A printing robot connection printing method

Abstract

The application discloses a printing robot connection printing method, comprising the following steps: S1, the printing robot acquires the image of the document to be printed before printing, and extracts feature points from the image of the document to be printed; S2, the printing robot acquires the position of the printing breakpoint by acquiring the image of the printed content on the paper through a camera; S3, the printing robot acquires the image of the position of the printing breakpoint through the camera, and extracts feature points from the image of the position of the printing breakpoint; S4, the printing robot determines the printing connection point according to the feature points of the image of the document to be printed and the feature points of the image of the position of the printing breakpoint. The printing robot determines the printing connection point through the feature points of the image of the document to be printed and the feature points of the image of the position of the printing breakpoint, so that the printing can be continued after the printing robot is interrupted, and the paper is not needed to be replaced for reprinting, thereby saving the use of paper.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent robots, and particularly relates to a printing robot continuous printing method. BACKGROUND

[0002] The existing printer needs AC power supply to provide power, and can only be fixedly placed in the office and home. When the user goes out to work, the printer cannot be carried. Moreover, the existing printer is fixedly arranged, and the paper moves in the body of the printer to realize printing. Therefore, the width of the body of the printer must be greater than the width of the paper to realize printing, which is another reason why the printer cannot be carried. Due to the above reasons, a portable and self-moving mobile printer robot appears. The existing printer robot needs to position the position of the robot before starting work, so that the document to be printed can be accurately and completely printed on the paper. The positioning method of the existing printer robot is to determine the initial printing position through the corner points of the paper. However, when the printer robot wants to continue printing after interrupting the printing, it cannot continue printing, needs to replace the paper and reposition the printing, and wastes the paper. SUMMARY

[0003] To solve the above problems, the present application provides a printing robot continuous printing method. The specific technical scheme of the present application is as follows:

[0004] A printing robot continuous printing method, the method comprising the following steps: S1: the printing robot obtains an image of a document to be printed before printing, and extracts feature points from the image of the document to be printed; S2: the printing robot obtains the position of a printing breakpoint by obtaining an image of the printed content on the paper through a camera; S3: the printing robot obtains an image of the position of the printing breakpoint through the camera, and extracts feature points from the image of the position of the printing breakpoint; S4: the printing robot determines a printing continuation point according to the feature points of the image of the document to be printed and the feature points of the image of the position of the printing breakpoint; S5: the printing robot determines the position of the next printed character on the paper according to the position of the printing continuation point on the document to be printed and the position of the printing breakpoint on the paper to continue printing.

[0005] Further, in step S1, the printing robot extracts feature points from the image of the document to be printed, including the following steps: the printing robot calculates the image of the document to be printed by a difference of Gauss function to obtain a Gauss difference pyramid; the printing robot obtains extreme points according to the pixel values of the pixel points of the response image in the Gauss difference pyramid; the printing robot performs curve fitting calculation based on the extreme points of each layer of the Gauss difference pyramid to obtain accurate points; the printing robot copies the accurate points, and then assigns direction values to the copied accurate points to obtain accurate points in different directions; and the printing robot performs histogram statistics on the accurate points in different directions to obtain feature points of the image of the document to be printed.

[0006] Further, the printing robot calculates the image of the document to be printed by a difference of Gauss function, including the following steps: the printing robot performs convolution operation on the image of the document to be printed and Gauss kernels with different scale space factors to obtain a plurality of Gauss space images, and subtracts adjacent Gauss space images to obtain a response image; the printing robot performs down-sampling on the image of the document to be printed to obtain a document image with a smaller scale, and then performs convolution operation on the document image with a smaller scale and Gauss kernels with different scale space factors to obtain a plurality of Gauss space images, and subtracts adjacent Gauss space images to obtain a response image; the printing robot repeats the above steps to obtain a plurality of response images, and then combines the plurality of response images into a Gauss difference pyramid.

