CN117197003A - Multi-condition control carton sample generation method - Google Patents
Multi-condition control carton sample generation method Download PDFInfo
- Publication number
- CN117197003A CN117197003A CN202311465794.4A CN202311465794A CN117197003A CN 117197003 A CN117197003 A CN 117197003A CN 202311465794 A CN202311465794 A CN 202311465794A CN 117197003 A CN117197003 A CN 117197003A
- Authority
- CN
- China
- Prior art keywords
- carton
- image
- legend
- area
- coordinate system
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 60
- 230000000295 complement effect Effects 0.000 claims abstract description 10
- 238000004458 analytical method Methods 0.000 claims abstract description 9
- 230000008569 process Effects 0.000 claims description 16
- 239000002390 adhesive tape Substances 0.000 claims description 13
- 238000004364 calculation method Methods 0.000 claims description 10
- 238000000605 extraction Methods 0.000 claims description 6
- 230000001788 irregular Effects 0.000 claims description 6
- 230000001502 supplementing effect Effects 0.000 claims description 4
- 238000002372 labelling Methods 0.000 abstract 1
- 230000008439 repair process Effects 0.000 description 8
- 238000004040 coloring Methods 0.000 description 3
- 238000010276 construction Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000010422 painting Methods 0.000 description 2
- 230000010076 replication Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000007789 sealing Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Abstract
The invention belongs to the technical field of image generation, and particularly relates to a multi-condition control carton sample generation method, which comprises the steps of obtaining various characteristic information data based on analysis of specific parts of a legend carton, generating a prompt text based on summary of the various characteristic information data, constructing an X-Y-Z three-dimensional coordinate system based on the legend carton, determining specific position relation and area of each part of the legend carton, obtaining a carton reference image, establishing a carton image complement model based on the carton reference image obtained by the X-Y-Z three-dimensional coordinate system constructed by the legend carton and extracting prompt text information, inputting a carton picture to be complemented into the carton image complement model to analyze and complete the carton image, solving the technical problems that the carton image is not completed and a complete system of the carton sample is generated at present by complementing the carton image, changing the image texture does not change the position of the carton, and labeling the position of the carton is not changed and is repeatedly used for one time.
Description
Technical Field
The invention belongs to the field of image generation, and particularly relates to a multi-condition control carton sample generation method.
Background
The digital image processing (Digital Image Processing), (Digital Image Processing) of the image restoration technique, which is also called computer image processing, refers to a process of converting an image signal into a digital signal and processing it by a computer, the first large computer capable of performing a meaningful image processing task appears in the early 6060 s of the 2020 th century, people begin to process graphics and image information by using a computer, the digital image processing is formed as a subject in the early 60 s of the 20 th century, the purpose of the early image processing is to improve the quality of an image, it is an object to improve the visual effect of people, in the image processing, an input is an image of low quality, an output is an image of improved quality, and common image processing methods are image enhancement, image restoration, image encoding, image compression, etc., the image restoration is a processing technique of improving the image quality, and is an image processing research field.
One patent application with publication number 201510177721 discloses a digital repair method for a tomb wall painting image, which belongs to the field of digital image repair, and comprises the steps of decomposing the tomb wall painting image by utilizing a total variation mathematical model to obtain a cartoon image, adding a structural factor item in priority calculation, driving priority calculation of an edge point to be repaired by utilizing the cartoon image, constructing a repair block by taking the point with the largest priority as the center, searching a sample block with the smallest Euclidean distance from the point in a known area, calculating the square sum of the mean pixel difference of the repair block and the sample block, comparing the square sum with a set threshold, adaptively adjusting the size of the repair block until a replication condition is met, carrying out sample block replication and edge update, and finally checking the area to be repaired, and iterating the steps until the repair is completed if the area is not empty.
In the prior art, only a single image is subjected to complement reduction, no better analysis and repair method exists for the image with the overlapped part or the overlapped part, the repair content is unknown, and the position relationship and the characteristics of the image part to be repaired cannot be determined.
Therefore, the invention provides a multi-condition control carton sample generation method.
