinate system correspond to the two mutually perpendicular degrees of freedom directions of the motion frame and / or the nozzle array. The coordinates of each nozzle are determined as follows: the angle between a certain axis vector of the Cartesian coordinate system of each nozzle and the corresponding axis vector of the global coordinate system of the motion frame is obtained as the nozzle offset angle of that nozzle. The coordinate position of each nozzle in the corresponding nozzle Cartesian coordinate system is multiplied by the nozzle offset angle of the corresponding nozzle, and the coordinates of the origin of the corresponding nozzle Cartesian coordinate system in the global coordinate system are added to obtain the coordinates of each nozzle in the global coordinate system.

[0066] The L vision systems can employ back-illuminated stroboscopic exposure to achieve simultaneous detection of L printheads within the printhead array. The number L of printheads to be detected simultaneously can be selected according to actual needs. In practical applications of this method, the vision system may include a camera, a lens, and a stroboscopic light source. All cameras are positioned at the same height and are used to collect, through their respective lenses, the flying ink droplets ejected from the nozzles of each printhead within the printhead array.

[0067] For the printhead array to be observed, the printhead setup in the actual printing system has the following characteristics: the printhead array is composed of multiple printheads 10 of the same specification spliced ​​together in the same direction, and the printhead array is installed on a moving part with Y1 and Z1 degrees of freedom, that is, the first horizontal longitudinal degree of freedom and the first vertical degree of freedom; multiple nozzles of the printhead are located on the lower surface E of the printhead, and the nozzles of each printhead in the printhead array are located on the same horizontal plane F, that is, after the printheads are spliced, E = F.

[0068] Based on the characteristics of this nozzle setup, this method considers that: theoretically, there should be no included angle between adjacent nozzles on a plane; however, in reality, there will be misalignment on the plane, with a certain angle between adjacent nozzles. The nozzle offset angle indicates the degree of misalignment. To achieve efficient and precise adjustment of the vision system position, it is assumed that there are M nozzles in the nozzle array. For all M nozzles in the nozzle array, they are all located on their lower surfaces E1…E M Internally defined nozzle Cartesian coordinate system And construct set R H =∪ i∈{1，…，M} R head，i ,for The actual physical positional relationship between the i-th nozzle and R head，j The actual physical positional relationship between the j-th nozzles is the same; a global Cartesian coordinate system for the flying droplet detection device is defined in plane F. basis vectors The direction is the same as the degree of freedom X1, and the basis vectors are... The direction is the same as the degree of freedom Y1; when E i / / F or E i When =F, the basis vectors and The included angle between them is defined as the nozzle array splicing offset angle β. i (Caused by installation error), where i∈{1,2,…,M}, such as Figure 2 As shown.

[0069] The nozzle position of the i-th nozzle in the above nozzle array is preferably determined using coordinate system R. head，i The coordinates below are represented using matrices. This represents the coordinate array of all nozzles in this nozzle, where N is the number of nozzles in a single nozzle; and preferably, the nozzle coordinate array A of the i-th nozzle is calculated in the following manner. i Rotation β i Then in the global Cartesian coordinate system R global The coordinate array B below i :

[0070] First, calculate matrix A based on the following expression. i Rotation β i The matrix A′ after i :

[0071]

[0072] Then, according to the coordinate system R of the i-th nozzle... head，i Origin O i In the global coordinate system R global coordinates below (x) i，0 y i，0 ), calculate the position of the i-th nozzle in the global coordinate system R based on the following expression. global The lower nozzle coordinate array B i :

[0073] B i =A′ i +D i ;

[0074] Where the matrix Sum matrix A′ i The dimension of both is 2xN.

[0075] This embodiment proposes a detection method for synchronous scanning of ink droplets ejected from a printhead array. By using parameters such as the printhead array splicing offset angle (theoretically, there should be no angle between adjacent printheads on a plane, but in reality, there will be splicing misalignment, with a certain angle between adjacent printheads; the printhead offset angle indicates the degree of misalignment) and the coordinates of the nozzle positions, the system calculates the motion step size of each degree of freedom during the scanning and detection of all nozzles in the printhead array. This allows multiple vision inspection systems to quickly align with the ink droplets ejected from their respective nozzles, synchronously performing the detection of multiple printhead nozzles in the printhead array. Compared to conventional flying droplet detection schemes, this overcomes the limitation of only being able to detect a single nozzle at a time, effectively improving observation efficiency. Furthermore, the synchronous scanning and detection method for ink droplets ejected from a printhead array designed in this invention does not limit the number of printheads that can be detected simultaneously. Only the addition of corresponding vision systems and motion mechanisms is needed to increase the number of printheads that can be detected simultaneously, enabling the detection of larger-scale printhead arrays and achieving high-efficiency detection.