[0007] Further, the printing robot obtains extreme points according to the pixel values of the pixel points of the response image in the Gauss difference pyramid, including the following steps: the printing robot constructs a comparison template with a size of 3*3*3 pixel points, and traverses each response image constituting the Gauss difference pyramid through the comparison template; when traversing through the comparison template, the printing robot compares the pixel values of the pixel points in the comparison template of the response image and the pixel points corresponding to the position of the adjacent response image above and below the comparison template; and the printing robot takes the pixel point with the maximum or minimum pixel value as an extreme point.

[0008] Further, the printing robot performs curve fitting calculation based on the extreme points of each layer of the Gauss difference pyramid to obtain accurate points, including the following steps: the printing robot calculates the principal curvatures of the extreme points according to the curve fitting calculation formula; and the printing robot takes the extreme points with principal curvatures greater than or equal to a set curvature threshold as accurate points.

[0009] Further, in step S2, the printing robot obtains the position of the printing breakpoint by obtaining an image of the printed content on the paper through the camera, including the following steps: the printing robot obtains an image of the paper through the camera, and then identifies the printed characters in the image of the paper; the printing robot determines whether the printing line below the currently printed characters has printed characters; if the printing line below the currently printed characters does not have printed characters, the printing robot takes the character located at the rightmost side of the printing line of the currently printed characters as the position of the printing breakpoint.

[0010] Further, in step S3, the printing robot obtains an image of the position of the printing breakpoint through the camera, and extracts feature points from the image of the position of the printing breakpoint, including the following steps: the printing robot calculates the image of the position of the printing breakpoint through a difference of Gaussian function to obtain a Gaussian difference pyramid; the printing robot obtains extreme points according to the pixel values of the pixel points of the response image in the Gaussian difference pyramid; the printing robot performs curve fitting calculation based on the extreme points of each layer of the Gaussian difference pyramid to obtain accurate points; the printing robot copies the accurate points, and then assigns direction values to the copied accurate points to obtain accurate points in different directions; the printing robot performs histogram statistics on the accurate points in different directions to obtain feature points of the image of the position of the printing breakpoint.

[0011] Further, the printing robot determines the printing continuation point according to the feature points of the image of the to-be-printed document and the feature points of the image of the position of the printing breakpoint, including the following steps: the printing robot selects matching points from the feature points of the image of the to-be-printed document and the feature points of the image of the position of the printing breakpoint to construct an initial matching pair set through an Euclidean distance calculation formula; the printing robot randomly selects 8 feature points from the initial matching pair set to construct a screening model; the printing robot detects the feature points in the initial matching pair set through the screening model, and constructs an inlier set with the feature points meeting the requirements of the screening model; the printing robot again randomly selects 8 feature points to construct a screening model to obtain an inlier set, repeats the step, obtains a plurality of inlier sets, and selects an inlier set with the most feature points; the printing robot takes the printed characters corresponding to the feature points in the inlier set with the most feature points as the printing continuation point.

[0012] Furthermore, the printing robot determines the position of the next printed character on the paper based on the position of the printing continuation point on the document to be printed and the position of the printing break point on the paper, including the following steps: The printing robot obtains the position of the printed character corresponding to the printing continuation point on the document to be printed, and then the printing robot takes the printed character to the right and adjacent to the printed character corresponding to the printing continuation point as the next printed character; The printing robot obtains the position of the next printed character on the document to be printed, and then obtains the positional relationship based on the position of the printed character corresponding to the printing continuation point on the document to be printed and the position of the next printed character on the document to be printed; The printing robot obtains the position of the printed character corresponding to the printing break point on the paper, and then the printing robot determines the position of the next printed character on the paper based on the position of the printed character corresponding to the printing break point on the paper and the positional relationship, and then continues printing.

[0013] Furthermore, when the printing robot obtains the next print character, if there is no print character to the right of the print character corresponding to the print continuation point, the printing robot will take the first print character of the next print line of the print continuation point in the document to be printed as the next print character.

