A printing packaging box management system and method based on color recognition

Abstract

The present invention discloses a printing packaging box management system and method based on color recognition, belongs to the field of printing packaging box management, relates to color recognition technology, and is used to solve the problem that it is difficult to distinguish the color of local or subtle parts using color recognition sensors in the prior art and the calculation workload is relatively large; from the perspective of image pixelation processing, firstly, a first target view image is pixelated to obtain a first target two-dimensional dot matrix image, and then a recognition processing module generates a first target pixel point A, and at the same time, a first rotation radius R1 is given, and the first target pixel point A is used as the coordinate origin to establish a temporary rotation coordinate, and the second target pixel point B is determined by rotating and drawing an arc, and the original target view images of all printed packaging boxes are processed synchronously, and the printed packaging boxes to be managed are identified through point-to-point comparison, thereby reducing the calculation workload.

Description

Technical Field

[0001] The invention belongs to the field of printing packaging box management, relates to color recognition technology, and specifically provides a printing packaging box management system and method based on color recognition. Background Art

[0002] Printed packaging boxes are a form of packaging widely used in the transportation and display of goods. They are usually made of paper materials, such as cardboard, corrugated cardboard, etc., and decorative patterns, designs or texts are applied to the surface through printing technology.

[0003] In some packaging factories, the same type of goods mostly use the same type of printed packaging boxes, which are mostly consistent or similar in size and style. Different colors are generally used to distinguish the specifications of the goods. After the goods are packaged, the means of distinction mostly falls on color recognition, which is generally achieved directly using color sensors.

[0004] Color recognition sensors in the prior art are devices specifically designed to sense and identify colors. They are usually composed of a light source, a receiver, and corresponding circuits. When light is irradiated onto a target object, the sensor receives the reflected light and converts it into electrical signals, which are then processed to determine the color of the object. Spectrometer-type sensors measure color by analyzing the spectral components of reflected light, while color filter-type sensors use color filters to filter specific wavelengths of light and determine color by comparing the light intensity of transmitted and reflected light.

[0005] However, the above color sensors do not have obvious advantages in local or subtle color recognition, and the subsequent recognition algorithms are generally based on the full size of the object to be recognized, which results in a large amount of computational workload.

[0006] To this end, the present invention proposes a printing packaging box management system and method based on color recognition. Summary of the invention

[0007] The purpose of this application is to provide a printing packaging box management system and method based on color recognition, which is used to solve the problem in the prior art that it is difficult to distinguish the color of local or subtle parts using color recognition sensors and the calculation workload is relatively large.

[0008] To achieve the above objectives, an embodiment of the present application provides a printing packaging box management system based on color recognition, including:

[0009] An image acquisition module for acquiring a first target view image of a printed packaging box to be managed and original target view images of all printed packaging boxes of a target factory;

[0010] A preprocessing module for performing pixelation processing on the first target view image and the original target view image to obtain a first target two-dimensional dot matrix image and an original target two-dimensional dot matrix image; and

[0011] The recognition processing module receives the first target two-dimensional dot matrix image, establishes a rectangular two-dimensional coordinate system, generates coordinate values ​​for each pixel point P, and marks them as P(x, y), where x is the horizontal coordinate and y is the vertical coordinate; obtains the RGB value of the corresponding pixel point and marks it as S P =(R P ,G P ,B P );

[0012] Generate a first target pixel point A, set a first rotation radius R1, and establish a temporary rotation coordinate with the first target pixel point A as the coordinate origin; draw an arc based on the coordinate origin and the first rotation radius R1, and obtain the pending pixel points n on the arc in the first quadrant in sequence according to the horizontal coordinates, where n is an integer, representing the horizontal coordinate of the pending pixel point, n=1,2...N; N represents the total number of the pending pixel points n;

[0013] Use the calculation formula to calculate the confirmed value QX of the undetermined pixel n and the first target pixel A An ;

[0014] Among them, the selected value QX An The calculation formula is:

[0015]

[0016] Wherein, α is the correction coefficient, 0<α<1;

[0017] The recognition processing module sets the selected value range (QX0, QX1), and the recognition processing module compares the selected value range (QX0, QX1) with the calculated selected value QX An Compare them one by one;