Disclosure of Invention
In order to overcome the deficiencies of the prior art, at least one technical problem presented in the background art is solved.
The technical scheme adopted for solving the technical problems is as follows: the invention discloses a multi-condition control carton sample generation method, which comprises the following steps: analyzing based on specific parts of the legend carton to obtain various item of characteristic information data, and generating a prompt text based on summarization of the various item of characteristic information data;
step two: constructing an X-Y-Z three-dimensional coordinate system for the legend paper box, constructing the X-Y-Z three-dimensional coordinate system based on the legend paper box, determining the specific position relationship and the area of each part of the legend paper box and obtaining a paper box reference image;
step three: carton reference images obtained based on an X-Y-Z three-dimensional coordinate system constructed by legend cartons and extraction of prompt text information are used for establishing carton image complement models;
step four: and inputting the carton picture to be complemented into a carton image complementing model for analysis, and complementing the carton image.
As a further aspect of the invention: the prompt text generation process comprises the following steps: based on the legend paper box, the legend paper box is disassembled into specific parts such as surfaces, bands, adhesive tapes and center joints, the specific parts such as the surfaces, the bands, the adhesive tapes and the center joints are analyzed to obtain specific part characteristics such as the surfaces, the bands, the adhesive tapes and the center joints, the legend paper box surface is divided into a plurality of reference surfaces, other specific parts of the paper box contained on each reference surface are marked in a one-to-one correspondence mode, and various characteristic information data are obtained and summarized to generate prompt texts.
As a further aspect of the invention: the process for determining the area of each part of the legend carton comprises the following steps:
step one: constructing an X-Y-Z three-dimensional coordinate system, acquiring the position coordinates of each vertex of the legend carton structure based on the constructed X-Y-Z three-dimensional coordinate system of the legend carton, and calculating the position coordinates of each vertex of the legend carton structure to obtain the surface area of each legend carton;
step two: calculating the area of other parts included in each side of the legend carton: firstly, edge vertexes of other parts included in each surface of the legend carton (wherein the edge vertexes are position points convenient for calculating the area) are obtained, then, the side lengths of the other parts included in each surface of the legend carton are obtained through position coordinates of the edge vertexes, and the areas of the other parts included in each surface of the carton are obtained through calculation and analysis of the side lengths.
As a further aspect of the invention: the position relation of each part of the legend paper box is determined, and the specific process is as follows: and calculating to obtain the distances between the edge vertices of the other parts contained in the surfaces of the legend cartons and the reference points by taking the vertices of the legend cartons as the reference points based on the obtained position coordinates of the vertices of the legend cartons and the position coordinates of the edge vertices of the other parts contained in the surfaces of the legend cartons, and determining the position relationship of the other parts contained in the surfaces of the legend cartons.
As a further aspect of the invention: the method for supplementing the carton image comprises the following steps:
the method comprises the following steps: counting and analyzing the number of the surfaces of the cartons related to the carton images to be complemented, calculating the area of the carton images to be complemented, and complementing the carton images according to the extracted prompt text information and the carton reference images to generate carton samples;
the second method is as follows: and acquiring an image cutting outline based on the constructed X-Y two-dimensional coordinate system, and re-complementing the carton image according to the prompt text information and the carton reference image to generate a carton sample.
As a further aspect of the invention: the method comprises the following specific implementation processes: comparing the required full-complement carton images with carton reference images, counting the number of the surfaces of the cartons related to the required full-complement carton images, setting pixel grids, and calculating the areas of the pixel grids of the carton images which do not need to be full-complement;
and (3) obtaining the area of the carton image to be complemented by taking the difference between the obtained pixel grid area of the carton image not to be complemented and the legend carton area, and complementing the carton image based on the obtained area of the carton image to be complemented, the information characteristics of each part contained in the prompt text and the position relation of each part determined in the reference image.