[0076] As a preferred embodiment, the above-mentioned method for fine-tuning the position of the overall adjustment of the motion frame and / or printhead array, and for fine-tuning the position of the vision system that has not reached the image acquisition position of the ejected ink droplets that are not detected after the overall adjustment, is as follows:

[0077] The spacing between two adjacent vision systems is set to an integer multiple of the spacing between adjacent nozzles in the nozzle array to be inspected; and a position fine-tuning mechanism is configured for each vision system.

[0078] A vision system is selected as the reference vision system. Based on the coordinates of its pre-detection nozzle, the motion frame and / or the printhead array to be detected are controlled to move in the two degrees of freedom, so that the reference vision system is located at the image acquisition position of the ejected ink droplets of its pre-detection nozzle. Based on the coordinates of the pre-detection nozzles of each other vision system, the position of each vision system is adjusted in the two degrees of freedom to the image acquisition position of the ejected ink droplets of its pre-detection nozzle through the position fine-tuning mechanism of each vision system.

[0079] Due to size limitations of the vision system, the spacing between adjacent fine-tuning mechanisms and the vision system is an integer multiple of the spacing between adjacent nozzles within the array; this multiple can be adjusted according to the nozzle size. The vision system can be mounted on the fine-tuning mechanism and can move with it.

[0080] The spacing between two adjacent vision systems is set to an integer multiple of the spacing between adjacent printheads within the printhead array to be inspected. This allows for simultaneous inspection of multiple printhead orifices during each inspection. Furthermore, each vision system is equipped with a position fine-tuning mechanism. One vision system is selected as a reference system and adjusted to the image acquisition position of the ink droplet ejection from its respective printhead. Other vision systems then perform position fine-tuning through their respective mechanisms, achieving efficient position adjustment of the vision systems for simultaneous inspection.

[0081] As a further preferred embodiment, for the above-mentioned fine-tuning mechanism, such as Figure 2 As shown, the coordinate system of each fine-tuning mechanism has two axes, U and V. The preferred method for calculating the motion step matrix C of the l-th fine-tuning mechanism when simultaneously observing L nozzles using L sets of vision systems is as follows. l To compensate for the splicing offset of the nozzle array (β) i ≠0) causes nozzle misalignment, reducing out-of-focus blur:

[0082] Choosing the k-th vision system as the baseline, i.e., keeping the fine-tuning mechanism of this vision system fixed, C k = 0, where l, k∈{1, 2, ..., L}. Then the motion step size matrix C of the two-degree-of-freedom fine-tuning mechanism of the l-th visual system is... l The following calculation method is preferred:

[0083]

[0084] In the formula, u l，n This indicates that when the position fine-tuning mechanism of the l-th vision system observes the n-th nozzle of the H(l)-th nozzle, it has a position in degree of freedom U. l The direction of motion, v l，n This indicates that when the position fine-tuning mechanism of the l-th vision system observes the n-th nozzle of the H(l)-th nozzle, it has a position in degree of freedom V. l The direction of motion, l∈{1, 2, ..., L}, n∈{1, 2, ..., N}, where L is the total number of vision systems and N is the total number of nozzles in each nozzle; A′ H(l) A represents the nozzle coordinate array matrix A of nozzle number H(l), consisting of the coordinates of all its nozzles in the Cartesian coordinate system for the nozzle. H(l) Rotation β i The matrix after that, A′ H(k) A represents the nozzle coordinate array matrix A, which is the coordinates of all nozzles of nozzle H(k) observed by the k-th vision system as a reference, in the Cartesian coordinate system for nozzles. H(k) Rotation β k The matrix after that.

[0085] In other embodiments, a threshold δ can be introduced, which is affected by the splicing offset of the nozzle array and the motion accuracy of the fine-tuning mechanism. If any element in the motion step matrix C of the fine-tuning mechanism has a calculation result less than the threshold, it can be set to zero.

[0086] As a further preferred embodiment, each vision system performs image acquisition by using stroboscopic exposure and performing double flashes on each ejected ink droplet within an exposure time period, so that the acquired image contains the projected images of the same ink droplet at two different times.