[0018] When QX An ∈(QX0, QX1), the corresponding undetermined pixel point n is marked as the second target preliminary point; and the point with the smallest horizontal coordinate among the second target preliminary points is determined as the second target pixel point B, and the RGB value of the second target pixel point B is obtained at the same time, S B =(R B ,G B ,B B );

[0019] The recognition processing module obtains a first original pixel point a and a second original pixel point b in the original target two-dimensional dot matrix image according to the generation rule of the first target pixel point A and the second target pixel point B;

[0020] The recognition processing module is also used to compare the RGB values ​​of the first target pixel point A and the second target pixel point B with the RGB values ​​of the corresponding first original pixel points a and the second original pixel points b respectively;

[0021] When the RGB values ​​of the first target pixel point A and the first original pixel point a, and the RGB values ​​of the second target pixel point B and the second original pixel point b are all the same, a plurality of identical original target two-dimensional dot matrix images are marked as target images.

[0022] Preferably, the image acquisition module acquires the first target view image of the printed packaging box to be managed through a CCD camera No. 1 connected thereto;

[0023] The original target view images of all printed packaging boxes of the target factory are collected by a No. 2 CCD camera connected thereto;

[0024] The model and specifications of the No. 1 CCD camera and the No. 2 CCD camera are the same.

[0025] Preferably, the target graph is one or more;

[0026] If there is only one target image, the single target image is fed back to the centralized management module, and the centralized management module determines the original target view image according to the received single target image, and further identifies the category of the printed packaging box to be managed.

[0027] Preferably, if there are multiple target images, the recognition processing module marks the multiple target images as two batches of two-dimensional dot matrix images, sets a second rotation radius R2, establishes temporary rotation coordinates with the second target pixel point B as the coordinate origin, and obtains the third target pixel point C, wherein the acquisition rule of the third target pixel point C is consistent with the acquisition rule of the second target pixel point B;

[0028] After the recognition processing module generates the second target pixel point B and the second target pixel point C, the recognition processing module obtains the second original pixel point b and the third original pixel point c in the two batches of two-dimensional dot matrix images according to the generation rule of the second target pixel point B and the second target pixel point C;

[0029] The recognition processing module also compares the RGB values ​​of the second target pixel point B and the second target pixel point C with the RGB values ​​of the corresponding second original pixel points b and the third original pixel point c respectively;

[0030] When the RGB values ​​of the second target pixel B and the second original pixel b, and the RGB values ​​of the third target pixel C and the third original pixel c are the same, a number of identical original target two-dimensional dot matrix images are marked as secondary target images.

[0031] Preferably, if there is only one secondary target image, the single target image is fed back to the centralized management module, and the centralized management module determines the original target view image according to the received single target image, and further identifies the category of the printed packaging box to be managed;

[0032] If there are multiple secondary target images, continue the above steps until the multiple target images are one, then feed back a single target image to the centralized management module, and the centralized management module determines the original target view image based on the received single target image, and then identifies the category of the printed packaging box to be managed.

[0033] Preferably, it also includes a server connected to the preprocessing module, and used to receive and store the original target two-dimensional dot matrix image sent by the preprocessing module;

[0034] The server is also used to receive a plurality of original target view images sent by the image acquisition module, and to make a one-to-one correspondence between the plurality of original target view images and the original target two-dimensional dot matrix images processed by the pre-processing module;

[0035] The preprocessing module is also used to send the original target two-dimensional dot matrix image after preprocessing to the server;

[0036] The image acquisition module is connected to the server and sends the acquired original target view images to the server.

[0037] Another embodiment of the present application provides a printed packaging box management method based on color recognition, comprising the following steps:

[0038] The image acquisition module acquires the first target view image of the printed packaging box to be managed and the original target view images of all printed packaging boxes of the target factory;

[0039] The preprocessing module performs pixelation processing on the first target view image and the original target view image to obtain a first target two-dimensional dot matrix image and an original target two-dimensional dot matrix image;

[0040] The recognition processing module receives the first target two-dimensional dot matrix image, establishes a rectangular two-dimensional coordinate system, generates coordinate values ​​for each pixel point P, and marks them as P(x, y), where x is the horizontal coordinate and y is the vertical coordinate; obtains the RGB value of the corresponding pixel point and marks it as S P =(R P ,G P ,B P );

