Ultraviolet printer and printing method based on artificial intelligence visual recognition

Abstract

The present application provides an ultraviolet printer and printing method based on artificial intelligence visual recognition. First, edge color difference is performed on the printing sub-image to obtain the edge ring shift recognition entropy of the printing sub-image; edge color difference is performed on the heat recognition sub-image to obtain the edge evaporation recognition vector of the heat recognition sub-image; joint distribution detection is performed on the edge evaporation recognition vector according to the edge ring shift recognition entropy to obtain the edge detection credibility of the printing pattern; scale edge segmentation is performed on the target printing pattern according to the edge detection credibility to obtain multiple target printing sub-images, and the printing rate is adjusted according to the marginalization level corresponding to each target printing sub-image. The present application adjusts the printing rate by the marginalization level corresponding to each target sub-image, thereby increasing the evaporation time of the ink solvent during the printing process, avoiding the thin film ring shift of the ink due to solvent evaporation during the printing process, and improving the clarity of the printed pattern of the ultraviolet printer.

Description

Technical Field

[0001] The present application relates to the field of printer technology, and more specifically, to an ultraviolet printer and a printing method based on artificial intelligence visual recognition. Background Art

[0002] Ultraviolet printers (UV, Ultraviolet) use ultraviolet light to cure ink or coating, so that it can quickly dry and cure during the printing process to achieve high-speed, high-quality printing effects. It is mainly suitable for printing non-absorbent materials, such as laser cardboard, aluminized paper and plastic and other non-absorbent packaging printing.

[0003] In the prior art, when a UV printer is used for printing, after the ink droplets fall onto the substrate, they will form a fixed contact angle on the substrate due to the combined effect of their own surface tension and the surface tension of the substrate. After the solvent continues to evaporate, the solute will precipitate and deposit on the solid surface. If the inward Marangoni flow in this process is less than the outward capillary flow imbalance, the deposited ink will move in a circle during the printing process to form a phenomenon of thick edges and thin middle (also called the coffee ring effect), which greatly affects the quality of the printed pattern of the UV printer and restricts the development of UV printing technology. Summary of the invention

[0004] The present application provides an ultraviolet printer and a printing method based on artificial intelligence visual recognition, so as to solve the technical problem that the clarity of the printed pattern is low due to the film shift of the ink during printing by the ultraviolet printer.

[0005] In a first aspect, the present application provides a printing method based on artificial intelligence visual recognition, which can be executed by a network device, or can also be executed by a chip configured in the network device, and the present application does not limit this.

[0006] Specifically, the method includes:

[0007] During the pattern printing process, a printed image is collected, and artificial intelligence edge detection is performed on the printed image to obtain a printed sub-image;

[0008] Collecting a heat recognition image, performing artificial intelligence edge detection on the heat recognition image, and obtaining a heat recognition sub-image;

[0009] Extracting the ring shift feature of the printed sub-image to obtain the edge ring shift recognition entropy of the printed sub-image;

[0010] Performing edge color difference on the heat recognition sub-graph to obtain an edge evaporation recognition vector of the heat recognition sub-graph;

[0011] Performing joint distribution detection on the edge evaporation recognition vector according to the edge ring shift recognition entropy to obtain edge detection credibility of the printed pattern;

[0012] When the edge detection credibility is higher than the credibility threshold, a target printing pattern is obtained, and scale edge segmentation is performed on the target printing pattern according to the edge detection credibility to obtain multiple target printing sub-images, and the marginalization level corresponding to each target printing sub-image is determined, and the printing rate is adjusted according to the marginalization level corresponding to each target printing sub-image.

[0013] In combination with the first aspect, in some implementations of the first aspect, performing ring shift feature extraction on the printed sub-image to obtain edge ring shift recognition entropy of the printed sub-image specifically includes:

[0014] Obtaining all edge feature points of the printed sub-image;

[0015] For each edge feature point of the printed sub-image, determining an edge ring shift recognition coefficient corresponding to the edge feature point according to the grayscale values ​​of non-edge feature points in its neighborhood;

[0016] The above steps are used to determine the edge ring shift recognition coefficients corresponding to all edge feature points, and the edge ring shift recognition entropy of the printed sub-image is determined according to the edge ring shift recognition coefficients corresponding to all edge feature points.