[0087] In other words, for the camera in the vision system, its exposure signal is set in the synchronization trigger controller to lag behind the ejection signal of its corresponding observation nozzle, and the camera's exposure lag time is independently adjustable. Correspondingly, for the stroboscopic source, the sum of its two flash signals and the interval between the two flashes is set to be included in the aforementioned exposure signal; that is, the sum of the width of its two flash pulses and the pulse interval is less than the width of the camera's exposure signal, and its flash pulse width, flash pulse interval, and flash lag time are also independently adjustable in the synchronization trigger controller. The two flash signals constitute a set of double flashes, which are reflected in the same frame image, showing the projected images of the same ink droplet at two different moments.

[0088] The proposed method for synchronous scanning and detection of ink droplets ejected by a printhead array employs a design where the light source flashes twice within a single camera exposure. This allows for the acquisition of projected images of the flying ink droplets at two different moments within a single frame image, with the time interval between the two flashes being adjustable. This image acquisition method of the vision system can calculate the droplet flight speed and angle using only a single frame image. Compared to conventional detection schemes that acquire two frames before and after acquisition and then compare them to calculate the flying ink droplet speed and angle, this significantly reduces measurement time and saves detection steps.

[0089] As a further preferred implementation, each vision system, when acquiring images of the ejected ink droplets from the corresponding nozzle, specifically does the following:

[0090] During an exposure period, multiple ink droplets before and after the ejection are subjected to double flashes to obtain an image of the ejected ink droplets from the nozzle. The projection image on top of the ejected ink droplet image is a superposition of the projection images captured at the previous moment when each ejected ink droplet is subjected to double flashes, and the projection image at the back of the ejected ink droplet image is a superposition of the projection images captured at the later moment when each ejected ink droplet is subjected to double flashes.

[0091] Specifically, such as Figure 3As shown, the exposure signal S2 of camera 21 lags behind the jet signal S1 of nozzle 10. The lag time t1 of signal S2 is independently adjustable. Due to hardware limitations, the frequency of the exposure signal S2 can be lower than that of the jet signal S1. The dual-pulse flash signal S3 of light source 20 is included in S2, that is, the first pulse width T of S3 is... 2_1 Pulse interval time T2 and second pulse width T 2_2 The sum of the pulse widths T1 and lag times t2 of the signals are less than that of S2, and these pulse widths are also independently adjustable. In other vision systems, the camera exposure signal S4 is synchronized with S2 (i.e., S2 = S4, but can also be asynchronous), and the light source flash signal S5 is synchronized with S3 (i.e., S3 = S5, but can also be asynchronous). The lag times t3 and t4 are also independently adjustable. The pulse flash signal width corresponding to S4 is T3, and the first pulse width of S5 is T... 4_1 Pulse interval time T4 and second pulse width T 4_2 The sum of these pulse widths is less than the pulse width T3 of S4. In other embodiments, the light source flash signal S3 can be set to the same frequency as the jet signal S1, extending the camera's exposure time T1 so that it contains multiple sets of double flashes within one exposure time period, allowing the camera to capture superimposed ink droplet images, which can improve image contrast.

[0092] As a further preferred implementation, firstly, printheads H(1) to H(L) simultaneously eject ink droplets at the same frequency; then, L sets of vision systems simultaneously expose and capture images; and edge detection is performed on the projected image of the ink droplets in the observation direction using a certain threshold. Based on the camera calibration value, the actual height h of the ink droplets is calculated; then, as... Figure 4 As shown, the ink droplet is discretized into n slices, each with a thickness of Δh. The cross-section of each slice is fitted into a circle with area A and diameter d. The actual volume V of the ink droplet is then calculated based on the following expression:

[0093]

[0094] Where, d i Let represent the diameter of the cross-section of the i-th slice layer, where i∈{1,2,…,n}.

[0095] As a further preferred embodiment, for the above-mentioned volume calculation method, it is preferable to use the Laplacian of Gaussian (LoG) transform to process the projected image regions of the same ink droplet at two different time points, corresponding to two flashes respectively, and to calculate the gray-level variance σ of the image region after LoG processing. 2 As a sharpness evaluation metric, a weighting coefficient λ is introduced as the weight for calculating the corresponding volume of each image region. The non-linear mapping relationship between the weighting coefficient λ and sharpness is defined based on the following expression:

[0096]

[0097] Where t is the gray-level variance σ of the image region. 2 The value is normalized to the interval [-10, 10]. The volume calculation method described above is further optimized by calculating the volume-weighted average of the two ink droplet image regions, and the volume V of the ink droplet is calculated based on the following expression:

[0098]

[0099] Where λ1 and λ2 represent the weighting coefficients of the two ink droplet image regions, respectively, and d i d represents the diameter of the cross-section of the i-th circular slice layer of the ink droplet in the first image region, i∈{1,2,…,n}. j Let represent the diameter of the cross-section of the j-th circular slice layer of the ink droplet in the second image region, where j∈{1,2,…,n}.