[0014] Compared with existing technologies, the advantages of the present invention are as follows: The printing robot described in this application determines the printing continuation point by using the feature points of the image of the document to be printed and the feature points of the image at the location of the printing breakpoint, so that printing can continue after the printing robot interrupts printing without replacing the paper to reprint, thus saving paper usage. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of a printing robot continuous printing method in one embodiment of the present invention. Implementation

[0016] The embodiments of the present invention are described in detail below. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout.

[0017] The following description, in conjunction with the accompanying drawings, further illustrates specific embodiments of the present invention, making the technical solution and its beneficial effects clearer and more explicit. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, but should not be construed as limiting the invention.

[0018] like Figure 1 As shown, a method for continuous printing by a printing robot, wherein the printed characters include, but are not limited to, Chinese characters, English characters, numbers, and symbols, etc., the method includes the following steps:

[0019] A printing robot continuous printing method, the method comprising the following steps: S1: the printing robot obtains an image of a document to be printed before printing, the printing robot receives the document to be printed sent by the user through a wireless module, and then converts the document to be printed into an image; or the printing robot directly receives the image of the document to be printed sent by the user through a wireless module, and then the printing robot extracts feature points from the image of the document to be printed, the printing robot can obtain the feature points of the image of the document to be printed by methods such as harris corner detection, Shi-Tomasi corner detection, FAST corner detection, scale invariant surf detection, scale invariant sift detection, and feature point description. S2: the printing robot obtains the position of the printing breakpoint by obtaining the image of the printed content on the paper through the camera, when the printing robot obtains the image of the paper through the camera, the printing robot can first move to the top of the paper, then move down from the top of the paper, and obtain the images of several papers during the movement, and then select the paper image with part of the printed characters and part of the blank from the obtained images of several papers; or the printing robot walks on the paper and obtains the images of several papers during the walking, and then splices the images to obtain a complete paper image; or the user lifts up the printing robot, and the printing robot obtains a complete paper image by shooting from top to bottom through the camera. After the printing robot obtains the image, the position of the printing breakpoint on the paper can be obtained according to the image recognition. S3: the printing robot obtains the image of the position of the printing breakpoint through the camera, and extracts feature points from the image of the position of the printing breakpoint, the printing robot can directly use the image in step S2 to extract feature points, or reobtain a clearer image of the position of the printing breakpoint to extract feature points. S4: the printing robot determines the position of the printing continuation point on the paper by matching the feature points of the image of the document to be printed and the feature points of the image of the position of the printing breakpoint. S5: the printing robot determines the position of the next printed character on the paper according to the position of the printing continuation point on the document to be printed and the position of the printing breakpoint on the paper to continue printing. Wherein, the printing breakpoint is the last printed character in the last printing process of the printing robot, and the printing continuation point is the corresponding character of the last printed character in the document to be printed. The printing robot determines the printing continuation point through the feature points of the image of the document to be printed and the feature points of the image of the position of the printing breakpoint, so that the printing robot can continue printing after the printing is interrupted, without replacing the paper to reprint, saving the use of paper.

[0020] As one of the embodiments, in step S1, the printing robot extracts feature points from the image of the document to be printed, including the following steps: the printing robot calculates the image of the document to be printed or the image of the paper by differential Gaussian function (differential Gaussian function is a function in SIFT (Scale Invariant Feature Transform) technology, which can obtain SIFT feature points by convolution operation of pixel matrix function of original image and Gaussian differential function, and these feature points are key points with Gaussian differential extreme points on the edge of the image) to obtain Gaussian differential pyramid; the printing robot obtains extreme points according to the pixel value of the pixel points of the response image in the Gaussian differential pyramid; the printing robot performs curve fitting calculation based on the extreme points of each layer of the Gaussian differential pyramid to obtain accurate points; the printing robot copies the accurate points, and then assigns direction values to the copied accurate points to obtain accurate points in different directions; the printing robot performs histogram statistics on the accurate points in different directions to obtain the feature points of the image of the document to be printed or the image of the paper. This method is SIFT feature extraction method. SIFT stands for Scale Invariant Feature Transform, which is a scale invariant feature transform. SIFT features remain invariant to rotation, scaling, brightness changes, etc., and are a very stable local feature. The printing robot obtains feature points by SIFT feature extraction method to improve the accuracy of feature point recognition.

