A scanning and printing system based on artificial intelligence

Abstract

The present application belongs to the technical field of data processing, and specifically relates to a scanning and printing system based on artificial intelligence, which comprises a scanning and printing device for printing files and scanning paper, and an artificial intelligence module for learning and managing various processes of the scanning and printing device, wherein a sound monitoring device is added in the scanning and printing device. The present application can perform layered processing on each multi-page file through the cooperation of a paper pushing module, an artificial intelligence module and a visual module. For example, the first group of files is moved 10 mm to the left through the paper pushing module, while the second group of files is moved 5 mm to the left through the paper pushing module, so that a 5-mm gap exists between the two groups of files, facilitating the separation of the two groups of files. The visual module can determine the poor printing effect of a single paper in the same file, and the paper can be directly pushed away from the file stacking range through the paper pushing module, and the page can be reprinted through the artificial intelligence module, which is beneficial to solving the situation that multiple files need to be manually classified after being printed and stacked in sequence.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, and particularly relates to a scanning and printing system based on artificial intelligence. BACKGROUND

[0002] The application of artificial intelligence in the printing system involves multiple aspects. The present application is mainly applied to using a machine learning model to evaluate the image of a printing component, identifying defects or design deviations, and performing early intervention to prevent defects, using artificial intelligence to analyze the operation data of the equipment, performing equipment maintenance in advance to reduce unplanned downtime, ensuring the efficient progress of the production process, and relying on artificial intelligence to complete data processing to further improve the accuracy and stability of the scanning and printing system.

[0003] Scanning and printing equipment is needed for daily office work of enterprises. In order to save costs, a large scanning and printing equipment is often used by multiple departments or even multiple companies. Such scanning and printing equipment may have the following problems in daily use:

[0004] The printed files cannot be taken away in time, and multiple files are stacked on the paper, which is difficult to distinguish;

[0005] No one is present to check the printing effect during printing, and poor printing effect causes the entire printed file to be wasted;

[0006] When printing multiple pages of files, the single paper printing effect is poor, and manual inspection is required to print the page separately, which wastes labor;

[0007] When the scanning and printing equipment has problems affecting the printing effect, no one can discover and stop in time, which not only wastes a large amount of paper, but also causes the problem of the scanning and printing equipment to be aggravated.

[0008] Therefore, it is necessary to provide a scanning and printing system based on artificial intelligence, which is beneficial to solve the problem of manual classification of multiple files printed and stacked in sequence, to eliminate manual inspection of the printing effect, to correct single bad printing paper for multiple page files, to ensure the printing quality of multiple page files, and to self-regulate the printing equipment. SUMMARY

[0009] To solve the above technical problems, the present application provides a scanning and printing system based on artificial intelligence, which solves the above technical problems.

[0010] The scanning and printing system based on artificial intelligence provided by the present application comprises a scanning and printing equipment for printing and scanning of files and paper;

[0011] An artificial intelligence module learns and manages each process of the scanning and printing equipment, wherein sound monitoring is added in the scanning and printing equipment, and vibration monitoring is performed on the moving printed paper.

[0012] Adding a visual module above the paper outlet of the scanning and printing device: visualizing the printed paper by the scanning and printing device, and monitoring the visual picture by the artificial intelligence module;

[0013] Adding a paper pushing module to one side of the paper outlet of the scanning and printing device: moving the single printed paper in the scanning and printing device, or moving the whole group of papers of the same file;

[0014] The artificial intelligence module detects the printed file completed by the scanning and printing device, wherein the printing effect is , the formula variables are: sound data , paper vibration frequency , ink depth , and text centering ;

[0015]

[0016] is a weight coefficient, used to adjust the influence degree of each variable on the printing effect; is a nonlinear index, used to simulate the nonlinear influence of each variable on the printing effect; is a constant term, which can represent the basic printing quality or system error.

[0017] Preferably, a sound detection sensor is added to the scanning and printing device to detect its working sound, and a ball-type vibration sensor is installed on the moving path of the paper in printing to detect the vibration of the moving paper.

[0018] Preferably, if the mechanical sound of the scanning and printing device during printing exceeds the threshold value, the visual module starts to visualize the printed paper, and analyzes the file in the picture by artificial intelligence, including ink depth and printing position, to judge the printing effect.

[0019] Preferably, if the paper vibration of the scanning and printing device during printing exceeds the threshold value, the visual module starts to visualize the printed paper, and analyzes the file in the picture by artificial intelligence, including ink depth and printing position, to judge the printing effect.