[0100] As a further preferred embodiment, the measurement process of the ink droplet velocity ejected by the printhead array is preferably performed in the following manner:

[0101] First, project the same ink droplet into two different time points within a single frame, corresponding to two flashes of light, with an interval of Δt between the two flashes. Then, segment the two projected image regions to obtain two ink droplet contour lines. Based on these two contour lines, define the image regions in the camera image coordinate system R... Cam The two centroid coordinates (x1, z1) and (x2, z2) are used to calculate the instantaneous velocity S of the ink droplet in the direction perpendicular to the camera's observation, based on the following expression:

[0102]

[0103] As a further preferred embodiment, for the above-mentioned speed calculation method, it is preferable to realize that there are Q ink droplets projected in a single frame of camera image at two consecutive time points. Two projected images of the same ink droplet are grouped together, and the upper and lower projections of each group of images correspond to two flashes, with an interval of Δt between the two flash signals. Then, the ink droplet contour lines are segmented from all projected images, and the coordinates of the ink droplets in the camera image coordinate system R are calculated based on the contour lines. Cam Multiple sets of centroid coordinates (x) 1，1 , z 1，1 (x) 1，2 , z 1，2 ), (x 2，1 , z 2，1 (x) 2，2 , z 2，2 ), ..., (x Q，1 , z Q，1 (x) Q，2 , z Q，2A weighting coefficient λ is introduced as the weight for calculating the velocity corresponding to each group of images. If Q is odd, then let the th... The velocity weighting coefficient λ for the first group of images is 1, and the weighting coefficient for the remaining groups of images is 0.8; if Q is even, then let the first group be... The velocity weighting coefficient λ for the two image groups is 1, and the weighting coefficient for the remaining image groups is 0.8. The above velocity calculation method is further optimized by calculating the weighted average velocity of multiple ink droplet images, and the velocity S of the ink droplet is calculated based on the following expression:

[0104]

[0105] According to another preferred embodiment of the present invention, based on the centroid coordinates (x1, z1) and (x2, z2) of the same ink droplet projected at two different times within a single frame of a camera image, the following expression is preferably used to perform the process of measuring the flight angle γ of the flying ink droplet in the direction perpendicular to the camera observation:

[0106]

[0107] As a further preferred embodiment, for the above angle calculation method, it is preferable to use the synchronous trigger controller in the control module to realize that there are W ink droplet projection images in one frame of camera image; then, the ink droplet contour lines are segmented from all the projection images respectively, and the coordinates in the camera image coordinate system R are obtained according to the contour lines respectively. Cam Multiple sets of centroid coordinates (x1, z1), (x2, z2), ..., (x w , z w Based on all W coordinates, a linear regression is used to fit a linear function f(x) = kz + b. The following expression is preferred for measuring the flight angle γ of the flying ink droplet in the direction perpendicular to the camera observation: γ = tan -1 k.

[0108] Example 2

[0109] A printhead array ink droplet detection system is provided for performing the printhead array ink droplet detection method described in Embodiment 1 above. The detection system includes a control module, a data processing module, a vision detection module, and a mechanical auxiliary module. The vision detection module includes multiple vision systems arranged in an array, with the spacing between two adjacent vision systems being an integer multiple of the spacing between adjacent printheads in the printhead array. The mechanical auxiliary module includes a motion frame and multiple position fine-tuning mechanisms configured on the motion frame for each vision system.

[0110] The control module is used to control the movement of the motion frame and / or the array of nozzles to be detected according to the coordinates of the multiple nozzles to be detected synchronously, and to control the position fine-tuning structure of each position to fine-tune the position of its corresponding vision system so that each vision system is located at the position of acquiring the ink droplet image of the nozzle to be detected, and to control multiple vision systems to simultaneously acquire ink droplet images of the nozzles of multiple nozzles; the data processing module is used to calculate the ink droplet flight state of the nozzle based on the ink droplet image corresponding to each nozzle.