[0017] In combination with the first aspect, in some implementations of the first aspect, performing edge color difference on the heat recognition subgraph to obtain an edge evaporation recognition vector of the heat recognition subgraph specifically includes:

[0018] Obtain all edge feature points of the heat recognition subgraph;

[0019] For each edge feature point of the heat recognition subgraph, feature screening is performed according to the grayscale values ​​of non-edge feature points in its neighborhood to obtain multiple evaporation mark points, and the edge evaporation recognition coefficient corresponding to the edge feature point is determined according to the grayscale value of each evaporation mark point;

[0020] The above steps are used to determine the edge evaporation recognition coefficients corresponding to all edge feature points, and the edge evaporation recognition vector of the heat recognition subgraph is determined according to the edge evaporation recognition coefficients corresponding to all edge feature points.

[0021] In conjunction with the first aspect, in certain implementations of the first aspect, determining the marginalization level corresponding to each target printing sub-graph specifically includes:

[0022] For any target printing subgraph, determine all connected areas in the target printing subgraph;

[0023] Determine the connected areas and edge ring shifts corresponding to each connected region, perform weighted fusion on the connected areas corresponding to each connected region according to the edge ring shifts corresponding to each connected region, and obtain the marginalization level corresponding to the target printing subgraph;

[0024] The above method is used to determine the marginalization level corresponding to each target printing sub-graph.

[0025] In conjunction with the first aspect, in certain implementations of the first aspect, adjusting the printing rate according to the marginalization levels corresponding to the target printing sub-graphs specifically includes:

[0026] Obtain the marginalization level corresponding to each target printing subgraph;

[0027] Get the current printing position;

[0028] The target printing sub-image to which the current printing position belongs is determined as a characteristic target printing sub-image, and the printing rate is adjusted according to the marginalization level of the characteristic target printing sub-image.

[0029] In combination with the first aspect, in some implementations of the first aspect, when the edge detection credibility is lower than a credibility threshold, a preset standard printing rate is acquired, and printing is performed according to the standard printing rate.

[0030] In combination with the first aspect, in certain implementations of the first aspect, an infrared camera is used to capture the thermal recognition image.

[0031] In a second aspect, the present application provides an ultraviolet printer based on artificial intelligence visual recognition, which includes a printing control unit, and the printing control unit includes:

[0032] The acquisition module is used to acquire a printed image during the pattern printing process, perform artificial intelligence edge detection on the printed image to obtain a printed sub-image; perform ring shift feature extraction on the printed sub-image to obtain edge ring shift recognition entropy of the printed sub-image;

[0033] The processing module collects a heat recognition image, performs artificial intelligence edge detection on the heat recognition image, and obtains a heat recognition sub-image; performs edge color difference on the heat recognition sub-image, and obtains an edge evaporation recognition vector of the heat recognition sub-image;

[0034] The processing module is further used to perform joint distribution detection on the edge evaporation recognition vector according to the edge ring shift recognition entropy to obtain edge detection credibility of the printed pattern;

[0035] The execution module is used to obtain a target printing pattern when the edge detection credibility is higher than a credibility threshold, perform scale edge segmentation on the target printing pattern according to the edge detection credibility to obtain multiple target printing sub-images, determine the marginalization level corresponding to each target printing sub-image, and adjust the printing rate according to the marginalization level corresponding to each target printing sub-image.

[0036] In a third aspect, the present application provides a computer terminal device, comprising a memory and a processor, wherein the memory stores a code, and the processor is configured to obtain the code and execute the above-mentioned printing method based on artificial intelligence visual recognition.

[0037] In a fourth aspect, the present application provides a computer-readable storage medium, which stores at least one computer program, and the computer program is loaded and executed by a processor to implement the operations performed by the above-mentioned printing method based on artificial intelligence visual recognition.

[0038] The technical solution provided by the embodiments disclosed in this application has the following beneficial effects:

[0039] The present application provides an ultraviolet printer and printing method based on artificial intelligence visual recognition. First, during the pattern printing process, a printing image is collected, and artificial intelligence edge detection is performed on the printing image to obtain a printing sub-image; a heat recognition image is collected, and artificial intelligence edge detection is performed on the heat recognition image to obtain a heat recognition sub-image; edge color difference is performed on the printing sub-image to obtain edge ring shift recognition entropy of the printing sub-image; edge color difference is performed on the heat recognition sub-image to obtain edge evaporation recognition vector of the heat recognition sub-image; joint distribution detection is performed on the edge evaporation recognition vector according to the edge ring shift recognition entropy to obtain edge detection credibility of the printing pattern; when the edge detection credibility is higher than the credibility threshold, a target printing pattern is obtained, and scale edge segmentation is performed on the target printing pattern according to the edge detection credibility to obtain multiple target printing sub-images, and the marginalization level corresponding to each target printing sub-image is determined. The printing rate is adjusted according to the marginalization level corresponding to each target printing sub-image. The present application adjusts the printing rate according to the marginalization level corresponding to each target sub-image, thereby improving the evaporation time of the ink solvent during the printing process, avoiding the film ring shift of the ink due to solvent evaporation during the printing process, and improving the clarity of the printed pattern of the ultraviolet printer. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 is an exemplary flow chart of a printing method based on artificial intelligence visual recognition according to some embodiments of the present application;