As a further aspect of the invention: the specific implementation process of the method II is as follows: acquiring all edge points of an image part to be complemented in an X-Y two-dimensional coordinate system of the carton image framework to be complemented, selecting four datum points in four sections of the coordinate system, performing line drawing processing on the datum points to obtain a rectangular outline of image cutting, performing area calculation on the cut image part, and re-complementing the image according to the rectangular outline and the reference image of the carton and extraction prompt text information.
The beneficial effects of the invention are as follows:
1. according to the multi-condition control carton sample generation method, the carton is disassembled into specific parts such as the surface, the ribbon, the adhesive tape and the middle seam based on the legend, the specific parts such as the surface, the ribbon, the adhesive tape and the middle seam are analyzed, specific part characteristics such as the surface, the ribbon, the adhesive tape and the middle seam are obtained, the legend carton surface is divided into a plurality of reference surfaces, other specific parts of the carton contained on each reference surface are marked in a one-to-one correspondence mode, various characteristic information data are obtained, prompt texts are generated in a summarizing mode, and basis is provided for specific contents of image completion according to the information data recorded in the prompt texts.
2. According to the multi-condition control carton sample generation method, through an X-Y-Z three-dimensional coordinate system based on construction of a legend carton, position coordinates of each vertex of the legend carton structure and each part of edge vertex are obtained, position coordinates of each item of the legend carton structure are calculated to obtain each surface area of the legend carton and other part areas contained in each surface of the carton, the mutual position relation of other parts contained in each surface of the legend carton is determined, and specific position representation is carried out on each part of an image to be completed.
3. The carton image complementing model is established based on the carton reference image and the extraction prompt text information obtained by the X-Y-Z three-dimensional coordinate system constructed by the legend carton, the carton image to be complemented is input into the carton image complementing model for analysis, the area of the carton image to be complemented is accurately obtained by calculating the pixel grid area through fixed integral adopted in the process of complementing the carton image by two methods, and the more complex complementing steps are greatly simplified by adopting the image contour cutting and the re-complementing method, so that the convenience of complementing the carton image is improved.
4. The method has the advantages that the method is capable of complementing the carton images by extracting the prompt text information and the legend carton reference images, controlling the conditions and changing the textures of the images without changing the positions of the cartons, marking the positions of the cartons without changing, marking once and utilizing for multiple times.
Drawings
The invention is further described below with reference to the accompanying drawings.
FIG. 1 is a flow chart of an embodiment of the present invention;
FIG. 2 is a schematic illustration of a method one of completing a carton image according to an embodiment of the invention;
FIG. 3 is a schematic diagram of a second method for supplementing an image of a carton according to an embodiment of the present invention;
fig. 4 is a schematic diagram of a method of calculating the area of the completed carton according to an embodiment of the invention.
Detailed Description
The invention is further described in connection with the following detailed description in order to make the technical means, the creation characteristics, the achievement of the purpose and the effect of the invention easy to understand.
As shown in fig. 1, a multi-condition controlled carton sample generation method according to an embodiment of the present invention includes:
step one: based on the legend, the carton is disassembled into specific parts such as a surface, a ribbon, a tape, a middle seam and the like, the specific part characteristics such as the surface, the ribbon, the tape, the middle seam and the like are obtained by analyzing the specific parts such as the surface, the ribbon, the tape, the middle seam and the like, the specific part characteristics such as the surface, the ribbon, the tape, the middle seam and the like are marked as A, B, C, D and E, the legend carton surface is divided into 7 faces (wherein the carton sealing faces are double-layered), the machine is marked as Ai1, ai2, ai3, ai4, ai5, ai6 and Ai7, the 7 faces of the legend paper box are taken as reference faces, the parts of the binding tape on different reference faces are marked as Bi1, bi2, bi3, bi4, bi5 and Bi6, the parts of the binding tape on different reference faces are marked as Ci1, ci2, ci3, ci4, ci5 and Ci6, and other parts of the legend paper box are marked as the same principle, so that the machine is not described;
each item of feature information data, such as Ai1-Bi1-ci1, ai2-Bi2-ci2, ai3-Bi3-ci3, is obtained with one-to-one correspondence of the obtained parts on the different reference surfaces.