[0041] Generate a first target pixel point A, set a first rotation radius R1, and establish a temporary rotation coordinate with the first target pixel point A as the coordinate origin; draw an arc based on the coordinate origin and the first rotation radius R1, and obtain the pending pixel points n on the arc in the first quadrant in sequence according to the horizontal coordinates, where n is an integer, representing the horizontal coordinate of the pending pixel point, n=1,2...N; N represents the total number of the pending pixel points n;

[0042] Use the calculation formula to calculate the confirmed value QX of the undetermined pixel n and the first target pixel A An ;

[0043] Among them, the selected value QX An The calculation formula is:

[0044]

[0045] Wherein, α is the correction coefficient, 0<α<1;

[0046] The recognition processing module sets the selected value range (QX0, QX1), and the recognition processing module compares the selected value range (QX0, QX1) with the calculated selected value QX An Compare them one by one;

[0047] When QX An ∈(QX0, QX1), the corresponding undetermined pixel point n is marked as the second target preliminary point; and the point with the smallest horizontal coordinate among the second target preliminary points is determined as the second target pixel point B, and the RGB value of the second target pixel point B is obtained at the same time, S B =(R B ,G B ,B B );

[0048] The recognition processing module obtains a first original pixel point a and a second original pixel point b in the original target two-dimensional dot matrix image according to the generation rule of the first target pixel point A and the second target pixel point B;

[0049] The recognition processing module is also used to compare the RGB values ​​of the first target pixel point A and the second target pixel point B with the RGB values ​​of the corresponding first original pixel points a and the second original pixel points b respectively;

[0050] When the RGB values ​​of the first target pixel point A and the first original pixel point a, and the RGB values ​​of the second target pixel point B and the second original pixel point b are all the same, marking a plurality of identical original target two-dimensional dot matrix images as target images;

[0051] If there are multiple target images, the recognition processing module marks the multiple target images as two batches of two-dimensional dot matrix images, sets a second rotation radius R2, establishes temporary rotation coordinates with the second target pixel point B as the coordinate origin, and obtains the third target pixel point C, wherein the acquisition rule of the third target pixel point C is consistent with the acquisition rule of the second target pixel point B;

[0052] After the recognition processing module generates the second target pixel point B and the second target pixel point C, the recognition processing module obtains the second original pixel point b and the third original pixel point c in the two batches of two-dimensional dot matrix images according to the generation rule of the second target pixel point B and the second target pixel point C;

[0053] The recognition processing module also compares the RGB values ​​of the second target pixel point B and the second target pixel point C with the RGB values ​​of the corresponding second original pixel points b and the third original pixel point c respectively;

[0054] When the RGB values ​​of the second target pixel B and the second original pixel b, and the RGB values ​​of the third target pixel C and the third original pixel c are the same, a number of identical original target two-dimensional dot matrix images are marked as secondary target images.

[0055] If there is only one secondary target image, the single target image is fed back to the centralized management module, and the centralized management module determines the original target view image according to the received single target image, and further identifies the category of the printed packaging box to be managed;

[0056] If there are multiple secondary target images, continue the above steps until the multiple target images are one, then feed back a single target image to the centralized management module, and the centralized management module determines the original target view image based on the received single target image, and then identifies the category of the printed packaging box to be managed.

[0057] Compared with the prior art, the present invention has the following beneficial effects:

[0058] From the perspective of image pixelation processing, the present invention first performs pixelation processing on the first target view image to obtain the first target two-dimensional dot matrix image, and then uses the recognition processing module to generate the first target pixel point A. At the same time, a first rotation radius R1 is given, and the first target pixel point A is used as the coordinate origin to establish a temporary rotation coordinate. The second target pixel point B is determined by rotating and drawing an arc, and the original target view images of all printed packaging boxes are processed synchronously. The printed packaging boxes to be managed are identified through point-to-point comparison, thereby reducing the workload of calculation.