[0111] Multiple vision systems are mounted on a motion frame. The coordinates of each nozzle are located in the global coordinate system of the motion frame. The two axes of the global coordinate system correspond to the two mutually perpendicular degrees of freedom of the motion frame and / or the nozzle array. The coordinates of each nozzle are determined as follows: the angle between a certain axis vector of the Cartesian coordinate system of each nozzle and the corresponding axis vector of the global coordinate system of the motion frame is obtained as the nozzle offset angle of the nozzle. The coordinate position of each nozzle in the corresponding nozzle Cartesian coordinate system is multiplied by the nozzle offset angle of the corresponding nozzle and the coordinates of the origin of the corresponding nozzle Cartesian coordinate system in the global coordinate system are added to obtain the coordinates of each nozzle in the global coordinate system.

[0112] Specifically, such as Figure 2 As shown, the printhead array ink droplet synchronous scanning detection system mainly includes a vision detection module, a control module, a mechanical auxiliary module, and a printhead array to be tested.

[0113] The visual inspection module includes L sets of vision systems using back-illuminated stroboscopic exposure to achieve synchronous inspection of L nozzles in the nozzle array. The number of nozzles L to be inspected synchronously can be selected according to actual needs. The vision system includes a camera 22, a lens 21, and a stroboscopic light source 20. All cameras are arranged at the same height and are used to collect flying ink droplets ejected from the nozzles in each nozzle in the nozzle array through their lenses.

[0114] The nozzle configuration has the following characteristics: the nozzle array is composed of multiple nozzles 10 of the same specification spliced ​​together in the same direction, and the nozzle array is installed on a moving part with Y1 and Z1 degrees of freedom, that is, the first horizontal longitudinal degree of freedom and the first vertical degree of freedom; multiple nozzle holes are located on the lower surface E of the nozzle, and the nozzle holes of each nozzle in the nozzle array are located on the same horizontal plane F, that is, after the nozzles are spliced, E = F.

[0115] For mechanical auxiliary modules, such as Figure 2 and Figure 5As shown, it includes a motion frame 31 and L sets of two-degree-of-freedom position fine-tuning mechanisms 30 arranged at equal intervals along the Y1 direction on the motion frame 31. The motion frame 31 has one degree of freedom X1, i.e., a first horizontal degree of freedom. Due to the size limitation of the vision system, the spacing between adjacent fine-tuning mechanisms is an integer multiple of the spacing between adjacent nozzles in the array, and this multiple can be adjusted according to the nozzle size. The vision system (camera 21, lens 22, strobe light source 20) is mounted on the position fine-tuning mechanism 30 and can move with it. The position fine-tuning mechanism 30 has two mutually perpendicular degrees of freedom U. l V l And the degree of freedom U l The direction of freedom X1 is the same as that of freedom V. l The direction is the same as that of degree of freedom Y1, where l∈{1,2,…,L}; the ink collection device 50 is used to collect ink droplets ejected from the nozzle of the printhead 10, and the size of the ink collection device 50 should be larger than the corresponding spray range of the printhead 10; specifically, the optical axes of the camera 21, lens 22 and stroboscopic light source 20 are at the same height and collinear, and are a few millimeters higher than the upper surface of the ink collection device 50.

[0116] like Figure 2 As shown, the control module 40 is connected to the printhead 10, light source 20, camera 21, and position fine-tuning mechanism 30 in the printhead array. It includes a motion controller, a synchronous trigger controller, and a printhead controller. The printhead controller is used to control the opening and closing of the printhead nozzles, thereby controlling the nozzles to spray and outputting a spray trigger signal to the synchronous trigger controller. The synchronous trigger controller receives the spray trigger signal and delays to trigger the visual detection module to work, thereby performing real-time detection of the flying ink droplets, including volume, velocity, and angle. The motion controller is used to control the position of the moving parts in each degree of freedom.

[0117] This embodiment uses L sets of vision systems arranged along the Y1 degree of freedom to simultaneously capture images of ink droplets ejected from multiple nozzles within the printhead array, obtaining information such as droplet volume, velocity, and angle. It is required that the L cameras 21 can convert pixel coordinates to world coordinates for images acquired from different targets, necessitating the calibration of each of the L cameras 21. For example, high-precision calibration spheres can be placed at the working distance of each camera 21. Then, the cameras 21 are controlled to capture images, and the conversion relationship between pixel coordinates and world coordinates is obtained based on the pixel coordinates in the image and the size of the calibration spheres, thus completing the calibration of the cameras 21.