[0020] Preferably, any printed paper is judged as failed by the artificial intelligence module, and the paper is pushed out to the other side by the paper pushing module, so as to be out of the predetermined range of retained paper, and the content of the paper is printed again by the artificial intelligence module.

[0021] Preferably, if two groups of files are printed continuously, the first group of files is moved as a whole by the paper pushing module after being completed by the artificial intelligence module.

[0022] Preferably, if the paper moves beyond the threshold value of the ball-type vibration sensor, and the printing has not started, the scanning and printing device is directly controlled by the artificial intelligence module to spit out the paper, and the artificial intelligence module is started to push the paper to the other side to make it out of the predetermined range of paper retention;

[0023] Preferably, the scanning and printing device is built-in with an ink cartridge, the ink cartridge is built-in with a pigment detection module for detecting the amount of printing pigment, the signal of the pigment detection module is transmitted to the artificial intelligence module, and the scanning and printing device is built-in with a paper box, the bottom of the paper box is provided with a paper detection module for detecting the number of papers, and the signal of the paper detection module is transmitted to the artificial intelligence module.

[0024] Preferably, after the file is uploaded to the scanning and printing device, the artificial intelligence module judges whether the file can be printed according to the amount of printing pigment reflected by the pigment detection module and the number of papers reflected by the paper detection module; if yes, the file is printed, otherwise, the file uploading end is reminded to add ink or paper.

[0025] Compared with the related art, the scanning and printing system based on artificial intelligence provided by the present application has the following beneficial effects:

[0026] The present application divides each multi-page file into layers by the paper pushing module, the artificial intelligence module and the visual module, for example, the first group of files is moved 10mm to the left by the paper pushing module, while the second group of files is moved 5mm to the left by the paper pushing module, so that there is a 5mm gap between the two groups of files, which facilitates the separation of the two groups of files, the visual module judges that the printing effect of a single paper in the same file is poor, and the paper pushing module directly pushes it away from the file stacking range, and the artificial intelligence module reprints the page, which is beneficial to solve the problem of manual classification after multiple files are printed and stacked in turn.

[0027] The present application supervises the scanning and printing by the visual module and the artificial intelligence module, completes the supervision of the paper conveying and the ink cartridge pigment in the scanning and printing device, and judges the printing quality according to the combination of the two groups of data and the visual module, which is beneficial to ensure the printing quality of multi-page files and to supervise the printing device. BRIEF DESCRIPTION OF DRAWINGS

[0028] Figure 1 The present application provides a scanning and printing system based on artificial intelligence, and the composition process schematic diagram of the scanning and printing system based on artificial intelligence is shown in the figure;

[0029] Figure 2 The present application provides a scanning and printing system based on artificial intelligence, and the composition process schematic diagram of the scanning and printing system based on artificial intelligence is shown in the figure;

[0030] Figure 3 Figure 6 shows a specific embodiment of a scanning and printing system based on artificial intelligence provided by the present application. DETAILED DESCRIPTION

[0031] The technical solutions of the present application will be described below in detail with reference to the drawings. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative labor fall within the scope of protection of the present application.

[0032] Since the fast and efficient scanning and printing device has a relatively high cost and requires a certain floor area, it is often shared by multiple departments, or even by multiple companies in a part of office buildings, which may result in the situation that the printing personnel are not in front of the device, and the problems mentioned in the background technology may occur. If the printing is performed in front of and behind multiple departments or the printed files are not taken away in time, the files may be mixed and taken away.

[0033] Embodiment 1: A scanning and printing system based on artificial intelligence, the scanning and printing device: for printing and scanning of files and paper;

[0034] The artificial intelligence module: learns and manages each process of the scanning and printing device, wherein a sound monitoring device is added to the scanning and printing device, and a vibration monitoring device is added to the moving paper in the printing process.

[0035] A sound detection sensor is added to the scanning and printing device to detect the working sound, and a ball-type vibration sensor is installed on the moving path of the paper in the printing process to detect the vibration of the moving paper.

[0036] A vision module is added above the paper outlet of the scanning and printing device: the printed paper of the scanning and printing device is visually imaged, and the vision image is monitored by the artificial intelligence module.

[0037] A paper pushing module is added to one side of the paper outlet of the scanning and printing device: the paper moving displacement of the single printed paper in the scanning and printing device is displaced; or the whole displacement of the same group of papers of the same file is displaced.

[0038] The artificial intelligence module detects the printed file of the scanning and printing device, wherein the printing effect is , the formula variables are: sound data , paper vibration frequency , ink depth , and text centering .