[0059] However, considering the local colors that are difficult to distinguish, the present application also uses this rule to generate other third target pixel points C, and obtains the second original pixel point b and the third original pixel point c in the two batches of two-dimensional dot matrix images according to the generation rule of the second target pixel point B and the second target pixel point C; and compares the RGB values ​​of the second target pixel point B and the second target pixel point C with the RGB values ​​of the corresponding second original pixel points b and the third original pixel points c respectively; when the RGB values ​​of the second target pixel point B and the second original pixel b, and the RGB values ​​of the third target pixel point C and the third original pixel point c are the same, the same original pixel points are compared. The two-dimensional dot matrix image is marked as a secondary target image; wherein the secondary target image may be one or more; if the secondary target image is one, then the single target image is fed back to the centralized management module, and the centralized management module determines the original target view image according to the received single target image, and further identifies the category of the printed packaging box to be managed; if the secondary target images are multiple, continue the above steps until the multiple target images are one, then the single target image is fed back to the centralized management module, and the centralized management module determines the original target view image according to the received single target image, and further identifies the category of the printed packaging box to be managed, thereby further improving the accuracy of color recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the prior art and the drawings required for use in the embodiments are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0061] Figure 1 This is a structural block diagram of a printing packaging box management system based on color recognition according to the present invention. DETAILED DESCRIPTION

[0062] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0063] Example 1

[0064] Please refer to Figure 1 , a printing packaging box management system based on color recognition, including an image acquisition module, a preprocessing module and an identification processing module;

[0065] The image acquisition module and the preprocessing module are both connected to the recognition processing module;

[0066] The image acquisition module acquires the first target view image of the printed packaging box to be managed through the No. 1 CCD camera connected thereto, and sends the acquired first target view image of the printed packaging box to the preprocessing module; and acquires the original target view images of all printed packaging boxes in the target factory through the No. 2 CCD camera connected thereto, and sends the acquired several original target view images to the preprocessing module;

[0067] A preprocessing module, performing preprocessing based on the received first target view image and the original target view image, wherein the preprocessing includes pixelation processing to obtain the first target two-dimensional dot matrix image and the original target two-dimensional dot matrix image;

[0068] In the present application, a server is also included, which is connected to the preprocessing module and is used to receive and store the original target two-dimensional dot matrix image sent by the preprocessing module; it should be noted that the server is also used to receive a plurality of original target view images sent by the image acquisition module, and one-to-one correspondence between the plurality of original target view images and the original target two-dimensional dot matrix image processed by the preprocessing module;

[0069] It is obvious that the preprocessing module is also used to send the original target two-dimensional dot matrix image after preprocessing to the server;

[0070] The image acquisition module is connected to the server and sends the acquired original target view images to the server;

[0071] The recognition processing module is used to send the extraction signal to the preprocessing module after receiving the recognition signal, and the preprocessing module feeds back the first target two-dimensional dot matrix image to the recognition processing module;

[0072] When the recognition processing module receives the first target two-dimensional dot matrix image, a rectangular two-dimensional coordinate system is established for the first target two-dimensional dot matrix image, and a coordinate value is generated for each pixel point P, and marked as P(x, y), where x is the horizontal coordinate and y is the vertical coordinate; the recognition processing module obtains the RGB value of the corresponding pixel point and marks it as S P =(R P ,G P ,B P ), where pixel point P is any coordinate point in a rectangular two-dimensional coordinate system;

[0073] The recognition processing module is also used to generate a first target pixel point A and obtain the RGB value of the first target pixel point, S A =(R A ,G A ,BA );

[0074] After the recognition processing module obtains the RGB value of the first target pixel point A, a first rotation radius R1 is given, and a temporary rotation coordinate is established with the first target pixel point A as the coordinate origin, wherein the X axis and Y axis of the rotation coordinate are respectively parallel to the X axis and Y axis of the rectangular two-dimensional coordinate system;

[0075] The recognition processing module draws an arc based on the coordinate origin and the first rotation radius R1, and sequentially obtains the undetermined pixel points n on the arc in the first quadrant according to the horizontal coordinates, where n is an integer representing the horizontal coordinate of the undetermined pixel point, n=1,2...N; N represents the total number of the undetermined pixel points n;

[0076] Synchronously obtain the RGB value of the pending pixel n, S n =(R n ,G n ,B n );

[0077] The recognition processing module uses the calculation formula to calculate the confirmed value QX of the undetermined pixel n and the first target pixel A An ;

[0078] Among them, the selected value QX An The calculation formula is:

[0079]

[0080] Wherein, α is the correction coefficient, 0<α<1;