[0118] Based on the detection system, synchronous scanning detection of ink droplets ejected from the printhead array can be performed according to the following process, which can be divided into two parts: calibration and scanning detection, as detailed below. Figure 6 As shown:

[0119] (a) First, coarse positioning: Using the control module, the first nozzle of the currently observed nozzle in the nozzle array ejects ink droplets. The frame 31 is moved in the X1 direction and the nozzle array 10 is moved in the Y1 direction. The nozzle array to be measured is moved in front of the vision inspection module so that there is a nozzle near the camera of each vision system. Then, precise positioning: The k-th vision system is selected as the reference. The first nozzle of the H(k)-th nozzle observed by the k-th vision system ejects ink droplets, and the k-th vision system observes the ink droplets ejected by it, where k∈{1,…,L}.

[0120] (b) Temporarily fix the position of the frame and printhead array, and use the control module to adjust the remaining fine-tuning mechanisms except for the kth one, so that all L sets of vision systems can detect the ink droplets ejected from the first nozzle of the corresponding printhead.

[0121] (c) A vision inspection module is used to perform real-time vision inspection of the ink droplets ejected by L printheads in the printhead array, including volume, velocity and angle.

[0122] (d) Using a control module, the frame is moved along the X1 degree of freedom and the nozzle array is moved along the Y1 degree of freedom according to the nozzle coordinate array of the H(k) nozzle. When switching to the next nozzle of the H(k) nozzle, the remaining fine adjustment mechanisms except for the k nozzle move to keep each camera aligned with the next nozzle.

[0123] (e) Using a control module to make the next nozzle of the printhead to be observed in the printhead array eject ink droplets;

[0124] (f) Repeat steps (c), (d) and (e) to sequentially traverse the remaining nozzles of printhead H(k) until the ink droplet detection of all nozzles is completed.

[0125] Multiple vision systems simultaneously observe multiple nozzles, and the observation progress (number of nozzles) of each nozzle can be the same, ultimately achieving the detection of all nozzles and all nozzles.

[0126] The relevant technical solutions are the same as in Embodiment 1, and will not be repeated here.

[0127] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for detecting ink droplets ejected from a printhead array, characterized in that, include: Control multiple vision systems to simultaneously acquire images of ink droplets ejected from the nozzles of multiple printheads; Based on the ejected ink droplet image corresponding to each nozzle, calculate the ejected ink droplet flight state of that nozzle. By combining the coordinates of the nozzles currently detected by each vision system with the coordinates of the nozzles to be detected next, the position of the moving frame and / or printhead array is adjusted as a whole, and the position of the vision system that has not reached the image acquisition position of the ink droplets of the nozzles to be detected after the overall adjustment is fine-tuned; image acquisition is repeated until the ink droplet detection of all nozzles to be detected in the printhead array is completed. The multiple vision systems are mounted on the motion frame. The coordinates of each nozzle are determined as follows: the angle between a certain axis vector of the pre-established nozzle Cartesian coordinate system of each nozzle and the corresponding axis vector of the global coordinate system is obtained as the nozzle offset angle of that nozzle; based on the nozzle offset angle of each nozzle, the coordinates of each nozzle in the corresponding nozzle Cartesian coordinate system are rotated by an angle and the coordinates of the origin of the corresponding nozzle Cartesian coordinate system in the global coordinate system are added to transform the coordinates of each nozzle in each nozzle to the global coordinate system, thus obtaining the coordinates of each nozzle in the global coordinate system; the two coordinate axes of the global coordinate system correspond to the two mutually perpendicular degrees of freedom directions of the motion frame and / or the nozzle array.

2. The method for detecting ink droplets ejected from a printhead array according to claim 1, characterized in that, The method for fine-tuning the position of the overall adjustment motion frame and / or printhead array, and for the vision system that, after overall adjustment, has not reached the image acquisition position of the ejected ink droplets at the nozzle to be detected, is as follows: The spacing between two adjacent vision systems is set to an integer multiple of the spacing between adjacent nozzles in the nozzle array to be inspected; and a position fine-tuning mechanism is configured for each vision system. A vision system is selected as the reference vision system. Based on the coordinates of its pre-detection nozzle, the motion frame and / or the printhead array to be detected are controlled to move in the two degrees of freedom, so that the reference vision system is located at the image acquisition position of the ejected ink droplets of its pre-detection nozzle. Based on the coordinates of the pre-detection nozzles of each other vision system, the position of each vision system is adjusted in the two degrees of freedom to the image acquisition position of the ejected ink droplets of its pre-detection nozzle through the position fine-tuning mechanism of each vision system.