[0039]

[0040] are weight coefficients used to adjust the influence of each variable on the printing effect; are non-linear exponents used to simulate the non-linear influence of each variable on the printing effect; is a constant term, which can represent the basic printing quality or system error.

[0041] To directly optimize the artificial intelligence calculation formula for the printer, a more comprehensive and refined mathematical model can be used, taking into account the main effect, interaction effect, and high-order effect of variables. The following formula can be proposed:

[0042]

[0043] : Same as the previous formula, representing the non-linear main effect of each variable.

[0044] to : The quadratic term of each variable, used to simulate the high-order effect of the variable itself.

[0045] to : Interaction term, used to simulate the interaction between different variables.

[0046] to : Three-factor interaction term, used to simulate the complex influence when three variables act together.

[0047] c: Constant term, representing the basic printing quality or other fixed influencing factors.

[0048] These weights and exponents need to be determined through data training. Multiple scanning of the running data of the printing device can be used to form a learning model, or methods such as polynomial regression, support vector machine regression, or neural network can be used to fit the experimental or actual use data, so as to find the best parameter value.

[0049] This optimized formula provides more flexibility and accuracy, and can better simulate the comprehensive influence of various factors on the printing effect in actual situations, especially considering the interaction and high-order effect between variables, as well as more complex multi-variable interaction.

[0050] The combination of artificial intelligence technology and printing systems is increasingly seen as a key way to achieve digital transformation and improve service capabilities. By integrating advanced sensors, automatic control systems, and artificial intelligence technology, intelligent printing equipment can achieve automation, intelligence, and digitization of the printing process. This not only improves production efficiency but also reduces manual intervention while ensuring print quality, enabling high-speed continuous printing and personalized custom printing.

[0051] The algorithm can distinguish the darkness of the ink in the picture by learning the image and artificial intelligence module algorithm.

[0052] To distinguish the darkness of the ink in the picture, first, the picture needs to be converted into digital information that can be recognized and processed by a computer. During this process, image segmentation and feature extraction techniques can be used to identify ink regions in the picture. Then, the algorithm of the artificial intelligence module classifies these regions by darkness.

[0053] Image segmentation is the process of subdividing an image into multiple constituent parts or pixel sets, with the goal of simplifying the image's representation and making it easier to analyze. For the problem of classifying the darkness of ink, semantic segmentation techniques can be used, which is a pixel-by-pixel classification method that can classify each pixel point into different categories. In the ink classification problem, ink of different darkness can be set as different categories, and the trained semantic segmentation model can be used to classify the ink regions pixel by pixel.

[0054] An effective image segmentation algorithm is the encoding-decoding model, such as U-Net. This model includes an encoder (convolutional layer) and a decoder (deconvolutional layer). The encoder is used to extract image features, and the decoder is used to generate a segmentation map from these features. When dealing with ink darkness, the encoder can capture different depth features of the ink region, and the decoder can convert these features into corresponding darkness classification results.

[0055] To train such a model, a large amount of labeled data is needed, i.e., pictures with manually labeled ink darkness. These data will be used to train a convolutional neural network (CNN), such as the aforementioned U-Net. CNN can automatically identify and classify new ink pictures by learning the ink features in the labeled pictures.

[0056] The algorithm can also distinguish whether the text in the paper is skewed or offset by learning the image and artificial intelligence module algorithm. Specific methods include the following:

[0057] Skew detection based on Hough transform

[0058] Principle: Detects straight lines in the image using the Hough transform to determine the orientation of the text lines. This method is suitable for document images that contain clear straight lines, such as text line edges.

[0059] Advantages: The algorithm is relatively simple and does not require complex computational resources.

[0060] Disadvantages: It performs poorly when the text lines are not clear or the straight line features are not obvious.

[0061] Skew angle detection based on minimum enclosing rectangle

[0062] Principle: Uses the minAreaRect function in OpenCV (an open-source computer vision library) to find the smallest enclosing rectangle that covers all the text regions and calculates the rotation angle of the rectangle to obtain the skew angle of the text.

[0063] Automatic correction based on deep learning

[0064] Principle: Uses a deep learning model, such as a convolutional neural network (CNN), to train a model that can automatically identify and correct text skew. This model can learn the features of text with different skew degrees and types from a large amount of labeled data.

[0065] Image correction based on affine transformation

[0066] Principle: Rotates and translates the image through affine transformation to correct the skew and offset of the text.

[0067] In summary, there are multiple methods to determine whether the text in the paper is skewed or offset using algorithms. In practical applications, the most suitable method can be chosen based on specific requirements and conditions.