[0041] Figure 2An exemplary flow chart of determining edge ring shift recognition entropy of a printed sub-image in some embodiments of the present application;

[0042] Figure 3 is a schematic diagram of exemplary hardware and / or software of a printing control unit according to some embodiments of the present application;

[0043] Figure 4 It is a structural schematic diagram of a computer terminal device that implements a printing method based on artificial intelligence visual recognition according to some embodiments of the present application. DETAILED DESCRIPTION

[0044] The present application acquires a printing image during the pattern printing process, performs artificial intelligence edge detection on the printing image, and obtains a printing sub-image; acquires a heat recognition image, performs artificial intelligence edge detection on the heat recognition image, and obtains a heat recognition sub-image; performs edge color difference on the printing sub-image, and obtains the edge ring shift recognition entropy of the printing sub-image; performs edge color difference on the heat recognition sub-image, and obtains the edge evaporation recognition vector of the heat recognition sub-image; performs joint distribution detection on the edge evaporation recognition vector according to the edge ring shift recognition entropy, and obtains the edge detection credibility of the printing pattern; when the edge detection credibility is higher than the credibility threshold, acquires the target printing pattern, performs scale edge segmentation on the target printing pattern according to the edge detection credibility, obtains multiple target printing sub-images, determines the marginalization level corresponding to each target printing sub-image, and adjusts the printing rate according to the marginalization level corresponding to each target printing sub-image. The present application adjusts the printing rate according to the marginalization level corresponding to each target sub-image, thereby improving the evaporation time of the ink solvent during the printing process, avoiding the film ring shift of the ink due to solvent evaporation during the printing process, and improving the clarity of the printed pattern of the ultraviolet printer.

[0045] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods. Figure 1 , which is an exemplary flow chart of a printing method based on artificial intelligence visual recognition according to some embodiments of the present application. The printing method 100 based on artificial intelligence visual recognition mainly includes the following steps:

[0046] In step S101, a print image is collected during the pattern printing process, and artificial intelligence edge detection is performed on the print image to obtain a print sub-image; ring shift feature extraction is performed on the print sub-image to obtain edge ring shift recognition entropy of the print sub-image.

[0047] It should be noted that the printed image is a printed pattern image collected on the printing plane after the ultraviolet printer completes the pattern printing, and the printed sub-image is an image of a partial area that only contains valid image information in the printed image, obtained after artificial intelligence edge detection is performed on the printed image.

[0048] Optionally, in some embodiments, a high-definition camera may be used to capture the printed image. In some other embodiments, other devices or equipment capable of image capture may also be used, which is not limited in this application.

[0049] Optionally, in some embodiments, a sobel edge detection operator may be used to perform edge detection on the printed image to obtain a printed sub-image. In some other embodiments, other algorithms or devices capable of implementing edge detection may also be used, which is not limited here.

[0050] Optionally, in some embodiments, reference Figure 2 As shown, this figure is an exemplary flow chart of determining the edge ring shift recognition entropy in some embodiments of the present application, performing ring shift feature extraction on the printed sub-image to obtain the edge ring shift recognition entropy of the printed sub-image specifically includes:

[0051] In step S1011, all edge feature points of the printed sub-image are obtained;

[0052] In step S1012, for each edge feature point of the printed sub-image, an edge ring shift recognition coefficient corresponding to the edge feature point is determined according to the grayscale values ​​of non-edge feature points in its neighborhood;

[0053] In step S1013, the above steps are used to determine the edge ring shift recognition coefficients corresponding to all edge feature points, and the edge ring shift recognition entropy of the printed sub-image is determined according to the edge ring shift recognition coefficients corresponding to all edge feature points.