3. The method for detecting ink droplets ejected from a printhead array according to claim 2, characterized in that, The motion step size matrix of the two degrees of freedom of the fine-tuning mechanism corresponding to the two degrees of freedom directions of each position is set as follows: In the formula, u l,n This indicates that when the position fine-tuning mechanism of the l-th vision system observes the n-th nozzle of the H(l)-th nozzle, it has a position in degree of freedom U. l The direction of motion, v l,n This indicates that when the position fine-tuning mechanism of the l-th vision system observes the n-th nozzle of the H(l)-th nozzle, it has a position in degree of freedom V. l The direction of motion, l∈{1,2,…,L}, n∈{1,2,…,N}, where L is the total number of vision systems and N is the total number of nozzles in each nozzle; A ′ H(l) A represents the nozzle coordinate array matrix A of nozzle number H(l), consisting of the coordinates of all its nozzles in the Cartesian coordinate system for the nozzle. H(l) Rotation β i The matrix after A ′ H(k) A represents the nozzle coordinate array matrix A, which is the coordinates of all nozzles of nozzle H(k) observed by the k-th vision system as a reference, in the Cartesian coordinate system for nozzles. H(k) Rotation β k The matrix after that.

4. The method for detecting ink droplets ejected from a printhead array according to claim 1, characterized in that, Each vision system uses stroboscopic exposure and double flashes for each ejected ink droplet within an exposure time period to perform image acquisition, so that the acquired image contains the projected images of the same ink droplet at two different times.

5. The method for detecting ink droplets ejected from a printhead array according to claim 4, characterized in that, When each vision system acquires images of the ejected ink droplets from the corresponding nozzle, the specific steps are as follows: During an exposure period, multiple ink droplets before and after the ejection are subjected to double flashes to obtain an image of the ejected ink droplets from the nozzle. The projection image on top of the ejected ink droplet image is a superposition of the projection images captured at the previous moment when each ejected ink droplet is subjected to double flashes, and the projection image at the back of the ejected ink droplet image is a superposition of the projection images captured at the later moment when each ejected ink droplet is subjected to double flashes.

6. The method for detecting ink droplets ejected from a printhead array according to claim 1, characterized in that, Based on the image of the ejected ink droplets corresponding to each nozzle, the actual volume of the ejected ink droplets for that nozzle is calculated as follows: Edge detection is performed on the upper and lower projected image regions of the ink droplet image corresponding to each nozzle, and the actual height of the ink droplet corresponding to each projected image region after edge detection is calculated based on the camera calibration value; the ink droplet corresponding to each projected image region is discretized into n slices with a thickness of Δh in the height direction; The Laplace transform of Gaussian is used to process each projected image region after edge detection, and the gray-level variance σ of the projected image region after the Laplace transform of Gaussian is calculated. 2 The weighting coefficient λ for each projected image region is determined based on the following expression: In the formula, t is the gray-level variance σ of the projected image region. 2 The value after normalization to the interval [-10, 10]; The actual volume V of the ejected ink droplets from this nozzle is calculated based on the following expression: Where λ1 and λ2 represent the weighting coefficients of the upper and lower projected image regions, respectively, and d i Let d represent the diameter of the cross-section of the i-th circular slice layer of an ink droplet in one of the projected image regions, i∈{1,2,…,n}. j Let represent the diameter of the cross-section of the j-th circular slice layer of the ink droplet in another projected image region, where j∈{1,2,…,n}.