[0081] The recognition processing module sets the selected value range (QX0, QX1), and the recognition processing module compares the selected value range (QX0, QX1) with the calculated selected value QX An Compare them one by one;

[0082] When QX An ∈(QX0, QX1), the corresponding undetermined pixel point n is marked as the second target preliminary point; and the point with the smallest horizontal coordinate among the second target preliminary points is determined as the second target pixel point B, and the RGB value of the second target pixel point B is obtained at the same time, S B =(R B ,G B ,B B );

[0083] After the recognition processing module generates the first target pixel point A and the second target pixel point B, it sends a source tracing signal to the server, and the server feeds back the original target two-dimensional dot matrix image to the recognition processing module;

[0084] The recognition processing module obtains a first original pixel point a and a second original pixel point b in the original target two-dimensional dot matrix image according to the generation rule of the first target pixel point A and the second target pixel point B;

[0085] The recognition processing module is also used to compare the RGB values ​​of the first target pixel point A and the second target pixel point B with the RGB values ​​of the corresponding first original pixel points a and the second original pixel points b respectively;

[0086] When the RGB values ​​of the first target pixel point A and the first original pixel point a, and the RGB values ​​of the second target pixel point B and the second original pixel point b are all the same, marking a plurality of identical original target two-dimensional dot matrix images as target images; wherein the target images may be one or more;

[0087] If there is only one target image, the single target image is fed back to the centralized management module, and the centralized management module determines the original target view image according to the received single target image, and further identifies the category of the printed packaging box to be managed;

[0088] If there are multiple target images, the recognition processing module marks the multiple target images as two batches of two-dimensional dot matrix images, sets a second rotation radius R2, establishes temporary rotation coordinates with the second target pixel point B as the coordinate origin, and obtains the third target pixel point C, wherein the acquisition rule of the third target pixel point C is consistent with the acquisition rule of the second target pixel point B;

[0089] After the recognition processing module generates the second target pixel point B and the second target pixel point C, the recognition processing module obtains the second original pixel point b and the third original pixel point c in the two batches of two-dimensional dot matrix images according to the generation rule of the second target pixel point B and the second target pixel point C;

[0090] The recognition processing module also compares the RGB values ​​of the second target pixel point B and the second target pixel point C with the RGB values ​​of the corresponding second original pixel points b and the third original pixel point c respectively;

[0091] When the RGB values ​​of the second target pixel B and the second original pixel b, and the RGB values ​​of the third target pixel C and the third original pixel c are the same, a plurality of identical original target two-dimensional dot matrix images are marked as secondary target images; wherein the secondary target images may be one or more;

[0092] If there is only one secondary target image, the single target image is fed back to the centralized management module, and the centralized management module determines the original target view image according to the received single target image, and further identifies the category of the printed packaging box to be managed;

[0093] If there are multiple secondary target images, continue the above steps until the multiple target images are one, then feed back a single target image to the centralized management module, and the centralized management module determines the original target view image based on the received single target image, and then identifies the category of the printed packaging box to be managed.

[0094] Obviously, the present application also includes a centralized management module, which is used to classify the printed packaging boxes identified to be managed to ensure orderly processing of the printed packaging boxes.

[0095] From the perspective of image pixelation processing, the present invention first performs pixelation processing on the first target view image to obtain the first target two-dimensional dot matrix image, and then uses the recognition processing module to generate the first target pixel point A. At the same time, a first rotation radius R1 is given, and the first target pixel point A is used as the coordinate origin to establish a temporary rotation coordinate. The second target pixel point B is determined by rotating and drawing an arc, and the original target view images of all printed packaging boxes are processed synchronously. The printed packaging boxes to be managed are identified through point-to-point comparison, thereby reducing the workload of calculation.