[0054] It should be noted that the edge ring shift recognition entropy is a quantitative value of the degree of film ring shift due to ink deposition in the printed sub-image, and the edge feature point is the edge point of the printed sub-image. The edge is an area where the grayscale value (or color) changes significantly, reflecting the boundary between different objects or different areas in the image. Images near the edge points are more prone to film ring shift during printing, resulting in lighter or darker colors of the printed images at that location, affecting the printing effect of the ultraviolet printer. Therefore, the edge ring shift recognition entropy of the printed sub-image can be extracted based on the grayscale values ​​of the pixels near the edge feature points of the printed sub-image.

[0055] In specific implementation, the standard deviation of the grayscale values ​​of non-edge feature points in the neighborhood corresponding to each edge feature point can be used as the edge ring shift recognition coefficient corresponding to the edge feature point, and the average value of the edge ring shift recognition coefficients corresponding to all edge feature points in the printed sub-image can be used as the edge ring shift recognition entropy of the printed sub-image.

[0056] In step S102, a heat recognition image is collected, and artificial intelligence edge detection is performed on the heat recognition image to obtain a heat recognition sub-image; edge color difference is performed on the heat recognition sub-image to obtain an edge evaporation recognition vector of the heat recognition sub-image.

[0057] It should be noted that the thermal recognition image is an infrared image of the printed pattern collected by the ultraviolet printer on the printing plane after the pattern is printed, which is used to perform thermal recognition on the printed pattern. The thermal recognition sub-image is obtained after performing artificial intelligence edge detection on the thermal recognition image, and only contains a partial area image of the valid image information in the thermal recognition image.

[0058] Optionally, in some embodiments, an infrared camera may be used to capture the thermal recognition image. In some other embodiments, other devices or equipment capable of image capture may also be used, which is not limited in this application.

[0059] Optionally, in some embodiments, artificial intelligence edge detection is performed on the heat recognition image to obtain a heat recognition sub-image in the following manner:

[0060] Grayscale processing is performed on the heat recognition image to obtain a heat recognition grayscale image;

[0061] Artificial intelligence edge detection is performed on the heat recognition grayscale image to obtain a heat recognition sub-image.

[0062] It should be noted that grayscale processing is the process of converting a color image into a grayscale image. For each pixel, the values ​​of its red, green, and blue (RGB) colors are converted into a single grayscale value. Common grayscale processing methods include the average method, the weighted average method, etc. In specific implementation, the artificial intelligence edge detection of the heat recognition grayscale image can utilize a deep learning model, such as a convolutional neural network. In some embodiments, edge detection models including HED (Holistically-Nested Edge Detection) and Canny Edge Detection can be used to perform artificial intelligence edge detection on the heat recognition grayscale image, and then the heat recognition sub-graph can be extracted through image segmentation technology. For example, according to the application scenario, simple threshold segmentation, K-means clustering, and other methods commonly used in image style segmentation can be used.

[0063] It should be noted that the edge evaporation recognition vector is a quantitative value of the difference between the evaporation degree of the ink solvent in the edge area and the evaporation degree of the ink solvent in the non-edge area in the thermal recognition sub-image, and is determined according to the grayscale value of the pixel points in the edge area of ​​the thermal recognition sub-image. Specifically, the different evaporation rates of the ink solvent in different areas during the printing process will cause the film to shift, and the evaporation of the solvent will cause the temperature to drop in a small area. The grayscale value of the pixel points in the edge area of ​​the thermal recognition sub-image of the present application reflects the temperature detail information of the area. Therefore, the edge evaporation recognition vector of the thermal recognition sub-image can be determined according to the grayscale value of the pixel points in the edge area of ​​the thermal recognition sub-image.

[0064] Optionally, in some embodiments, edge color difference is performed on the heat recognition sub-graph to obtain the edge evaporation recognition vector of the heat recognition sub-graph in the following manner:

[0065] Obtain all edge feature points of the heat recognition subgraph;

[0066] For each edge feature point of the heat recognition subgraph, feature screening is performed according to the grayscale values ​​of non-edge feature points in its neighborhood to obtain multiple evaporation mark points, and the edge evaporation recognition coefficient corresponding to the edge feature point is determined according to the grayscale value of each evaporation mark point;

[0067] The above steps are used to determine the edge evaporation recognition coefficients corresponding to all edge feature points, and the edge evaporation recognition vector of the heat recognition subgraph is determined according to the edge evaporation recognition coefficients corresponding to all edge feature points.

[0068] In a specific implementation, in the process of determining the edge evaporation recognition coefficient corresponding to the edge feature point, for any edge feature point, all pixel points whose grayscale values ​​among non-edge feature points in its neighborhood are higher than that of the edge feature point are used as evaporation mark points, and then the pixel differences between each evaporation mark point and the edge feature point are obtained, and the standard deviation of all pixel differences is used as the edge evaporation recognition coefficient of the edge feature point, and then the edge evaporation recognition coefficients corresponding to all edge feature points can be combined into an edge evaporation recognition vector of the heat recognition subgraph according to a time sequence.