[0021] As one of the embodiments, the printing robot calculates the image of the document to be printed by differential Gaussian function, including the following steps: the printing robot convolves the image of the document to be printed with Gaussian kernels of different scale space factors to obtain a plurality of Gaussian space images, and subtracts adjacent Gaussian space images to obtain response images; the printing robot performs down-sampling on the image of the document to be printed to obtain a smaller-scale image of the document to be printed, and then convolves the smaller-scale image of the document to be printed with Gaussian kernels of different scale space factors to obtain a plurality of Gaussian space images, and subtracts adjacent Gaussian space images to obtain response images; the printing robot repeats the above steps to obtain a plurality of response images, and then the printing robot combines the plurality of response images into a Gaussian differential pyramid. The printing robot matches the feature points by the Gaussian differential pyramid to improve the accuracy of feature point recognition.

[0022] As one of the embodiments, the printing robot obtains extreme points according to pixel values of pixel points of response images in a Gaussian difference pyramid, including the following steps: the printing robot constructs a comparison template of 3*3*3 pixel points, and traverses each response image constituting the Gaussian difference pyramid through the comparison template; when traversing through the comparison template, the printing robot compares pixel values of pixel points in the comparison template of the response image and pixel points of the comparison template at corresponding positions of adjacent response images above and below; the printing robot takes the pixel point with the maximum or minimum pixel value as an extreme point. The printing robot takes the pixel point with the maximum or minimum pixel value as an extreme point, which has high accuracy.

[0023] As one of the embodiments, the printing robot obtains accurate points through curve fitting calculation based on extreme points of each layer of the Gaussian difference pyramid, including the following steps: the printing robot calculates principal curvatures of the extreme points according to a curve fitting calculation formula; the printing robot takes the extreme point with a principal curvature greater than or equal to a set curvature threshold as an accurate point. The printing robot obtains accurate points through principal curvatures, which has high practicability.

[0024] As one of the embodiments, in step S2, the printing robot obtains the position of the printing breakpoint by obtaining an image of the printed content on the paper through the camera, including the following steps: the printing robot obtains an image of the paper through the camera, and then identifies the printed characters in the image of the paper; the printing robot determines whether the printing line below the current printed character has printed characters; if the printing line below the current printed character has no printed characters, the printing robot takes the character located at the rightmost side of the printing line of the current printed character as the position of the printing breakpoint. The printing robot determines the latest printing line according to whether the printing line below has printed characters, so as to determine the last printed character before the printing robot interrupts printing, i.e. the printing breakpoint, which can quickly help the printing robot find the position of the printing breakpoint.

[0025] As one of the embodiments, in step S3, the printing robot obtains an image of the position of the printing breakpoint through the camera, and extracts feature points from the image of the position of the printing breakpoint, including the following steps: the printing robot calculates the image of the position of the printing breakpoint through a difference Gaussian function to obtain a Gaussian difference pyramid; the printing robot obtains extreme points according to pixel values of pixel points of response images in the Gaussian difference pyramid; the printing robot obtains accurate points through curve fitting calculation based on extreme points of each layer of the Gaussian difference pyramid; the printing robot copies the accurate points, and then assigns direction values to the copied accurate points to obtain accurate points in different directions; the printing robot obtains feature points of the image of the position of the printing breakpoint through histogram statistics of the accurate points in different directions. The printing robot obtains feature points through the difference pyramid, which improves the accuracy of feature point identification.