[0068] Example two, this embodiment mainly solves the situation that in multi-page printing, a page appears printing error, for example, printing overlap, unclear printing, and overall layout skew, etc. This situation is often accompanied by ink cartridge ink flow and paper feeding.

[0069] The file upload end uploads the file to the scanning and printing end through the local area network. If the mechanical sound of the scanning and printing device exceeds the threshold value during printing, the vision module starts to visualize the printed paper, analyzes the file in the picture through artificial intelligence, including ink depth and printing position, to judge the printing effect. The artificial intelligence module combines algorithms to judge the printing effect. If any printed paper is judged as failed by the artificial intelligence module, the paper is pushed out of the predetermined retention paper range by the paper pushing module, and the content of the paper is printed again by the artificial intelligence module.

[0070] In the process of normal operation of the scanning and printing device, the sound waves of its operation are often within a fluctuation range. The sound sensor mainly identifies the sound frequency by detecting the sound wave vibration. The sound wave causes air pressure changes when it propagates in the air. These changes are captured by the sound sensor and converted into electrical signals, which are then processed by the corresponding circuit and algorithm to extract the frequency information.

[0071] Example three, the difference from example two is that if the scanning and printing device is not smooth in paper feeding, it means that the paper feeding structure is abnormal, and the corresponding paper vibration will also appear abnormal situation.

[0072] If the paper vibration of the scanning and printing device exceeds the threshold value during printing, the visual module starts to visualize the printed paper, analyzes the file in the picture through artificial intelligence, including ink depth and printing position, to judge the printing effect. The artificial intelligence module combines the algorithm to judge the printing effect. Any printed paper is judged as failed by the artificial intelligence module. The paper is pushed out to the other side by the paper pushing module, so that it is out of the predetermined range of retained paper, and the content of the paper is printed again by the artificial intelligence module.

[0073] Another vibration abnormality, if the ball-type vibration sensor detects that the paper movement vibration exceeds the threshold value and the printing has not started, the scanning and printing device is directly ejected by the artificial intelligence module, and the paper pushing module is started by the artificial intelligence module to push the paper to the other side, so that it is out of the predetermined range of retained paper.

[0074] The working principle of the ball-type vibration sensor is mainly based on the movement of metal balls or cylindrical metal conductors inside the sensor, thereby changing the on or off state of the circuit; when the device is vibrated or angularly deflected, the metal balls or metal conductors inside the sensor will collide with each other, which causes the external lead electrodes to change from closed to open or from open to closed. After the circuit detects this change in switching quantity, it can control the circuit to send a corresponding response.

[0075] Example four, in examples two and three, how to handle a single page of paper with poor printing effect:

[0076] Any printed paper is judged as failed by the artificial intelligence module. The paper is pushed out to the other side by the paper pushing module, so that it is out of the predetermined range of retained paper, and the content of the paper is printed again by the artificial intelligence module.

[0077] Example five, multiple departments print different files in front and back, or files cannot be taken away in time, and multiple file papers appear to be stacked.

[0078] If two groups of files are printed continuously, the first group of files is completed by the artificial intelligence module to start the paper feed module, and the first file is displaced as a whole, and each multi-page file is processed in layers.

[0079] For example, the first group of files is moved 10mm to the left by the paper feed module, and the second group of files is moved 5mm to the left by the paper feed module, and there is a 5mm gap between the two groups of files, which facilitates the separation of the two groups of files, so that the boundary position of the two groups of files can be quickly separated.

[0080] Embodiment six, the current market scanning and printing equipment is provided with an ink cartridge, and the ink cartridge is provided with a pigment detection module for detecting the remaining amount of printing pigment. The pigment detection module signal is transmitted to the artificial intelligence module.

[0081] The scanning and printing equipment is provided with a paper box, and the bottom of the paper box is provided with a paper detection module for detecting the number of paper sheets. The signal of the paper detection module is transmitted to the artificial intelligence module.

[0082] After the file is uploaded to the scanning and printing equipment, the artificial intelligence module judges whether the file can be printed according to the remaining amount of pigment reflected by the pigment detection module and the number of paper sheets reflected by the paper detection module.

[0083] Yes, the file is printed, otherwise, the file uploading end is reminded to add ink cartridge or paper.

[0084] The information of insufficient paper or insufficient pigment can be transmitted back to the file uploading end through the artificial intelligence module, or the warning device of the scanning and printing equipment can be started to warn.