[0069] In step S103, a joint distribution detection is performed on the edge evaporation recognition vector according to the edge ring shift recognition entropy to obtain the edge detection reliability of the printed pattern.

[0070] It should be noted that in the present application, the edge ring shift recognition entropy reflects the overall color ring shift degree of the printed image, and the edge evaporation recognition vector reflects the abnormal evaporation degree of the solvent in different edge areas. Since the color ring shift of the printed image is caused by the rapid evaporation of the solvent in the edge area, when there is actually a film ring shift phenomenon in the printing process, there is a certain degree of similarity between the data distributions of the two. Therefore, the edge detection credibility of the printed pattern can be determined based on the degree of similarity between the edge ring shift recognition entropy and the edge evaporation recognition vector. The edge detection credibility reflects the possible degree of film ring shift of the ultraviolet printer during the image printing process.

[0071] Optionally, in some embodiments, a long short-term memory neural network is used to train the correlation between the edge ring shift recognition entropy and the edge evaporation recognition vector to obtain the edge detection credibility of the printed pattern.

[0072] Preferably, in some embodiments, the process of performing joint distribution detection on the edge evaporation recognition vector according to the edge shift recognition entropy to obtain the edge detection credibility of the printed pattern may specifically include: normalizing the edge shift recognition entropy and the edge evaporation recognition vector, extracting vector elements in the edge evaporation recognition vector and the edge shift recognition entropy to form a credibility detection sequence, and using the autocorrelation coefficient of the credibility monitoring sequence as the edge detection credibility.

[0073] In step S104, when the edge detection credibility is higher than the credibility threshold, the target printing pattern is obtained, and the target printing pattern is scaled and segmented according to the edge detection credibility to obtain multiple target printing sub-images, and the marginalization level corresponding to each target printing sub-image is determined, and the printing rate is adjusted according to the marginalization level corresponding to each target printing sub-image.

[0074] Optionally, in some embodiments, when the edge detection credibility is lower than a credibility threshold, a preset standard printing rate is acquired, and printing is performed according to the standard printing rate.

[0075] Optionally, in some embodiments, the credibility threshold is calibrated as a constant based on experience. In some embodiments, the credibility of edge detection when printing with an ultraviolet printer under a standard working environment can also be used as the credibility threshold, which is not limited here.

[0076] It should be noted that the target printing pattern is an image stored in the UV printer that needs to be printed in the target area. Optionally, in some embodiments, the target printing pattern is scaled and edge segmented according to the edge detection credibility to obtain multiple target printing sub-images, which specifically includes: obtaining a preset printing direction, and evenly segmenting the target printing image according to the preset printing direction and the edge detection credibility to obtain multiple target printing sub-images.

[0077] It should be noted that a threshold interval can be preset, and the number of divisions corresponding to the edge detection credibility can be determined according to the threshold interval corresponding to the edge detection credibility, and the target printing image is evenly divided according to the number of divisions. For example, when the number of divisions is 5, the target printing image is evenly divided into 5 target printing sub-images in the preset printing direction.

[0078] Optionally, in some embodiments, the marginalization level corresponding to each target printing sub-graph is determined in the following manner:

[0079] For any target printing subgraph, determine all connected areas in the target printing subgraph;

[0080] Determine the connected areas and edge ring shifts corresponding to each connected region, perform weighted fusion on the connected areas corresponding to each connected region according to the edge ring shifts corresponding to each connected region, and obtain the marginalization level corresponding to the target printing subgraph;

[0081] The above method is used to determine the marginalization level corresponding to each target printing sub-graph.

[0082] In specific implementation, the number of pixels in a connected region can be used as the connected area of ​​the connected region, and the edge ring shift of the connected region is the average of the proportion of pixels in the connected region in the neighborhood of all edge points of the connected region. Then, the connected areas corresponding to each connected region are weightedly fused according to the edge ring shift of each connected region to obtain a fusion feature, and then the threshold interval corresponding to the fusion feature is determined to determine the marginalization level corresponding to the target print sub-image. In some embodiments, the threshold interval of the fusion feature is preset based on experience and will not be repeated here.