Summarizing the acquired characteristic information data to generate a prompt text;
step two: constructing an X-Y-Z three-dimensional coordinate system by taking the top point of the bottom surface of the legend paper box as an origin, wherein a X, Y, Z axis is respectively overlapped with three side lines of the length, the width and the height of the legend paper box, constructing the X-Y-Z three-dimensional coordinate system based on the legend paper box, and determining the specific position relation and the area of each part of the legend paper box, wherein the specific process comprises the following steps:
s101, obtaining each vertex of the legend carton according to the legend carton, marking each vertex of the legend carton as Si1, si2, si3, si4, si5, obtaining the position of each vertex of the legend carton structure based on the X-Y-Z three-dimensional coordinate system of the construction of the legend carton, marking the position of each vertex of the legend carton as Si1 (X1, Y1, Z1), si2 (X2, Y2, Z2), si3 (X3, Y3, Z3), si4 (X4, Y4, Z4), and calculating the area of each face of the legend carton according to the position coordinates obtained each vertex, for example:
presetting Si1 (x 1, y1, z 1), si2 (x 2, y2, z 2), si3 (x 3, y3, z 3) and Si4 (x 4, y4, z 4) as four vertexes of the upper surface of the carton, wherein the area is calculated according to coordinates of the vertexes, specifically, the length and the width of the upper surface are obtained through difference of xi, yi and zi among the points, and the obtained length and the obtained width of the upper surface are multiplied to obtain the area of the upper surface;
s102, calculating the area of other parts included in each side of the legend carton: firstly, edge vertexes of other parts included in each face of the legend carton (wherein the edge vertexes are position points convenient for calculating the area) are obtained, then, the side lengths of the other parts included in each face of the legend carton are obtained through position coordinates of the edge vertexes, and the area is obtained through calculation and analysis of the side lengths;
s103, calculating to obtain distances between edge vertices of other parts contained in the legend carton and the reference points based on the obtained position coordinates of the vertices of the legend carton and the position coordinates of the edge vertices of other parts contained in the faces of the legend carton, and determining the positions of the other parts contained in the faces of the legend carton, wherein the position relationship of the parts can be specifically represented by selecting different nearest reference points, for example, assuming that a part of the upper surface of the carton needs to be subjected to image complementation (the image area of the complementation part is smaller than the upper surface area of the carton), selecting the reference point nearest to the complementation part, calculating the distances between the selected reference point and the edge vertices of the required complementation part, and representing the position relationship by the calculated distances and the position coordinates;
summarizing the information data obtained in S101, S102 and S103 to form a carton reference image;
step three: obtaining a carton reference image and extracting prompt text information based on an X-Y-Z three-dimensional coordinate system constructed by a legend carton, and establishing a carton image complement model;
step four: inputting a carton picture to be complemented into a carton image complementing model for analysis, and complementing the carton image, wherein the specific complementing process is as follows:
the method comprises the following steps: comparing the image to be complemented with the carton image, counting the number of each surface of the carton related to the image to be complemented, and supposing that the image complementation relates to the surfaces of the cartons Ai1, ai2 and Ai3, and illustrating by Ai 1;
setting the pixel grid of the Ai1 surface, calculating the area of the pixel grid which does not need to be complemented by the Ai1 surface, and obtaining the area of the pixel grid which needs to be complemented by the difference between the area of the pixel grid which does not need to be complemented by the Ai1 surface and the original complete Ai1 area, (wherein the area of the pixel grid which needs to be complemented is the area of the carton image which needs to be complemented);
the pixel grid area calculation process without complement to the Ai1 surface comprises the following steps:
counting the number of colored complete pixel grids, calculating the areas of all the complete pixel grids, and marking the areas as Li;