[0085] The scanning and printing equipment has a scanning function, which does not increase the cost of the visual module. The essence is that the artificial intelligence can achieve separate printing effect through multiple data training.

[0086] A line scan camera, an encoder, a light source and a lens combination can be used.

[0087] The scanning camera constructs the final two-dimensional image by scanning line by line, which is suitable for scanning fast-moving objects or large objects.

[0088] The encoder is used to control the relative motion between the camera and the object, to ensure the stability and continuity of image acquisition.

[0089] Selecting appropriate light sources and lenses is crucial to improving the quality of scanned images, especially in terms of high-speed acquisition and high-resolution imaging.

[0090] The above merely illustrates the embodiments of the present application, and does not limit the patent scope of the present application, and any equivalent structure or equivalent process transformation, or direct or indirect application in other related technical fields, which are made by using the content of the present application specification and drawings, are also included in the patent protection scope of the present application.

Claims

1. An artificial intelligence-based scan-print system, characterized by, The application relates to a scanning and printing device, which comprises the following parts: a scanning and printing device for printing and scanning of files and paper; an artificial intelligence module for learning and managing the processes of the scanning and printing device, wherein sound monitoring is added to the scanning and printing device, and vibration monitoring is added to the moving printing paper; a visual module added above the paper outlet of the scanning and printing device, which is used for visual imaging of the printed paper of the scanning and printing device, and the visual image is monitored by the artificial intelligence module, image segmentation and feature extraction technology are used to identify the ink area in the image, and the artificial intelligence module is used for deep and shallow classification of the areas; a paper pushing module added to one side of the paper outlet of the scanning and printing device, which is used for moving displacement of the single printed paper in the scanning and printing device or for overall displacement of the same group of paper of the same file; the artificial intelligence module is used for detecting the printed file of the scanning and printing device, wherein the printing effect is E, and the formula variables are as follows: sound data a, paper vibration frequency b, ink depth m and text centering n; w1, w2, w3, w4 are weight coefficients for adjusting the influence degree of each variable on the printing effect; a, b, Y, d are nonlinear indexes for simulating the nonlinear influence of each variable on the printing effect; c is a constant term representing the basic printing quality or system error; a sound detection sensor is added to the scanning and printing device to detect the working sound, and a ball type vibration sensor is installed on the moving path of the paper in the printing process to detect the vibration of the moving paper; if the mechanical sound of the scanning and printing device exceeds the threshold value during the printing process, the visual module is started to visually image the printed paper, the file in the image is analyzed by the artificial intelligence, including ink depth and printing position, and the printing effect is judged; whether the text is skewed or offset in the paper is distinguished by an algorithm, and the same is realized by the image of the visual module and the learning algorithm of the artificial intelligence module, the rotation angle of the rectangular paper is calculated, so that the skew angle of the text is obtained, the image is rotated and translated through affine transformation, and the skew and offset of the text are corrected; if the paper vibration of the scanning and printing device exceeds the threshold value during the printing process, the visual module is started to visually image the printed paper, the file in the image is analyzed by the artificial intelligence, including ink depth and printing position, and the printing effect is judged; if any printed paper is judged as failed by the artificial intelligence module, the paper is pushed out to the other side by the paper pushing module, so that the paper is separated from the predetermined paper storage range, and the content of the paper is printed again by the artificial intelligence module; if two groups of files are continuously printed, the first group of files is integrally displaced by the paper pushing module started by the artificial intelligence module after the first group of files is completed; if the vibration of the ball type vibration sensor exceeds the threshold value, and the printing has not started, the scanning and printing device is directly controlled by the artificial intelligence module to spit out the paper, and the paper is pushed out to the other side by the paper pushing module started by the artificial intelligence module, so that the paper is separated from the predetermined paper storage range.

2. The artificial intelligence-based scanning and printing system according to claim 1, wherein, The scanning and printing device is internally provided with an ink cartridge, the ink cartridge is internally provided with a colorant detection module for detecting the residual amount of printing colorant, the colorant detection module signal is transmitted to the artificial intelligence module, the scanning and printing device is internally provided with a paper box, the bottom of the paper box is provided with a paper detection module for detecting the number of papers, and the signal of the paper detection module is transmitted to the artificial intelligence module.

3. The artificial intelligence-based scanning and printing system according to claim 1, wherein, After the file is uploaded to the scanning and printing device, the artificial intelligence module determines whether the file can be printed according to the ink level reflected by the ink detection module and the paper quantity reflected by the paper detection module; if yes, the file is printed; if no, the file uploading end is reminded to add ink cartridges or paper.

Images