[0096] Example 2

[0097] A method for managing printed packaging boxes based on color recognition comprises the following steps:

[0098] The image acquisition module acquires the first target view image of the printed packaging box to be managed and the original target view images of all printed packaging boxes of the target factory;

[0099] The preprocessing module performs pixelation processing on the first target view image and the original target view image to obtain a first target two-dimensional dot matrix image and an original target two-dimensional dot matrix image;

[0100] The recognition processing module receives the first target two-dimensional dot matrix image, establishes a rectangular two-dimensional coordinate system, generates coordinate values ​​for each pixel point P, and marks them as P(x, y), where x is the horizontal coordinate and y is the vertical coordinate; obtains the RGB value of the corresponding pixel point and marks it as S P =(R P ,G P ,B P );

[0101] Generate a first target pixel point A, set a first rotation radius R1, and establish a temporary rotation coordinate with the first target pixel point A as the coordinate origin; draw an arc based on the coordinate origin and the first rotation radius R1, and obtain the pending pixel points n on the arc in the first quadrant in sequence according to the horizontal coordinates, where n is an integer, representing the horizontal coordinate of the pending pixel point, n=1,2...N; N represents the total number of the pending pixel points n;

[0102] Use the calculation formula to calculate the confirmed value QX of the undetermined pixel n and the first target pixel A An ;

[0103] Among them, the selected value QX An The calculation formula is:

[0104]

[0105] Wherein, α is the correction coefficient, 0<α<1;

[0106] The recognition processing module sets the selected value range (QX0, QX1), and the recognition processing module compares the selected value range (QX0, QX1) with the calculated selected value QX An Compare them one by one;

[0107] When QX An ∈(QX0, QX1), the corresponding undetermined pixel point n is marked as the second target preliminary point; and the point with the smallest horizontal coordinate among the second target preliminary points is determined as the second target pixel point B, and the RGB value of the second target pixel point B is obtained at the same time, S B =(R B ,G B ,B B );

[0108] The recognition processing module obtains a first original pixel point a and a second original pixel point b in the original target two-dimensional dot matrix image according to the generation rule of the first target pixel point A and the second target pixel point B;

[0109] The recognition processing module is also used to compare the RGB values ​​of the first target pixel point A and the second target pixel point B with the RGB values ​​of the corresponding first original pixel points a and the second original pixel points b respectively;

[0110] When the RGB values ​​of the first target pixel point A and the first original pixel point a, and the RGB values ​​of the second target pixel point B and the second original pixel point b are all the same, marking a plurality of identical original target two-dimensional dot matrix images as target images;

[0111] If there are multiple target images, the recognition processing module marks the multiple target images as two batches of two-dimensional dot matrix images, sets a second rotation radius R2, establishes temporary rotation coordinates with the second target pixel point B as the coordinate origin, and obtains the third target pixel point C, wherein the acquisition rule of the third target pixel point C is consistent with the acquisition rule of the second target pixel point B;

[0112] After the recognition processing module generates the second target pixel point B and the second target pixel point C, the recognition processing module obtains the second original pixel point b and the third original pixel point c in the two batches of two-dimensional dot matrix images according to the generation rule of the second target pixel point B and the second target pixel point C;

[0113] The recognition processing module also compares the RGB values ​​of the second target pixel point B and the second target pixel point C with the RGB values ​​of the corresponding second original pixel points b and the third original pixel point c respectively;

[0114] When the RGB values ​​of the second target pixel B and the second original pixel b, and the RGB values ​​of the third target pixel C and the third original pixel c are the same, a number of identical original target two-dimensional dot matrix images are marked as secondary target images.

[0115] If there is only one secondary target image, the single target image is fed back to the centralized management module, and the centralized management module determines the original target view image according to the received single target image, and further identifies the category of the printed packaging box to be managed;

[0116] If there are multiple secondary target images, continue the above steps until the multiple target images are one, then feed back a single target image to the centralized management module, and the centralized management module determines the original target view image based on the received single target image, and then identifies the category of the printed packaging box to be managed.

[0117] The above formulas are all calculated by removing dimensions and taking numerical values. The formula is a formula that is closest to the actual situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained by simulating a large amount of data.