for incomplete colored pixel gridPerforming area calculation, as shown in fig. 2, selecting an irregular part edge point in a single incomplete coloring pixel grid, marking the irregular part edge point as Di1, di2, di3, di4 and Di6, performing curve connection on the marked edge point, taking an incomplete pixel grid as an example, constructing an X-Y coordinate system by taking a pixel grid edge line as an axis, and coloring a part in the coordinate system by the formula:the area is calculated as shown in fig. 4, where the fixed integral geometry has the following meaning: when (when)When not less than 0, integrateGeometrically represented by y =Curved trapezoid area surrounded by x=a, x=b and x axis;
when (when)When less than or equal to 0, integrateGeometrically represented by y =A curved trapezoid surrounded by x=a, x=b and the x axis is positioned below the x axis;
counting the number of all incomplete coloring pixel grids, obtaining the areas of all incomplete pixel grids, and marking the areas as Ki;
summing the obtained areas Li of all complete pixel grids with the areas and Ki of all incomplete pixel grids to obtain the area of the pixel grid which is not required to be complemented by the Ai1 surface, and obtaining the area of the image which is required to be complemented by the Ai1 surface by making a difference between the area of the pixel grid which is not required to be complemented by the Ai1 surface and the area of the complete Ai1 surface recorded by the original reference image;
the above-mentioned only calculates the image area to be complemented of the Ai1 surface, and other partial areas to be complemented contained in the Ai1 surface are obtained in the same way as the above-mentioned, and are not described one by one here;
the method comprises the steps of filling a carton image by using the obtained area of an Ai1 surface needing to be filled with the image, the area of other parts contained in the Ai1 surface needing to be filled with the image, and the position information of each part recorded by a legend carton reference image and the carton part characteristics recorded by a prompt text to generate a carton sample;
the second method is as follows: based on an image needing to be complemented in a complex and irregular manner, as shown in fig. 3, an X-Y two-dimensional coordinate system is constructed on a carton image needing to be complemented, an origin of the X-Y two-dimensional coordinate system is positioned at the center of a part needing to be complemented in the carton image, all edge points needing to be complemented in the part needing to be complemented are obtained, an edge point with the farthest vertical distance from the X axis is selected in each of the upper interval and the lower interval of the X axis, the selected edge points are marked as Zi1 and Zi2, zi1 and Zi2 are used as datum points, two straight lines respectively passing through the datum points Zi1 and Zi2 and parallel to the X axis are made, one edge point with the farthest vertical distance from the Y axis is selected in each of the left interval and the right interval of the Y axis, the selected edge points are marked as Zi3 and Zi4, and Zi3 are used as datum points, and two straight lines respectively passing through the datum points Zi3 and Zi4 and parallel to the Y axis are made;
the method comprises the steps of intersecting four straight lines to form a rectangular outline, cutting out an image according to the rectangular outline, cutting out an image part needing to be complemented, then, re-complementing the image according to the regular rectangular outline, firstly, calculating the size of the rectangular outline according to the position coordinates of four selected datum points in a coordinate system, and then, according to a carton reference image and extraction prompt text information, re-complementing the image according to the rectangular outline, so as to generate a carton sample.
The method comprises the steps of firstly decomposing an illustration paper box, splitting the illustration paper box into parts such as surfaces, bands, adhesive tapes and center joints, taking the disassembled paper box surfaces as reference surfaces, then correspondingly marking other parts such as the bands, the adhesive tapes and the center joints contained on each reference surface, and summarizing marked contents to form a prompt text of a paper box image;
based on the legend paper box, an X-Y-Z three-dimensional coordinate system is built in the legend paper box to obtain each vertex of the paper box, each vertex is used as a datum point, the area of each surface of the paper box is obtained through the position coordinate calculation of each datum point, the areas of other parts such as a paper box ribbon, an adhesive tape and a center seam can also be obtained through the calculation of the X-Y-Z three-dimensional coordinate system, and the position relation among other parts such as the ribbon, the adhesive tape and the center seam contained in each surface of the paper box can be determined through the position coordinates in the coordinate system;
and (3) obtaining a carton reference image and extracting prompt text information based on an X-Y-Z three-dimensional coordinate system constructed by the legend carton, establishing a carton image complement model, and complementing the inputted carton image to be complemented through the carton image complement model to generate a carton sample.