[0026] As one of the embodiments, the printing robot determines the printing continuation point according to the feature points of the image of the document to be printed and the feature points of the image of the printing breakpoint, including the following steps: the printing robot selects matched feature points from the feature points of the image of the document to be printed and the feature points of the image of the paper by using the Euclidean distance calculation formula, and then forms an initial matching pair set by using all the matched feature points. The Euclidean distance is the actual distance between two points in an n-dimensional space, and the similarity of feature points in an image can be calculated by using the Euclidean distance. The smaller the Euclidean distance is, the greater the similarity is. The core of the Euclidean distance algorithm is that the image matrix has n elements (n pixel points), and the feature group (all pixel points in the pixel point matrix) of the image is formed by using the n element values (x1, x2,..., xn). The feature group forms an n-dimensional space (the Euclidean distance is for a multi-dimensional space), and the feature code (each pixel point) in the feature group constitutes the value of each dimension, that is, x1 (the first pixel point) corresponds to one dimension, x2 (the second pixel point) corresponds to two dimensions,..., and xn (the nth pixel point) corresponds to n dimensions. In the n-dimensional space, two image matrices form a point, and then the Euclidean distance formula is used to calculate the distance between the two points. The Euclidean distance formula is: point A = (x1, x2,..., xn); point B = (y1, y2,..., yn); AB^2= (x1-y1)^2+(x2-y2)^2+...+(xn-yn)^2; AB is the distance between the two points A and B in the multi-dimensional space, and when the distance between A and B is less than the set threshold, A and B are determined as matched feature points. The printing robot randomly selects 8 feature points from the initial matching pair set to construct a screening model; the printing robot detects the feature points in the initial matching pair set by using the screening model, and constructs an inner point set by using the feature points meeting the requirements of the screening model; the screening model is constructed by using the RANSAC algorithm, and RANSAC is the abbreviation of “RANdom SAmple Consensus”. It can estimate the parameters of a mathematical model through an iterative method from a set of observation data containing “outliers”. The input of the RANSAC algorithm needs a data set and a parameterized model representing the data set, and some confidence parameters.The RANSAC algorithm obtains the final result by iteratively selecting a random subset from the original data, assuming that the subset of data is an inlier, and then verifying this assumption by the following steps: fitting a data model from the assumed inliers, i.e. reconstructing all free parameters of the model from the inliers; all other data is verified against the fitted model, and if a point fits the resulting model, it is considered an inlier; if there are enough inliers, the estimated model is considered good; then the model is re-estimated from all inliers, because the previous model was only estimated from the initial inliers; and finally the model is evaluated by the error of the inliers and the model. The printing robot randomly selects 8 feature points to construct a screening model to obtain an inlier set, and after repeating this step several times, a plurality of inlier sets are obtained, and the inlier set with the most feature points is selected; the printing robot takes the printing character corresponding to the feature point in the inlier set with the most feature points as the printing continuation point. The printing robot removes noise points by constructing a screening model, so that the matching result of the feature points is more accurate.

[0027] As one of the embodiments, the printing robot determines the position of the next printing character on the paper according to the position of the printing continuation point on the to-be-printed document and the position of the printing breakpoint on the paper to continue printing, including the following steps: the printing robot obtains the position of the printing character corresponding to the printing continuation point on the to-be-printed document, and then the printing robot takes the printing character adjacent to the right of the printing character corresponding to the printing continuation point as the next printing character; the printing robot obtains the position of the next printing character on the to-be-printed document, and then obtains the positional relationship according to the position of the printing character corresponding to the printing continuation point on the to-be-printed document and the position of the next printing character on the to-be-printed document; the printing robot obtains the position of the printing character corresponding to the printing breakpoint on the paper, and then the printing robot determines the position of the next printing character on the paper according to the position of the printing character corresponding to the printing breakpoint on the paper and the positional relationship to continue printing. The printing robot determines the printing position of the next printing character according to the positional relationship between the two characters, which is highly practical.