[0083] Preferably, in some embodiments, adjusting the printing rate according to the marginalization levels corresponding to the target printing sub-graphs specifically includes:

[0084] Obtain the marginalization level corresponding to each target printing subgraph;

[0085] Get the current printing position;

[0086] The target printing sub-image to which the current printing position belongs is determined as a characteristic target printing sub-image, and the printing rate is adjusted according to the marginalization level of the characteristic target printing sub-image.

[0087] Optionally, in some embodiments, the target printing sub-image has five marginalization levels, namely A, B, C, D and E, and each marginalization level has a corresponding printing rate. Therefore, when the marginalization degree is large, the printing rate of the ultraviolet printer is reduced, thereby reducing the ink deposition rate of the edge part of the printed pattern during the printing process, thereby avoiding the technical problem of low clarity of the printed pattern caused by the thin film ring shift of the ink deposition when the external printer is printing.

[0088] Preferably, in some embodiments, adjusting the printing rate according to the marginalization levels corresponding to the target printing sub-graphs specifically includes:

[0089] Obtain the marginalization level corresponding to each target printing subgraph;

[0090] Get the current printing position;

[0091] The target printing sub-image to which the current printing position belongs is determined as a characteristic target printing sub-image, and the printing rate is adjusted according to the marginalization level of the characteristic target printing sub-image and the edge ring shift recognition entropy.

[0092] In specific implementation, a proportional-integral-derivative (PID) controller can be used to adjust the printing rate. For example, the difference between the edge ring shift recognition entropy and the preset input is used as the PID controller input, and the corresponding controller open-loop gain is determined according to the marginalization level, and then the printing rate is adjusted by the PID controller output control voltage.

[0093] In addition, in another aspect of the present application, in some embodiments, the present application provides an ultraviolet printer based on artificial intelligence visual recognition, the printer includes a printing control unit, reference Figure 3 , which is a schematic diagram of exemplary hardware and / or software of a printing control unit according to some embodiments of the present application, the printing control unit 200 includes: a collection module 201, a processing module 202 and an execution module 203, which are described as follows:

[0094] In some specific embodiments of the present application, the acquisition module 201 is used to acquire a printed image during pattern printing, perform artificial intelligence edge detection on the printed image to obtain a printed sub-image; perform ring shift feature extraction on the printed sub-image to obtain edge ring shift recognition entropy of the printed sub-image;

[0095] Processing module 202, in some specific embodiments of the present application, the processing module 202 is used to collect a heat recognition image, perform artificial intelligence edge detection on the heat recognition image, and obtain a heat recognition sub-graph; perform edge color difference on the heat recognition sub-graph, and obtain an edge evaporation recognition vector of the heat recognition sub-graph;

[0096] In some embodiments, the processing module 202 is further used to perform a joint distribution detection on the edge evaporation recognition vector according to the edge ring shift recognition entropy to obtain the edge detection credibility of the printed pattern;

[0097] Execution module 203, in some specific embodiments of the present application, the execution module 203 is used to obtain the target printing pattern when the edge detection credibility is higher than the credibility threshold, perform scale edge segmentation on the target printing pattern according to the edge detection credibility, obtain multiple target printing sub-images, determine the marginalization level corresponding to each target printing sub-image, and adjust the printing rate according to the marginalization level corresponding to each target printing sub-image. The above describes in detail an example of an ultraviolet printer and a printing method based on artificial intelligence visual recognition provided by an embodiment of the present application. It can be understood that in order to realize the above functions, the corresponding device includes a hardware structure and / or software module corresponding to each function.

[0098] Those skilled in the art should easily realize that, in combination with the units and algorithm steps of each example described in the embodiments disclosed in this document, the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a certain function in the application is executed in hardware or in a computer software-driven hardware manner depends on the specific application and design constraints of the technical solution. Therefore, professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0099] In addition, the present application also provides a computer terminal device, which includes a memory and a processor, the memory stores a code, and the processor is configured to obtain the code and execute the above-mentioned printing method based on artificial intelligence visual recognition.

[0100] In some embodiments, reference Figure 4 , which is a schematic diagram of the structure of a computer terminal device using a printing method based on artificial intelligence visual recognition according to some embodiments of the present application. The printing method based on artificial intelligence visual recognition in the above embodiment can be Figure 4 The computer terminal device 300 shown in the figure is implemented, and the computer terminal device 300 includes at least one communication bus 301, a communication interface 302, a processor 303 and a memory 304.

[0101] The processor 303 may be a general-purpose central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more for controlling the execution of the printing method based on artificial intelligence visual recognition in the present application.

[0102] The communication bus 301 may include a path for transmitting information between the above-mentioned components.