[0118] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A printing packaging box management system based on color recognition, characterized in that: include: An image acquisition module for acquiring a first target view image of a printed packaging box to be managed and original target view images of all printed packaging boxes of a target factory; A preprocessing module for performing pixelation processing on the first target view image and the original target view image to obtain a first target two-dimensional dot matrix image and an original target two-dimensional dot matrix image; as well as The recognition processing module receives the first target two-dimensional dot matrix image, establishes a rectangular two-dimensional coordinate system, generates coordinate values ​​for each pixel point P, and marks them as P(x, y), where x is the horizontal coordinate and y is the vertical coordinate; obtains the RGB value of the corresponding pixel point and marks it as S P =(R P ,G P ,B P ); Generate a first target pixel point A, set a first rotation radius R1, and establish a temporary rotation coordinate with the first target pixel point A as the coordinate origin; draw an arc based on the coordinate origin and the first rotation radius R1, and obtain the pending pixel points n on the arc in the first quadrant in sequence according to the horizontal coordinates, where n is an integer, representing the horizontal coordinate of the pending pixel point, n=1,2...N; N represents the total number of the pending pixel points n; Use the calculation formula to calculate the confirmed value QX of the undetermined pixel n and the first target pixel A An ; Among them, the selected value QX An The calculation formula is: Wherein, α is the correction coefficient, 0<α<1; The recognition processing module sets the selected value range (QX0, QX1), and the recognition processing module compares the selected value range (QX0, QX1) with the calculated selected value QX An Compare them one by one; When QX An ∈(QX0, QX1), the corresponding undetermined pixel point n is marked as the second target preliminary point; and the point with the smallest horizontal coordinate among the second target preliminary points is determined as the second target pixel point B, and the RGB value of the second target pixel point B is obtained at the same time, S B =(R B ,G B ,B B ); The recognition processing module obtains a first original pixel point a and a second original pixel point b in the original target two-dimensional dot matrix image according to the generation rule of the first target pixel point A and the second target pixel point B; The recognition processing module is also used to compare the RGB values ​​of the first target pixel point A and the second target pixel point B with the RGB values ​​of the corresponding first original pixel points a and the second original pixel points b respectively; When the RGB values ​​of the first target pixel point A and the first original pixel point a, and the RGB values ​​of the second target pixel point B and the second original pixel point b are all the same, a plurality of identical original target two-dimensional dot matrix images are marked as target images.

2. A printing packaging box management system based on color recognition as claimed in claim 1, characterized in that: The image acquisition module acquires a first target view image of the printed packaging box to be managed through a CCD camera No. 1 connected thereto; The original target view images of all printed packaging boxes of the target factory are collected by a No. 2 CCD camera connected thereto; The model and specifications of the No. 1 CCD camera and the No. 2 CCD camera are the same.

3. A printing packaging box management system based on color recognition as claimed in claim 1, characterized in that: The target graph is one or more; If there is only one target image, the single target image is fed back to the centralized management module, and the centralized management module determines the original target view image according to the received single target image, and further identifies the category of the printed packaging box to be managed.

4. A printing packaging box management system based on color recognition as claimed in claim 3, characterized in that: If there are multiple target images, the recognition processing module marks the multiple target images as two batches of two-dimensional dot matrix images, sets a second rotation radius R2, establishes temporary rotation coordinates with the second target pixel point B as the coordinate origin, and obtains the third target pixel point C, wherein the acquisition rule of the third target pixel point C is consistent with the acquisition rule of the second target pixel point B; After the recognition processing module generates the second target pixel point B and the second target pixel point C, the recognition processing module obtains the second original pixel point b and the third original pixel point c in the two batches of two-dimensional dot matrix images according to the generation rule of the second target pixel point B and the second target pixel point C; The recognition processing module also compares the RGB values ​​of the second target pixel point B and the second target pixel point C with the RGB values ​​of the corresponding second original pixel points b and the third original pixel point c respectively; When the RGB values ​​of the second target pixel B and the second original pixel b, and the RGB values ​​of the third target pixel C and the third original pixel c are the same, a number of identical original target two-dimensional dot matrix images are marked as secondary target images.

5. A printing packaging box management system based on color recognition as claimed in claim 4, characterized in that: If there is only one secondary target image, the single target image is fed back to the centralized management module, and the centralized management module determines the original target view image according to the received single target image, and further identifies the category of the printed packaging box to be managed; If there are multiple secondary target images, continue the above steps until the multiple target images are one, then feed back a single target image to the centralized management module, and the centralized management module determines the original target view image based on the received single target image, and then identifies the category of the printed packaging box to be managed.