The foregoing has shown and described the basic principles, principal features and advantages of the invention. It will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, and that the above embodiments and descriptions are merely illustrative of the principles of the present invention, and various changes and modifications may be made without departing from the spirit and scope of the invention, which is defined in the appended claims. The scope of the invention is defined by the appended claims and equivalents thereof.
Claims (7)
1. A multi-condition control carton sample generation method is characterized in that: the method comprises the following steps: disassembling the paper box, marking and corresponding each specific part of the disassembled paper box, obtaining each item of characteristic information data, and summarizing each item of characteristic information data to generate a prompt text;
step two: constructing an X-Y-Z three-dimensional coordinate system for the legend carton, determining specific coordinates of vertexes of each part of the legend carton based on the X-Y-Z three-dimensional coordinate system, determining specific position relations and area sizes of each part through the specific coordinates of the vertexes of each part, and obtaining a carton reference image, wherein the area size determining process comprises the following steps:
constructing an X-Y-Z three-dimensional coordinate system, acquiring the position coordinates of each vertex of the legend carton structure based on the constructed X-Y-Z three-dimensional coordinate system of the legend carton, and calculating the position coordinates of each vertex of the legend carton structure to obtain the surface area of each legend carton;
calculating the area of other parts included in each side of the legend carton: firstly, edge vertexes of other parts included in each surface of the legend carton are obtained, wherein the edge vertexes are position points convenient for calculating the area, then the side lengths of the other parts included in each surface of the legend carton are obtained through position coordinates of the edge vertexes, and the areas of the other parts included in each surface of the carton are obtained through calculation and analysis of the side lengths;
step three: carton reference images obtained based on an X-Y-Z three-dimensional coordinate system constructed by legend cartons and extraction of prompt text information are used for establishing carton image complement models;
step four: inputting a carton picture to be complemented into a carton image complement model for analysis, wherein the method comprises the following steps: dividing a damaged area region of the carton image into a regular area region and an irregular area region by a constructed coordinate system and setting pixel grids, marking edge points of the area region in the carton damaged region, calculating the area of the regular area region by calculating the area and the number of single pixel grids, calculating the area of the irregular area region by determining integral points, overlapping and summing the area of the regular area region and the area of the irregular area region of the carton to obtain the damaged area region of the carton image, complementing the carton image and generating a carton sample according to the damaged area region of the carton image, a prompt text and a reference image of the carton;
the second method is as follows: and (3) constructing an X-Y two-dimensional coordinate system in the carton image to be complemented, obtaining all edge points of the image part to be complemented, selecting four datum points in four sections of the coordinate system, processing the datum points to obtain a rectangular outline of image cutting, calculating the area of the cut image part, and re-complementing the image according to the rectangular outline and the reference image of the carton and extraction prompt text information.
2. A multi-condition controlled carton sample generation method according to claim 1, wherein: the prompt text generation process comprises the following steps: based on the legend paper box, the legend paper box is disassembled into specific parts such as surfaces, bands, adhesive tapes and center joints, the specific parts such as the surfaces, the bands, the adhesive tapes and the center joints are analyzed to obtain specific part characteristics such as the surfaces, the bands, the adhesive tapes and the center joints, the legend paper box surface is divided into a plurality of reference surfaces, other specific parts of the paper box contained on each reference surface are marked in a one-to-one correspondence mode, and various characteristic information data are obtained and summarized to generate prompt texts.
3. A multi-condition controlled carton sample generation method according to claim 1, wherein: the specific process for determining the position relation of each part of the legend carton comprises the following steps: and calculating to obtain the distances between the edge vertices of the other parts contained in each surface of the legend carton and the reference points by taking the vertices of the legend carton as the reference points based on the obtained position coordinates of the vertices of the legend carton and the position coordinates of the edge vertices of the other parts contained in each surface of the legend carton, and determining the position relationship of the other parts contained in each surface of the legend carton.
4. A multi-condition controlled carton sample generation method according to claim 1, wherein: the method for supplementing the carton image comprises the following steps:
and counting and analyzing the number of the surfaces of the cartons related to the carton images to be complemented, calculating the area of the carton images to be complemented, and complementing the carton images according to the extracted prompt text information and the carton reference images to generate carton samples.