[0028] As one of the embodiments, when the printing robot obtains the next printing character, if there is no printing character to the right of the printing character corresponding to the printing continuation point, the printing robot takes the first printing character of the next printing line of the printing character corresponding to the printing continuation point in the to-be-printed document as the next printing character. When the printing robot does not find a printing character to the right of the same printing line, the printing robot takes the first printing character of the next line as the next printing character, which is highly flexible.

[0029] A chip for storing a program configured to perform the printing robot continuation printing method described above.

[0030] A printing robot, the printing robot comprising a master control chip, the master control chip being the chip described above.

[0031] Compared with the prior art, the printing robot has the beneficial effects that the printing robot determines the printing continuation point through the feature points of the image of the to-be-printed document and the feature points of the image of the position of the printing breakpoint, so that the printing can be continued after the printing robot interrupts the printing, and the paper is not needed to be replaced for reprinting, thereby saving the use of paper.

[0032] In the description of the specification, the description of the terms "in combination with an embodiment", "preferably", "example", "specific example" or "some examples" and the like means that the specific features, structures, materials or characteristics described in combination with the embodiment or example are included in at least one embodiment or example of the present application, and the illustrative description of the above terms in the specification does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. The connection mode in the description of the specification has obvious effects and practical effects.

[0033] Through the description of the structure and principle described above, those skilled in the art should understand that the present application is not limited to the specific embodiments described above, and improvements and substitutions of the technology known in the art on the basis of the present application all fall within the protection scope of the present application, which should be defined by the claims.

Claims

1. A printing robot continuous printing method characterized by, The method comprises the following steps: S1: The printing robot obtains an image of a document to be printed before printing, and extracts feature points from the image of the document to be printed; S2: The printing robot obtains the position of the printing breakpoint by obtaining the image of the printed content on the paper through the camera; S3: The printing robot obtains the image of the position of the printing breakpoint through the camera, and extracts feature points from the image of the position of the printing breakpoint; S4: The printing robot determines the printing continuation point according to the feature points of the image of the document to be printed and the feature points of the image of the position of the printing breakpoint; S5: The printing robot determines the position of the next printed character on the paper according to the position of the printing continuation point on the document to be printed and the position of the printing breakpoint on the paper to continue printing.

2. The method of claim 1, wherein, In step S1, the printing robot extracts feature points from the image of the document to be printed, comprising the following steps: The printing robot calculates the image of the document to be printed by a difference Gaussian function to obtain a Gaussian difference pyramid; The printing robot obtains extreme points according to the pixel values of the pixel points of the response image in the Gaussian difference pyramid; The printing robot performs curve fitting calculation based on the extreme points of each layer of the Gaussian difference pyramid to obtain accurate points; The printing robot copies the accurate points, and then assigns direction values to the copied accurate points to obtain accurate points in different directions; The printing robot performs histogram statistics on the accurate points in different directions to obtain the feature points of the image of the document to be printed.

3. The method of claim 2, wherein, The printing robot calculates the image of the document to be printed by a difference Gaussian function, comprising the following steps: The printing robot convolves the image of the document to be printed with Gaussian kernels of different scale space factors to obtain a plurality of Gaussian space images, and subtracts adjacent Gaussian space images to obtain a response image; The printing robot down-samples the image of the document to be printed to obtain a document to be printed with a smaller scale, and then convolves the document to be printed with a smaller scale with Gaussian kernels of different scale space factors to obtain a plurality of Gaussian space images, and subtracts adjacent Gaussian space images to obtain a response image; The printing robot repeats the above steps to obtain a plurality of response images, and then combines the plurality of response images into a Gaussian difference pyramid.