[0103] The memory 304 may be a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, an optical disc storage (including a compressed optical disc, a laser disc, an optical disc, a digital versatile disc, a Blu-ray disc, etc.), a disk or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of an instruction or data structure and can be accessed by a computer, but is not limited thereto. The memory 304 may exist independently and be connected to the processor 303 via the communication bus 301. The memory 304 may also be integrated with the processor 303.

[0104] The memory 304 is used to store the program code for executing the solution of the present application, and the execution is controlled by the processor 303. The processor 303 is used to execute the program code stored in the memory 304. The program code may include one or more software modules. The determination of the edge evaporation identification vector in the above embodiment can be implemented by the processor 303 and one or more software modules in the program code in the memory 304.

[0105] The communication interface 302 uses any transceiver or other device for communicating with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area networks (WLAN), etc.

[0106] Optionally, the computer terminal device 300 may further include a power supply 305 for providing power to various devices or circuits in the real-time computer terminal device.

[0107] In a specific implementation, as an embodiment, a computer terminal device may include multiple processors, each of which may be a single-CPU processor or a multi-CPU processor. The processor here may refer to one or more devices, circuits, and / or processing cores for processing data (e.g., computer program instructions).

[0108] The computer terminal device may be a general-purpose computer terminal device or a dedicated computer terminal device. In a specific implementation, the computer terminal device may be a desktop computer, a portable computer, a network server, a personal digital assistant (PDA), a mobile phone, a tablet computer, a wireless terminal device, a communication device or an embedded device. The embodiment of the present application does not limit the type of computer terminal device.

[0109] In addition, in other aspects of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores at least one computer program, and the computer program is loaded and executed by a processor to implement the operations performed by the above-mentioned printing method based on artificial intelligence visual recognition.

[0110] In summary, in an ultraviolet printer and printing method based on artificial intelligence visual recognition disclosed in an embodiment of the present application, firstly, a printing image is collected during the pattern printing process, and artificial intelligence edge detection is performed on the printing image to obtain a printing sub-image; a heat recognition image is collected, and artificial intelligence edge detection is performed on the heat recognition image to obtain a heat recognition sub-image; edge color difference is performed on the printing sub-image to obtain the edge ring shift recognition entropy of the printing sub-image; edge color difference is performed on the heat recognition sub-image to obtain the edge evaporation recognition vector of the heat recognition sub-image; joint distribution detection is performed on the edge evaporation recognition vector according to the edge ring shift recognition entropy to obtain the printing sub-image. The edge detection credibility of the pattern; when the edge detection credibility is higher than the credibility threshold, the target printing pattern is obtained, and the target printing pattern is scaled and segmented according to the edge detection credibility to obtain multiple target printing sub-images, and the marginalization level corresponding to each target printing sub-image is determined. The printing rate is adjusted according to the marginalization level corresponding to each target printing sub-image. The present application adjusts the printing rate by the marginalization level corresponding to each target sub-image, thereby increasing the evaporation time of the ink solvent during the printing process, avoiding the film ring shift of the ink due to solvent evaporation during the printing process, and improving the clarity of the printed pattern of the ultraviolet printer.

[0111] The above is only an embodiment of the present application, and the common knowledge such as the specific technical scheme or characteristics in the scheme is not described in detail here. It should be pointed out that for those skilled in the art, without departing from the technical scheme of the present application, several modifications and improvements can be made, which should also be regarded as the scope of protection of the present application, and these will not affect the effect of the implementation of the present application and the practicality of the patent.

[0112] The scope of protection claimed by this application shall be based on the content of its claims. The specific implementation methods and other records in the specification can be used to interpret the content of the claims. Obviously, those skilled in the art can make various changes and modifications to this application without departing from the spirit and scope of this application. Thus, if these modifications and variations of this application fall within the scope of the claims of this application and their equivalent technologies, this application is also intended to include these modifications and variations.