5. A multi-condition controlled carton sample generation method according to claim 1, wherein: the method for supplementing the carton image further comprises the following steps:
and acquiring an image clipping contour based on the constructed X-Y two-dimensional coordinate system, and re-complementing the carton image according to the prompt text information and the reference image through the image clipping contour to generate a carton sample.
6. The multi-condition controlled carton sample generation method of claim 5, wherein: the method comprises the following specific implementation processes: and comparing the required full-complement carton images with the carton reference images, counting the number of the surfaces of the cartons related to the required full-complement carton images, setting the pixel grid, and calculating the pixel grid area of the carton images which do not need to be full.
7. The multi-condition controlled carton sample generation method of claim 5, wherein: the specific implementation process of the method further comprises the following steps: and (3) subtracting the pixel grid area of the obtained carton image which does not need to be complemented from the reference image area of the carton to obtain the area of the carton image which needs to be complemented, and complementing the carton image based on the obtained area of the carton image which needs to be complemented, the information characteristics of each part contained in the prompt text and the position relation of each part determined in the reference image.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311465794.4A CN117197003B (en) | 2023-11-07 | 2023-11-07 | Multi-condition control carton sample generation method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311465794.4A CN117197003B (en) | 2023-11-07 | 2023-11-07 | Multi-condition control carton sample generation method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN117197003A true CN117197003A (en) | 2023-12-08 |
CN117197003B CN117197003B (en) | 2024-02-27 |
Family
ID=88994675
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202311465794.4A Active CN117197003B (en) | 2023-11-07 | 2023-11-07 | Multi-condition control carton sample generation method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117197003B (en) |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4969595A (en) * | 1989-07-21 | 1990-11-13 | Oscar Mayer Foods Corporation | Convertible distribution and carry-out carton |
CN108564527A (en) * | 2018-04-04 | 2018-09-21 | 百度在线网络技术(北京)有限公司 | The method and device of the completion of panorama sketch content and reparation based on neural network |
CN110390258A (en) * | 2019-06-05 | 2019-10-29 | 东南大学 | Image object three-dimensional information mask method |
US20210110194A1 (en) * | 2019-10-14 | 2021-04-15 | Hangzhou Dianzi University | Method for automatic extraction of data from graph |
WO2022052620A1 (en) * | 2020-09-10 | 2022-03-17 | 北京达佳互联信息技术有限公司 | Image generation method and electronic device |
WO2023044605A1 (en) * | 2021-09-22 | 2023-03-30 | 深圳先进技术研究院 | Three-dimensional reconstruction method and apparatus for brain structure in extreme environments, and readable storage medium |
CN116402713A (en) * | 2023-04-04 | 2023-07-07 | 国网甘肃省电力公司电力科学研究院 | Electric three-dimensional point cloud completion method based on two-dimensional image and geometric shape |
CN116703895A (en) * | 2023-08-02 | 2023-09-05 | 杭州灵西机器人智能科技有限公司 | Small sample 3D visual detection method and system based on generation countermeasure network |
WO2023207963A1 (en) * | 2022-04-29 | 2023-11-02 | 北京字跳网络技术有限公司 | Image processing method and apparatus, electronic device, and storage medium |
-
2023
- 2023-11-07 CN CN202311465794.4A patent/CN117197003B/en active Active
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4969595A (en) * | 1989-07-21 | 1990-11-13 | Oscar Mayer Foods Corporation | Convertible distribution and carry-out carton |
CN108564527A (en) * | 2018-04-04 | 2018-09-21 | 百度在线网络技术(北京)有限公司 | The method and device of the completion of panorama sketch content and reparation based on neural network |