4. The method of claim 3, wherein, The printing robot obtains extreme points according to the pixel values of the pixel points of the response image in the Gaussian difference pyramid, comprising the following steps: The printing robot constructs a comparison template of 3*3*3 pixel points, and traverses each response image constituting the Gaussian difference pyramid through the comparison template; When traversing through the comparison template, the printing robot compares the pixel values of the pixel points in the comparison template of the response image and the pixel points in the comparison template at the corresponding positions of the adjacent response images above and below; The printing robot takes the pixel point with the maximum or minimum pixel value as the extreme point.

5. The method of claim 4, wherein, The printing robot performs curve fitting calculation based on the extreme points of each layer of the Gaussian difference pyramid to obtain accurate points, comprising the following steps: The printing robot calculates the principal curvatures of the extreme points according to the curve fitting calculation formula; The printing robot takes the extreme points with principal curvatures greater than or equal to a set curvature threshold as accurate points.

6. The method of claim 2, wherein, In step S2, the printing robot obtains the position of the printing breakpoint by the camera obtaining the image of the printed content on the paper, including the following steps: The printing robot obtains the image of the paper by the camera, and then identifies the printed characters in the image of the paper; The printing robot determines whether the printing line below the currently printed character has printed characters; If the printing line below the currently printed character has no printed characters, the printing robot takes the rightmost character in the printing line of the currently printed character as the position of the printing breakpoint.

7. The method of claim 6, wherein, In step S3, the printing robot obtains the image of the position of the printing breakpoint by the camera, and extracts feature points from the image of the position of the printing breakpoint, including the following steps: The printing robot calculates the image of the position of the printing breakpoint by the difference Gaussian function to obtain a Gaussian difference pyramid; The printing robot obtains extreme points according to the pixel values of the pixel points in the response image in the Gaussian difference pyramid; The printing robot performs curve fitting calculation based on the extreme points of each layer of the Gaussian difference pyramid to obtain accurate points; The printing robot copies the accurate points, and then assigns direction values to the copied accurate points to obtain accurate points in different directions; The printing robot performs histogram statistics on the accurate points in different directions to obtain feature points of the image of the position of the printing breakpoint.

8. The method of claim 7, wherein, The printing robot determines the printing continuation point according to the feature points of the image of the to-be-printed document and the feature points of the image of the position of the printing breakpoint, including the following steps: The printing robot selects matching points from the feature points of the image of the to-be-printed document and the feature points of the image of the position of the printing breakpoint to construct an initial matching pair set by the Euclidean distance calculation formula; The printing robot randomly selects 8 feature points from the initial matching pair set to construct a screening model; The printing robot detects the feature points in the initial matching pair set by the screening model, and constructs an inner point set with the feature points meeting the requirements of the screening model; The printing robot randomly selects 8 feature points to construct a screening model to obtain an inner point set, repeats the step, obtains several inner point sets, and selects the inner point set with the most feature points; The printing robot takes the printed characters corresponding to the feature points in the inner point set with the most feature points as the printing continuation point.

9. The method of claim 8, wherein, The printing robot determines the position of the next printed character on the paper according to the position of the printing continuation point on the to-be-printed document and the position of the printing breakpoint on the paper to continue printing, including the following steps: The printing robot obtains the position of the printed character corresponding to the printing continuation point on the to-be-printed document, and then the printing robot takes the printed character adjacent to the right of the printed character corresponding to the printing continuation point as the next printed character; The printing robot obtains the position of the next printed character on the to-be-printed document, and then obtains the positional relationship according to the position of the printed character corresponding to the printing continuation point on the to-be-printed document and the position of the next printed character on the to-be-printed document; The printing robot obtains the position of the printed character corresponding to the printing breakpoint on the paper, and then the printing robot determines the position of the next printed character on the paper according to the position of the printed character corresponding to the printing breakpoint on the paper and the positional relationship to continue printing.

10. The method of claim 9, wherein, When the printing robot acquires the next printing character, if there is no printing character to the right of the printing character corresponding to the printing continuation point, the printing robot takes the first printing character of the next printing line of the printing character corresponding to the printing continuation point in the document to be printed as the next printing character.

Images