7. The method for detecting ink droplets ejected from a printhead array according to claim 5, characterized in that, When each nozzle's corresponding droplet image contains only the upper and lower projection images of a single droplet, the method for calculating the flight velocity of the droplet from that nozzle based on its corresponding droplet image is as follows: Determine the time interval Δt between applying two consecutive flash signals to the same ink droplet when acquiring the image of the ejected ink droplet corresponding to each nozzle; Edge detection is performed on the upper and lower projected image regions of the ink droplet image corresponding to each nozzle. Based on the two contour lines corresponding to the upper and lower projected image regions after edge detection, the coordinates of the two contours in the camera image coordinate system R are determined. Cam The coordinates of the two centroids (x1, z1) and (x2, z2) are given; the instantaneous velocity of the ink droplet in the camera's observation direction is calculated based on the following expression. Alternatively, if each nozzle's corresponding droplet image contains two projected images of Q droplets, then the flight velocity of the droplets ejected from that nozzle is calculated based on the droplet image corresponding to that nozzle as follows: Determine the time interval Δt between applying two flash signals to the same ink droplet when acquiring the image of the ejected ink droplet corresponding to the nozzle; Taking the two projected image regions above and below the same ink droplet as a group, the ink droplet contour lines are segmented from all projected image regions corresponding to Q ink droplets in the image of the ejected ink droplets corresponding to the nozzle. The coordinates in the camera image system R are determined based on the contour lines. Cam Multiple sets of centroid coordinates (x) 1, ,z 1, (x) 1, ,z 1, ),(x 2, ,z 2, (x) 2, ,z 2, ),…,(x Q,1 ,z Q,1 (x) Q,2 ,z Q,2 ); Calculate the flight velocity of the ejected ink droplets from the nozzle. In the formula, λ i The weight of the projection image region corresponding to the i-th ink droplet is used to calculate the flight speed of the ejected ink droplets in the nozzle. The value is determined according to the position of the projection image region corresponding to the i-th ink droplet in the ejected ink droplet image of the nozzle. The weight is greater when the region is located in the middle of the ejected ink droplet image of the nozzle than when it is located at the edge of the ejected ink droplet image of the nozzle.

8. The method for detecting ink droplets ejected from a printhead array according to claim 4, characterized in that, Each nozzle's corresponding image contains two projected images of the ink droplet, one above the other. Therefore, based on the image of the ink droplet ejected from each nozzle, the flight angle of the ejected ink droplet for that nozzle is calculated as follows: Based on the centroid coordinates (x1, z1) and (x2, z2) of the projected images of the same ink droplet at two consecutive moments within a single camera frame, the flight angle γ of the flying ink droplet in the camera's observation direction is calculated using the following expression:

9. A method for detecting ink droplets ejected from a printhead array according to claim 5, characterized in that, If each nozzle contains W projected images of the ejected droplets, the method for calculating the flight angle of the ejected droplets from each nozzle based on the ejected droplet images is as follows: In the image of the ejected ink droplets corresponding to the nozzle, the ink droplet contour lines are segmented from W projected images, and the coordinates in the camera image system R are determined based on the corresponding ink droplet contour lines. Cam The centroid coordinates of the following points are (x1, z1), (x2, z2), ..., (x w ,z w Based on W centroid coordinates, a linear regression is used to fit a linear function f(x) = kz + b. The following expression is used to measure the flight angle γ of the flying ink droplet in the camera observation direction: γ = tan -1 k.

10. A printhead array ink droplet detection system, characterized in that, For performing a method for detecting ink droplets ejected from a printhead array as described in any one of claims 1 to 9, the detection system includes: a control module, a data processing module, a vision detection module, and a mechanical auxiliary module; the vision detection module includes multiple vision systems arranged in an array, the spacing between two adjacent vision systems being an integer multiple of the spacing between adjacent printheads within the printhead array; the mechanical auxiliary module includes a motion frame and multiple position fine-tuning mechanisms configured on the motion frame for each vision system; The control module is used to control the movement of the motion frame and / or the printhead array to be detected according to the coordinates of the multiple nozzles to be detected synchronously, and to control the position fine-tuning structure of each position to fine-tune the position of its corresponding vision system so that each vision system is located at the image acquisition position of the ejected ink droplets of its nozzle to be detected, and to control the multiple vision systems to synchronously acquire images of ejected ink droplets from the nozzles of multiple printheads; the data processing module is used to calculate the flight state of ejected ink droplets of each nozzle based on the ejected ink droplet image corresponding to each nozzle. The multiple vision systems are mounted on the motion frame. The coordinates of each nozzle are determined as follows: the angle between a certain axis vector of each nozzle's Cartesian coordinate system and the corresponding axis vector of the global coordinate system of the motion frame is obtained as the nozzle offset angle of that nozzle; the coordinate position of each nozzle in the corresponding nozzle's Cartesian coordinate system is multiplied by the corresponding nozzle offset angle, and the coordinates of the origin of the corresponding nozzle's Cartesian coordinate system in the global coordinate system are added to obtain the coordinates of each nozzle in the global coordinate system. The two coordinate axes of the global coordinate system correspond to the same two mutually perpendicular degrees of freedom directions of the motion frame and / or the nozzle array.

Images