Claims

1. A printing method based on artificial intelligence visual recognition, characterized in that: include: During the pattern printing process, a print image is collected, artificial intelligence edge detection is performed on the print image to obtain a print sub-image, and ring shift feature extraction is performed on the print sub-image to obtain edge ring shift recognition entropy of the print sub-image; Collect a heat recognition image, perform artificial intelligence edge detection on the heat recognition image to obtain a heat recognition sub-image, perform edge color difference on the heat recognition sub-image to obtain an edge evaporation recognition vector of the heat recognition sub-image; Performing joint distribution detection on the edge evaporation recognition vector according to the edge ring shift recognition entropy to obtain edge detection credibility of the printed pattern; When the edge detection credibility is higher than the credibility threshold, a target printing pattern is acquired, scale edge segmentation is performed on the target printing pattern according to the edge detection credibility to obtain a plurality of target printing sub-graphs, a marginalization level corresponding to each target printing sub-graph is determined, and a printing rate is adjusted according to the marginalization level corresponding to each target printing sub-graph; The step of extracting the ring shift feature of the printed sub-image to obtain the edge ring shift recognition entropy of the printed sub-image specifically includes: Obtaining all edge feature points of the printed sub-image; For each edge feature point of the printed sub-image, determining an edge ring shift recognition coefficient corresponding to the edge feature point according to the grayscale values ​​of non-edge feature points in its neighborhood; The above steps are used to determine the edge ring shift recognition coefficients corresponding to all edge feature points, and the edge ring shift recognition entropy of the printed sub-image is determined according to the edge ring shift recognition coefficients corresponding to all edge feature points; The step of performing edge color difference on the heat recognition sub-graph to obtain the edge evaporation recognition vector of the heat recognition sub-graph specifically includes: Obtain all edge feature points of the heat recognition subgraph; For each edge feature point of the heat recognition subgraph, feature screening is performed according to the grayscale values ​​of non-edge feature points in its neighborhood to obtain multiple evaporation mark points, and the edge evaporation recognition coefficient corresponding to the edge feature point is determined according to the grayscale value of each evaporation mark point; The above steps are used to determine the edge evaporation recognition coefficients corresponding to all edge feature points, and the edge evaporation recognition vector of the heat recognition subgraph is determined according to the edge evaporation recognition coefficients corresponding to all edge feature points; Wherein, determining the marginalization level corresponding to each target printing sub-graph specifically includes: For any target printing subgraph, determine all connected areas in the target printing subgraph; Determine the connected areas and edge ring shifts corresponding to each connected region, perform weighted fusion on the connected areas corresponding to each connected region according to the edge ring shifts corresponding to each connected region, and obtain the marginalization level corresponding to the target printing subgraph; Determine the marginalization level corresponding to each target printing sub-graph respectively by the above method; The step of adjusting the printing rate according to the marginalization levels corresponding to the target printing sub-graphs specifically includes: Obtain the marginalization level corresponding to each target printing subgraph; Get the current printing position; The target printing sub-image to which the current printing position belongs is determined as a characteristic target printing sub-image, and the printing rate is adjusted according to the marginalization level of the characteristic target printing sub-image.

2. The method according to claim 1, characterized in that When the edge detection credibility is lower than the credibility threshold, a preset standard printing rate is acquired, and printing is performed according to the standard printing rate.

3. The method according to claim 1, characterized in that An infrared camera is used to collect the thermal recognition image.

4. An ultraviolet printer based on artificial intelligence visual recognition, which uses the method described in any one of claims 1 to 3 to perform printing based on artificial intelligence visual recognition, and the ultraviolet printer based on artificial intelligence visual recognition includes a printing control unit, characterized in that: The printing control unit comprises: The acquisition module is used to acquire a printed image during the pattern printing process, perform artificial intelligence edge detection on the printed image to obtain a printed sub-image; perform ring shift feature extraction on the printed sub-image to obtain edge ring shift recognition entropy of the printed sub-image; The processing module collects a heat recognition image, performs artificial intelligence edge detection on the heat recognition image, and obtains a heat recognition sub-image; performs edge color difference on the heat recognition sub-image, and obtains an edge evaporation recognition vector of the heat recognition sub-image; The processing module is further used to perform joint distribution detection on the edge evaporation recognition vector according to the edge ring shift recognition entropy to obtain edge detection credibility of the printed pattern; The execution module is used to obtain a target printing pattern when the edge detection credibility is higher than a credibility threshold, perform scale edge segmentation on the target printing pattern according to the edge detection credibility to obtain multiple target printing sub-images, determine the marginalization level corresponding to each target printing sub-image, and adjust the printing rate according to the marginalization level corresponding to each target printing sub-image.

5. A computer terminal device, characterized in that: The computer terminal device includes a memory and a processor, the memory stores a code, and the processor is configured to obtain the code and execute a printing method based on artificial intelligence visual recognition as described in any one of claims 1 to 3.

6. A computer-readable storage medium storing at least one computer program, characterized in that: The computer program is loaded and executed by a processor to implement the operations performed by a printing method based on artificial intelligence visual recognition as described in any one of claims 1 to 3.

Images