6. A printing packaging box management system based on color recognition as claimed in claim 1, characterized in that: It also includes a server connected to the preprocessing module and used to receive and store the original target two-dimensional dot matrix image sent by the preprocessing module; The server is also used to receive a plurality of original target view images sent by the image acquisition module, and to make a one-to-one correspondence between the plurality of original target view images and the original target two-dimensional dot matrix images processed by the pre-processing module; The preprocessing module is also used to send the original target two-dimensional dot matrix image after preprocessing to the server; The image acquisition module is connected to the server and sends the acquired original target view images to the server.

7. A printing packaging box management method of a printing packaging box management system based on color recognition according to any one of claims 1 to 6, characterized in that: The following steps are involved: The image acquisition module acquires the first target view image of the printed packaging box to be managed and the original target view images of all printed packaging boxes of the target factory; The preprocessing module performs pixelation processing on the first target view image and the original target view image to obtain a first target two-dimensional dot matrix image and an original target two-dimensional dot matrix image; The recognition processing module receives the first target two-dimensional dot matrix image, establishes a rectangular two-dimensional coordinate system, generates coordinate values ​​for each pixel point P, and marks them as P(x, y), where x is the horizontal coordinate and y is the vertical coordinate; obtains the RGB value of the corresponding pixel point and marks it as S P =(R P ,G P ,B P ); Generate a first target pixel point A, set a first rotation radius R1, and establish a temporary rotation coordinate with the first target pixel point A as the coordinate origin; draw an arc based on the coordinate origin and the first rotation radius R1, and obtain the pending pixel points n on the arc in the first quadrant in sequence according to the horizontal coordinates, where n is an integer, representing the horizontal coordinate of the pending pixel point, n=1,2...N; N represents the total number of the pending pixel points n; Use the calculation formula to calculate the confirmed value QX of the undetermined pixel n and the first target pixel A An ; Among them, the selected value QX An The calculation formula is: Wherein, α is the correction coefficient, 0<α<1; The recognition processing module sets the selected value range (QX0, QX1), and the recognition processing module compares the selected value range (QX0, QX1) with the calculated selected value QX An Compare them one by one; When QX An ∈(QX0, QX1), the corresponding undetermined pixel point n is marked as the second target preliminary point; and the point with the smallest horizontal coordinate among the second target preliminary points is determined as the second target pixel point B, and the RGB value of the second target pixel point B is obtained at the same time, S B =(R B ,G B ,B B ); The recognition processing module obtains a first original pixel point a and a second original pixel point b in the original target two-dimensional dot matrix image according to the generation rule of the first target pixel point A and the second target pixel point B; The recognition processing module is also used to compare the RGB values ​​of the first target pixel point A and the second target pixel point B with the RGB values ​​of the corresponding first original pixel points a and the second original pixel points b respectively; When the RGB values ​​of the first target pixel point A and the first original pixel point a, and the RGB values ​​of the second target pixel point B and the second original pixel point b are all the same, marking a plurality of identical original target two-dimensional dot matrix images as target images; If there are multiple target images, the recognition processing module marks the multiple target images as two batches of two-dimensional dot matrix images, sets a second rotation radius R2, establishes temporary rotation coordinates with the second target pixel point B as the coordinate origin, and obtains the third target pixel point C, wherein the acquisition rule of the third target pixel point C is consistent with the acquisition rule of the second target pixel point B; After the recognition processing module generates the second target pixel point B and the second target pixel point C, the recognition processing module obtains the second original pixel point b and the third original pixel point c in the two batches of two-dimensional dot matrix images according to the generation rule of the second target pixel point B and the second target pixel point C; The recognition processing module also compares the RGB values ​​of the second target pixel point B and the second target pixel point C with the RGB values ​​of the corresponding second original pixel points b and the third original pixel point c respectively; When the RGB values ​​of the second target pixel B and the second original pixel b, and the RGB values ​​of the third target pixel C and the third original pixel c are the same, marking a number of identical original target two-dimensional dot matrix images as secondary target images; If there is only one secondary target image, the single target image is fed back to the centralized management module, and the centralized management module determines the original target view image according to the received single target image, and further identifies the category of the printed packaging box to be managed; If there are multiple secondary target images, continue the above steps until the multiple target images are one, then feed back a single target image to the centralized management module, and the centralized management module determines the original target view image based on the received single target image, and then identifies the category of the printed packaging box to be managed.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, a printing packaging box management method based on color recognition as described in claim 7 is implemented.

Images