CN110390258A (en) * | 2019-06-05 | 2019-10-29 | 东南大学 | Image object three-dimensional information mask method |
US20210110194A1 (en) * | 2019-10-14 | 2021-04-15 | Hangzhou Dianzi University | Method for automatic extraction of data from graph |
WO2022052620A1 (en) * | 2020-09-10 | 2022-03-17 | 北京达佳互联信息技术有限公司 | Image generation method and electronic device |
WO2023044605A1 (en) * | 2021-09-22 | 2023-03-30 | 深圳先进技术研究院 | Three-dimensional reconstruction method and apparatus for brain structure in extreme environments, and readable storage medium |
WO2023207963A1 (en) * | 2022-04-29 | 2023-11-02 | 北京字跳网络技术有限公司 | Image processing method and apparatus, electronic device, and storage medium |
CN116402713A (en) * | 2023-04-04 | 2023-07-07 | 国网甘肃省电力公司电力科学研究院 | Electric three-dimensional point cloud completion method based on two-dimensional image and geometric shape |
CN116703895A (en) * | 2023-08-02 | 2023-09-05 | 杭州灵西机器人智能科技有限公司 | Small sample 3D visual detection method and system based on generation countermeasure network |
Non-Patent Citations (4)
Title |
---|
HAORUO ZHANG,QIXIN CAO: ""Holistic and local patch framework for 6D object pose estimation in RGB-D images"", 《COMPUTER VISION AND IMAGE UNDERSTANDING》 * |
刘昶,王玉晗,张鑫,刘丽萍: "" 基于投影轮廓的纸箱尺寸单目测量"", 《沈阳理工大学学报》, vol. 37, no. 05 * |
唐露新; 张宇维; 宋有聚; 王小桂; 于丽敏: ""基于形态和小波的低纹理图序列高光修复研究"", 《中国测试》, vol. 45, no. 04 * |
肖海鸿,吴秋遐, 李玉琼,康文雄: ""三维补全关键技术研究综述"", 《光学精密工程》, vol. 31, no. 05 * |
Also Published As
Publication number | Publication date |
---|---|
CN117197003B (en) | 2024-02-27 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US20200258301A1 (en) | Generating a customized three-dimensional mesh from a scanned object | |
KR101955035B1 (en) | Method for designing a geometrical three-dimensional modeled object | |
KR102040988B1 (en) | Creating a surface from a plurality of 3d curves | |
US9158297B2 (en) | Computing device and method for generating measurement program of product | |
US20050093875A1 (en) | Synthesis of progressively-variant textures and application to arbitrary surfaces | |
CN109816669A (en) | A kind of improvement Mask R-CNN image instance dividing method identifying power equipments defect | |
EP1082703A1 (en) | Generation of three-dimensional, textured models | |
JPS63145577A (en) | Pattern data generating method | |
CN102663800A (en) | City building complex and rendering method considering city image | |
US10832376B2 (en) | Generating enhanced digital content using piecewise parametric patch deformations | |
Zhang et al. | Dual contouring for domains with topology ambiguity | |
US10706500B2 (en) | Generating enhanced digital content using piecewise parametric patch deformations | |
US20200098086A1 (en) | Generating enhanced digital content using piecewise parametric patch deformations | |
CN113536617B (en) | Method for rapidly generating fine finite element model with complex structure | |
CN114429535A (en) | Smooth cutting method and device for triangular mesh curved surface | |
CN117197003B (en) | Multi-condition control carton sample generation method | |
Shesh et al. | Efficient and dynamic simplification of line drawings | |
CN110543733B (en) | Method and device for metal plate ordering processing, and equipment and storage medium thereof | |
CN110533714A (en) | Method and system based on image processing techniques detection target object maximum inscribed circle | |
Melero et al. | On the Interactive 3D Reconstruction of Iberian Vessels. | |
CN108763767A (en) | Big data quantity IGS industry pattern POLYGON conversion methods towards VR engines | |
US20040164982A1 (en) | Method and apparatus for editing three-dimensional model, and computer readable medium | |
Li et al. | Enhancing extraction of two-dimensional engineering drawings from three-dimensional data of existing buildings | |
JPH06168223A (en) | Pipe network analyzing device, data display device for the same and graphic generating method | |
CN117272495B (en) | Image and data organization, fusion loading and display method and system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |