CN113610851A - Packaging decoration quality inspection method based on machine vision - Google Patents
Packaging decoration quality inspection method based on machine vision Download PDFInfo
- Publication number
- CN113610851A CN113610851A CN202111177798.3A CN202111177798A CN113610851A CN 113610851 A CN113610851 A CN 113610851A CN 202111177798 A CN202111177798 A CN 202111177798A CN 113610851 A CN113610851 A CN 113610851A
- Authority
- CN
- China
- Prior art keywords
- image
- information
- decoration
- obtaining
- product
- 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
- 238000005034 decoration Methods 0.000 title claims abstract description 221
- 238000004806 packaging method and process Methods 0.000 title claims abstract description 104
- 238000007689 inspection Methods 0.000 title claims abstract description 71
- 238000000034 method Methods 0.000 title claims abstract description 63
- 238000011156 evaluation Methods 0.000 claims abstract description 119
- 238000013441 quality evaluation Methods 0.000 claims abstract description 58
- 238000007781 pre-processing Methods 0.000 claims abstract description 26
- 238000000605 extraction Methods 0.000 claims abstract description 15
- 239000000203 mixture Substances 0.000 claims description 39
- 230000005540 biological transmission Effects 0.000 claims description 23
- 230000009467 reduction Effects 0.000 claims description 15
- 230000004931 aggregating effect Effects 0.000 claims description 7
- 238000004590 computer program Methods 0.000 claims description 7
- 238000012549 training Methods 0.000 claims description 6
- 239000005022 packaging material Substances 0.000 abstract description 12
- 238000001514 detection method Methods 0.000 abstract description 7
- 230000000694 effects Effects 0.000 description 19
- 230000008569 process Effects 0.000 description 12
- 230000007613 environmental effect Effects 0.000 description 8
- 238000005516 engineering process Methods 0.000 description 7
- 238000012545 processing Methods 0.000 description 7
- 238000012986 modification Methods 0.000 description 6
- 230000004048 modification Effects 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 238000001303 quality assessment method Methods 0.000 description 5
- 238000004458 analytical method Methods 0.000 description 4
- 238000013461 design Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 238000004519 manufacturing process Methods 0.000 description 4
- 238000003062 neural network model Methods 0.000 description 4
- 239000007921 spray Substances 0.000 description 3
- 230000003044 adaptive effect Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 230000001737 promoting effect Effects 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000003925 brain function Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010438 heat treatment Methods 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000005312 nonlinear dynamic Methods 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30168—Image quality inspection
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Biophysics (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Software Systems (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Physics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Quality & Reliability (AREA)
- Auxiliary Devices For And Details Of Packaging Control (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The application discloses a packaging decoration quality inspection method based on machine vision, which comprises the following steps: acquiring a first product decoration image through a first image sensor, inputting the first product decoration image into a first preprocessing module, and acquiring a first image set; sequentially performing feature extraction on the first image set to obtain a first image feature set; obtaining a first display surface according to the first product; constructing a first plane coordinate system based on the first display surface to obtain first image position identification information; obtaining a first evaluation result based on the first decoration quality evaluation model; obtaining a first evaluation threshold value, and judging whether the first evaluation result meets the first evaluation threshold value; when the first evaluation result satisfies the first evaluation threshold, first qualified label information is identified for the first product decoration image. The technical problems of low efficiency, wrong detection and missed detection of the product packaging materials and the printing and decorating quality in the prior art are solved.
Description
Technical Field
The application relates to the field of artificial intelligence, in particular to a packaging decoration quality inspection method based on machine vision.
Background
Sales packaging and transport packaging are responsible for different tasks in commodity circulation, with transport packaging mainly serving as a bridge for production and sales, and sales packaging mainly serving as a medium for sales and consumption. The sales package decoration is an advertisement that can be seen everywhere on the market, is a tool for directly transmitting information to the existing market and the potential market, is a powerful weapon for improving the competitiveness of the commodity, and is a typical mode for promoting marketing. The effect of a successful sales package on increasing sales and price is undoubtedly enormous. In actual production, the quality inspection of packaging decoration is carried out manually, the time is long, the inspection efficiency is low, and meanwhile, the phenomenon of missing inspection easily occurs manually. The design and development of a system for intelligently detecting the packaging decoration quality and effect have important practical significance.
In the process of implementing the technical solution in the embodiment of the present application, the inventor of the present application finds that the above-mentioned technology has at least the following technical problems:
the inspection efficiency of the product packaging material and the printing decoration quality is low, and the technical problems of wrong inspection and missed inspection exist in the prior art.
Disclosure of Invention
The application aims to provide a packaging decoration quality inspection method based on machine vision, which is used for solving the technical problems of low inspection efficiency, wrong inspection and missed inspection of product packaging materials and printing decoration quality in the prior art.
In view of the above problems, the embodiments of the present application provide a packaging decoration quality inspection method based on machine vision.
In a first aspect, the present application provides a machine vision-based package decoration quality inspection method implemented by a machine vision-based package decoration quality inspection system, wherein the method comprises: acquiring a first product decoration image through a first image sensor, inputting the first product decoration image into a first preprocessing module, and acquiring a first image set, wherein the first image set is a multi-dimensional acquisition result; sequentially performing feature extraction on the first image set to obtain a first image feature set, wherein the first image feature set comprises a first appearance feature subset and a first composition feature subset; obtaining a first display surface according to the first product, wherein the first display surface is a display area of the first image set; constructing a first plane coordinate system based on the first display surface, and carrying out position identification on the first image set to obtain first image position identification information; inputting the first appearance feature subset, the first composition feature subset and the first image position identification information into a first decoration quality evaluation model to obtain a first evaluation result; obtaining a first evaluation threshold value, and judging whether the first evaluation result meets the first evaluation threshold value; when the first evaluation result satisfies the first evaluation threshold, first qualified label information is identified for the first product decoration image.
In another aspect, the present application further provides a machine vision-based package decoration quality inspection system for performing a machine vision-based package decoration quality inspection method according to the first aspect, wherein the system comprises: a first obtaining unit: the first obtaining unit is used for collecting a first product decoration image through a first image sensor, inputting the first product decoration image into a first preprocessing module, and obtaining a first image set, wherein the first image set is a multi-dimensional collecting result; a second obtaining unit: the second obtaining unit is used for sequentially performing feature extraction on the first image set to obtain a first image feature set, wherein the first image feature set comprises a first appearance feature subset and a first composition feature subset; a third obtaining unit: the third obtaining unit is configured to obtain a first display surface according to the first product, where the first display surface is a display area of the first image set; a fourth obtaining unit: the fourth obtaining unit is configured to construct a first planar coordinate system based on the first display surface, perform position identification on the first image set, and obtain first image position identification information; a fifth obtaining unit: the fifth obtaining unit is used for inputting the first appearance feature subset, the first composition feature subset and the first image position identification information into a first decoration quality evaluation model to obtain a first evaluation result; a first judgment unit: the first judging unit is used for obtaining a first evaluation threshold value and judging whether the first evaluation result meets the first evaluation threshold value; a first identification unit: the first identification unit is used for identifying first qualified label information for the first product decoration image when the first evaluation result meets the first evaluation threshold value.
In a third aspect, embodiments of the present application further provide a machine vision-based packaging decoration quality inspection system, including a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor implements the steps of the method according to the first aspect when executing the program.
One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:
1. acquiring a first product decoration image through a first image sensor, inputting the first product decoration image into a first preprocessing module, and acquiring a first image set, wherein the first image set is a multi-dimensional acquisition result; sequentially performing feature extraction on the first image set to obtain a first image feature set, wherein the first image feature set comprises a first appearance feature subset and a first composition feature subset; obtaining a first display surface according to the first product, wherein the first display surface is a display area of the first image set; constructing a first plane coordinate system based on the first display surface, and carrying out position identification on the first image set to obtain first image position identification information; inputting the first appearance feature subset, the first composition feature subset and the first image position identification information into a first decoration quality evaluation model to obtain a first evaluation result; obtaining a first evaluation threshold value, and judging whether the first evaluation result meets the first evaluation threshold value; when the first evaluation result satisfies the first evaluation threshold, first qualified label information is identified for the first product decoration image. The quality of the product packaging material and the quality of the printing decoration are intelligently monitored in real time based on a computer vision technology, so that the unqualified packaging decoration is identified and timely adjusted, the packaging decoration quality inspection efficiency and the quality inspection accuracy are further improved, and the technical effect that the product packaging decoration meets the requirements is finally ensured.
2. And intelligently analyzing points of the product packaging decoration which do not meet the related requirements of the first evaluation threshold, and adaptively adjusting the equipment printing process parameters based on the corresponding process steps, so that the standard level of the subsequent product packaging decoration in the corresponding printing equipment is guaranteed, and the technical effect of intelligently adjusting the product packaging decoration process parameters is achieved.
3. The model parameters corresponding to each factory are obtained by aggregating parameter data provided by a plurality of factories, one model parameter is finally determined based on the plurality of model parameters, so that the model constructed based on the final model parameter is more reasonable, the evaluation result is more accurate and effective, and meanwhile, the model parameters provided by each factory are encrypted, thereby ensuring that the model parameters of each factory are not leaked.
The foregoing description is only an overview of the technical solutions of the present application, and the present application can be implemented according to the content of the description in order to make the technical means of the present application more clearly understood, and the following detailed description of the present application is given in order to make the above and other objects, features, and advantages of the present application more clearly understandable.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only exemplary, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
FIG. 1 is a schematic flow chart of a machine vision-based packaging decoration quality inspection method according to an embodiment of the present application;
fig. 2 is a schematic flow chart illustrating the adjustment of the first printing device by the first parameter adjustment information in the machine vision-based packaging decoration quality inspection method according to the embodiment of the present application;
fig. 3 is a schematic flow chart of inputting a primary image noise and the primary noise-reduced image set into the first preprocessing module to match a second noise-reduction channel in the packaging decoration quality inspection method based on machine vision according to the embodiment of the present application, so as to obtain the first image set;
fig. 4 is a schematic flow chart illustrating a process of updating the first initial model by using the first encryption parameter to obtain the first decoration quality evaluation model in a machine vision-based packaging decoration quality inspection method according to an embodiment of the present application;
FIG. 5 is a schematic structural diagram of a machine vision based packaging decoration quality inspection system according to an embodiment of the present application;
fig. 6 is a schematic structural diagram of an exemplary electronic device according to an embodiment of the present application.
Description of reference numerals:
a first obtaining unit 11, a second obtaining unit 12, a third obtaining unit 13, a fourth obtaining unit 14, a fifth obtaining unit 15, a first judging unit 16, a first identification unit 17, a bus 300, a receiver 301, a processor 302, a transmitter 303, a memory 304, and a bus interface 305.
Detailed Description
The embodiment of the application provides a packaging decoration quality inspection method based on machine vision, and solves the technical problems that in the prior art, the inspection efficiency of product packaging materials and printing decoration quality is low, and error detection and omission detection exist simultaneously. The quality of the product packaging material and the quality of the printing decoration are intelligently monitored in real time based on a computer vision technology, so that the unqualified packaging decoration is identified and timely adjusted, the packaging decoration quality inspection efficiency and the quality inspection accuracy are further improved, and the technical effect that the product packaging decoration meets the requirements is finally ensured.
In the following, the technical solutions in the embodiments of the present application will be clearly and completely described with reference to the accompanying drawings, and it is to be understood that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments of the present application, and it should be understood that the present application is not limited by the example embodiments described herein. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the present application without making any creative effort, shall fall within the protection scope of the present application. It should be further noted that, for the convenience of description, only some but not all of the elements relevant to the present application are shown in the drawings.
The technical scheme provided by the application has the following general idea:
the application provides a machine vision-based packaging decoration quality inspection method, which is applied to a machine vision-based packaging decoration quality inspection system, wherein the method comprises the following steps: acquiring a first product decoration image through a first image sensor, inputting the first product decoration image into a first preprocessing module, and acquiring a first image set, wherein the first image set is a multi-dimensional acquisition result; sequentially performing feature extraction on the first image set to obtain a first image feature set, wherein the first image feature set comprises a first appearance feature subset and a first composition feature subset; obtaining a first display surface according to the first product, wherein the first display surface is a display area of the first image set; constructing a first plane coordinate system based on the first display surface, and carrying out position identification on the first image set to obtain first image position identification information; inputting the first appearance feature subset, the first composition feature subset and the first image position identification information into a first decoration quality evaluation model to obtain a first evaluation result; obtaining a first evaluation threshold value, and judging whether the first evaluation result meets the first evaluation threshold value; when the first evaluation result satisfies the first evaluation threshold, first qualified label information is identified for the first product decoration image.
Having thus described the general principles of the present application, various non-limiting embodiments thereof will now be described in detail with reference to the accompanying drawings.
Example one
Referring to fig. 1, an embodiment of the present application provides a machine vision-based packaging decoration quality inspection method, where the method is applied to a machine vision-based packaging decoration quality inspection system, and the method specifically includes the following steps:
step S100: acquiring a first product decoration image through a first image sensor, inputting the first product decoration image into a first preprocessing module, and acquiring a first image set, wherein the first image set is a multi-dimensional acquisition result;
particularly, the packaging decoration quality inspection method based on the machine vision is applied to the packaging decoration quality inspection system based on the machine vision, and the quality of a product packaging material and the quality of printing decoration can be intelligently monitored in real time based on a computer vision technology, so that the packaging decoration quality inspection efficiency and the quality inspection accuracy are improved, and the product packaging decoration is finally ensured to meet the requirements. The sensor is a detection device which can sense the measured information and convert the sensed information into an electric signal or other information in a required form according to a certain rule to output so as to meet the requirements of information transmission, processing, storage, display, recording, control and the like. The package decoration refers to the modeling and surface design of the package, and is decorated and beautified on the basis of scientific and reasonable design, so that each element of the appearance, patterns, colors, characters, trademark and brand of the package forms an artistic whole, and the functions of transmitting commodity information, expressing commodity characteristics, publicizing commodities, beautifying commodities, promoting sales, facilitating consumption and the like are achieved. The first image sensor is a sensor device for capturing images of product packaging and decoration in real time. Based on the first image sensor, multi-angle and multi-distance acquisition is carried out on the product packaging decoration image, noise reduction preprocessing is carried out on the acquired product packaging decoration image through the first preprocessing module, and therefore a multi-dimensional acquisition result of the first product packaging decoration is obtained, namely the first image set.
The image information of the product package decoration can be accurately collected in real time through the first image sensor, and a sufficient and effective image data basis is provided for the subsequent quality inspection of the product package decoration quality.
Step S200: sequentially performing feature extraction on the first image set to obtain a first image feature set, wherein the first image feature set comprises a first appearance feature subset and a first composition feature subset;
specifically, the appearance feature and composition feature extraction is sequentially carried out on the product packaging decoration images acquired at different dimensions, different angles and different distances in the first image set. All appearance features and all composition features of each product packaging decoration image in the first image set can be obtained through sequential image feature extraction, wherein all appearance features of each product packaging decoration image in the first image set form a first appearance feature subset, and all composition features of each product packaging decoration image in the first image set form a first composition feature subset. Further, the first subset of appearance features and the first subset of composition features constitute all image features of the first product decoration image, i.e., the first set of image features.
The method has the advantages that the real-time characteristics of the product packaging decoration images are obtained by intelligently acquiring the appearance and composition conditions of the product packaging decoration images in all the acquired images, an accurate parameter basis is provided for judging whether the product packaging decoration meets related requirements, a data basis is provided for subsequent quality inspection results, and the technical effect of improving the quality inspection accuracy is achieved.
Step S300: obtaining a first display surface according to the first product, wherein the first display surface is a display area of the first image set;
step S400: constructing a first plane coordinate system based on the first display surface, and carrying out position identification on the first image set to obtain first image position identification information;
specifically, a machine vision-based packaging decoration quality inspection system automatically screens to obtain a first display surface of the first product, wherein the first display surface is a product packaging decoration display area in the first image set. And further constructing a corresponding plane coordinate system based on the plane structure of the first display surface, namely the first plane coordinate system. And sequentially carrying out position identification on each product package image in the first image set based on the first plane coordinate system, thereby obtaining accurate position coordinates of all product package images in the first image set in the first plane coordinate system, automatically carrying out position marking, and finally obtaining the position identification information of the first image. Based on the first plane coordinate system, the position of all the product packaging decoration images in the first image set is intelligently determined and automatically marked, so that the technical effect of intelligently analyzing the positions of the product packaging decoration images based on the same position standard is achieved.
Step S500: inputting the first appearance feature subset, the first composition feature subset and the first image position identification information into a first decoration quality evaluation model to obtain a first evaluation result;
specifically, the first decoration quality evaluation model is a neural network model having characteristics of the neural network model. The neural network model is a neural network model in machine learning, reflects many basic characteristics of human brain functions, is a deep feedforward neural network with the characteristics of local connection, weight sharing and the like, and is a highly complex nonlinear dynamic learning system. The first decoration quality evaluation model can continuously perform self-training learning according to training data, each group of data in the plurality of groups of training data comprises the first appearance feature subset, the first composition feature subset and the first image position identification information, the first decoration quality evaluation model continuously corrects the first decoration quality evaluation model by self, and when the output information of the first decoration quality evaluation model reaches a preset accuracy rate/convergence state, the supervision learning process is ended.
By carrying out data training on the first decoration quality evaluation model, the first decoration quality evaluation model can process input data more accurately, the output first evaluation result is more accurate, accurate data information acquisition is achieved, and the intelligent technical effect of the evaluation result is improved.
Step S600: obtaining a first evaluation threshold value, and judging whether the first evaluation result meets the first evaluation threshold value;
step S700: when the first evaluation result satisfies the first evaluation threshold, first qualified label information is identified for the first product decoration image.
Specifically, the first evaluation threshold is the lowest criteria that the machine vision-based package decoration quality inspection system integrates the first product packaging material, package design, and decoration condition to ultimately determine to meet the first product package decoration requirement. The packaging decoration quality inspection system based on machine vision intelligently compares an output result of a first decoration quality evaluation model with a first evaluation threshold value, and automatically marks the first product decoration image to be qualified when the output result of the first decoration quality evaluation model, namely the first evaluation result meets the first evaluation threshold value, namely the first qualified label information is marked to the corresponding first product decoration image. The method comprises the steps of presetting a first evaluation threshold value in advance by integrating multiple requirements of product packaging, intelligently judging whether an evaluation result of actual packaging and decoration of a product reaches the preset first evaluation threshold value or not by a packaging and decoration quality inspection system further based on machine vision, and automatically judging whether a first product decoration image meets related requirements and carrying out qualified marking by the system only if the intelligent judgment result is reached.
The quality of the product packaging material and the quality of the printing decoration are intelligently monitored in real time based on a computer vision technology, so that the unqualified packaging decoration is identified and timely adjusted, the packaging decoration quality inspection efficiency and the quality inspection accuracy are further improved, and the technical effect that the product packaging decoration meets the requirements is finally ensured.
Further, as shown in fig. 2, step S600 in this embodiment of the present application further includes:
step S610: when the first evaluation result does not satisfy the first evaluation threshold, identifying first non-qualified label information for the first product decoration image;
step S620: obtaining first error information according to the first non-qualified tag information, wherein the first error information comprises first characteristic error information or/and first position error information;
step S630: matching the first printing device through the first error information, wherein the first printing device is a device generating errors;
step S640: obtaining first parameter adjustment information through the first characteristic error information or/and the first position error information;
step S650: and adjusting the first printing equipment through the first parameter adjustment information.
Specifically, the machine vision-based packaging decoration quality inspection system intelligently compares an output result of a first decoration quality evaluation model with the first evaluation threshold, and automatically performs unqualified identification on the first product decoration image when the output result of the first decoration quality evaluation model, namely the first evaluation result, does not meet the first evaluation threshold, namely, marks first unqualified label information to the corresponding first product decoration image. And further according to the first non-qualified label information, intelligently analyzing the first product decoration image marked with the first non-qualified label information by a machine vision-based packaging decoration quality inspection system, so as to obtain an error condition which does not meet related requirements in the first product decoration image, namely the first error information. The first error information may exist in three different situations, and the first error situation is that the information in the product packaging decoration image does not meet the relevant requirement of the first evaluation threshold, namely the first characteristic error information; the second error condition is that the position of the product packaging decoration image does not meet the relevant requirement of the first evaluation threshold value, namely the first position error information; the third error condition is that the product packaging decoration image information and the position do not meet the related requirements of the first evaluation threshold, namely the first characteristic error information and the first position error information.
Further, based on the three error analysis results, the packaging and decoration links causing the corresponding errors and the correspondingly used equipment information are respectively matched, wherein the correspondingly matched first printing equipment is the equipment generating the errors. And analyzing error information corresponding to the three error analysis results, respectively obtaining equipment adjustment data information corresponding to different errors, namely the first parameter adjustment information, and adjusting the first decoration equipment based on the first parameter adjustment information.
And intelligently analyzing points of the product packaging decoration which do not meet the related requirements of the first evaluation threshold, and adaptively adjusting the equipment printing process parameters based on the corresponding process steps, so that the standard level of the subsequent product packaging decoration in the corresponding printing equipment is guaranteed, and the technical effect of intelligently adjusting the product packaging decoration process parameters is achieved.
Further, step S640 in the embodiment of the present application further includes:
step S641: constructing a first space coordinate system based on the first plane coordinate system, carrying out position identification on the first printing equipment, and obtaining first equipment operation position identification information, wherein the first equipment position identification information comprises position identification information of a plurality of operation elements of the first printing equipment;
step S642: obtaining first product decoration standard information, and determining first parameter adjustment direction information of the plurality of operation elements according to the first product decoration standard information;
step S643: obtaining first parameter adjustment scalar information of the plurality of operating elements through the first characteristic error information or/and the first position error information, wherein the first parameter adjustment scalar information and the first parameter adjustment direction information are in one-to-one correspondence;
step S644: using the position identification information of the plurality of operation elements as first parameter adjustment start point information;
step S645: setting the first parameter adjustment direction information, the first parameter adjustment scalar information, and the first parameter adjustment start point information as the first parameter adjustment information.
Specifically, a corresponding space coordinate system, namely the first space coordinate system, is further constructed based on a plane coordinate system constructed by the first product corresponding to the first display surface. Further, labeling the printing equipment which does not meet the first evaluation threshold value in the first space coordinate system, namely obtaining the identification information of the operation position of the first equipment. Wherein the first apparatus position identification information includes position identification information of a plurality of operation elements of the first printing apparatus. The first product decoration standard information refers to specific information of the first product meeting acceptance requirements, which is determined according to comprehensive analysis of national standards, industry standards or factory standards of production enterprises, special requirements of customers and the like in the field of product packaging decoration. And determining ideal setting parameters of a plurality of operating elements corresponding to the first equipment based on the first product decoration standard information, and further correspondingly adjusting the parameter setting in the actual printing process.
Obtaining an adjustment direction corresponding to the position of the first device, namely information of the adjustment direction of the first parameter, based on comparison between the actual printing position and the ideal printing position of the first device; and respectively carrying out adaptive adjustment on the parameters of the plurality of operating elements based on three different error analysis conditions to obtain adjustment quantities corresponding to equipment distances, namely the first parameter adjustment scalar information. Wherein the first parameter adjustment scalar information corresponds to the first parameter adjustment direction information one to one. For example, when the printing position of the product package has errors and offsets with the standard printing position, the direction of the spray head of the printing equipment is correspondingly adjusted; and when the printing color of the product package has an error with the standard printing color, correspondingly adjusting the distance of the spray head of the printing equipment.
And using the position identification information of the plurality of operation elements as a starting point of device parameter adjustment, and simultaneously using the first parameter adjustment direction information, the first parameter adjustment scalar information and the first parameter adjustment starting point information as the first parameter adjustment information. The direction and the distance of the printing equipment are respectively adjusted in an adaptive manner based on different printing error information, so that the technical effect of intelligently adjusting the parameter setting data of the printing equipment based on different printing errors is achieved.
Further, as shown in fig. 3, step S100 in the embodiment of the present application further includes:
step S110: obtaining first environment noise, and setting the first environment noise as primary image noise, wherein the first environment noise is an influence factor of the acquisition environment of the decoration image of the first product on the image quality;
step S120: obtaining first transmission noise, and setting the first transmission noise as secondary image noise, wherein the first transmission noise is an influence factor of the transmission process of the decoration image of the first product on the image quality;
step S130: inputting the secondary image noise and the first product decoration image into the first preprocessing module to be matched with a first noise reduction channel, and obtaining a primary noise reduction image set;
step S140: and inputting the primary image noise and the primary noise reduction image set into the first preprocessing module to match with a second noise reduction channel to obtain the first image set.
Specifically, the first environmental noise is external factors that affect the packaging decoration image of the product, such as the illumination intensity, the air humidity, and the angle at which the first image sensor captures the image in the surrounding environment when the first product is printed for packaging decoration. The first transmission noise refers to internal factors which influence the product packaging decoration image, such as friction and heating of a spray head part during transmission of each equipment element when the first product is subjected to packaging decoration printing.
The packaging decoration quality inspection system based on machine vision intelligently collects relevant external and internal factors to obtain the first environmental noise and the first transmission noise, the first environmental noise is set as first-level image noise, and the first transmission noise is set as second-level image noise. The first environmental noise is an influence factor of the acquisition environment of the first product decoration image on the image quality; the first transmission noise is an influence factor of the first product decoration image transmission process on the image quality. Inputting the secondary image noise and the first product decoration image into the first preprocessing module to be matched with a first noise reduction channel to obtain a primary noise reduction image set, further inputting the primary image noise and the primary noise reduction image set into the first preprocessing module to be matched with a second noise reduction channel, and finally obtaining the first image set. The first denoising channel is a model used for processing transmission noise in the first preprocessing module, and the second denoising channel is a model used for processing environmental noise in the first preprocessing module.
The first product decoration image collected by the first image sensor is subjected to environment noise reduction and transmission noise reduction through the first preprocessing module, and finally the first image set is obtained, so that the technical effect of reducing the influence degree on product packaging decoration during transmission of an external environment and internal equipment is achieved.
Further, step S700 in the embodiment of the present application further includes:
step S710: obtaining a first appearance quality evaluation result, a first composition quality evaluation result and a first image position quality evaluation result according to the first evaluation result;
step S720: respectively judging whether the first appearance quality evaluation result, the first composition quality evaluation result and the first image position quality evaluation result meet the first evaluation threshold value, and obtaining a first judgment result, a second judgment result and a third judgment result;
step S730: when the first determination result, the second determination result, and the third determination result are simultaneously satisfied, the first evaluation result satisfies the first evaluation threshold.
Specifically, according to the first evaluation result intelligently output by the first decoration quality evaluation model, the evaluation results of the first decoration quality evaluation model on the appearance, composition and position of the first product package decoration can be respectively obtained, namely the first appearance quality evaluation result, the first composition quality evaluation result and the first image position quality evaluation result. And sequentially judging whether the shape, the composition and the position of the first product in the packaging decoration meet the relevant requirements in the first evaluation threshold value, and correspondingly obtaining a first judgment result, a second judgment result and a third judgment result respectively. The first evaluation result satisfies the first evaluation threshold only if the first determination result, the second determination result, and the third determination result satisfy the corresponding requirements at the same time. Through judging each feature point in the first product packaging decoration respectively, when all feature points all meet the requirements, just affirm first product packaging decoration is up to standard, has reached based on image appearance, composition and position, and whether intelligent judgement first evaluation result satisfies the technological effect of first evaluation threshold value.
Further, step S700 in the embodiment of the present application further includes:
step S740: extracting features of the first display surface to obtain first texture feature information;
step S750: inputting the first texture feature information into the first decoration quality evaluation model to obtain a first texture quality evaluation result;
step S760: and judging whether the first texture quality evaluation result meets the first evaluation threshold value or not, and obtaining a fourth judgment result.
Specifically, feature extraction is performed on the display area image in the first image set corresponding to the first product, and texture features in the corresponding image, that is, the first texture feature information, are obtained. The texture is a visual feature reflecting the homogeneity phenomenon in the image, and embodies the surface structure organization arrangement attribute with slow change or periodic change on the surface of the object. And inputting the first texture feature information into the first decoration quality evaluation model, intelligently obtaining a first texture quality evaluation result by the first decoration quality evaluation model, and further judging whether the first texture quality evaluation result meets the first evaluation threshold value, so as to obtain a fourth judgment result. The technical effect of inspecting the packaging decoration quality based on the texture features in the packaging decoration image of the first product is achieved.
Further, as shown in fig. 4, the embodiment of the present application further includes:
step S810: downloading a first encryption mode from a first collaboration system;
step S820: training and constructing a first initial model by using a plurality of groups of the first appearance feature subsets, the first composition feature subsets and the first image position identification information to obtain first model parameters;
step S830: encrypting the first model parameter by using the first encryption mode to obtain a first encryption result;
step S840: uploading the first encryption result to the first collaboration system, and acquiring the first encryption parameter by the first collaboration system by aggregating the first encryption result and the second encryption result until a Kth encryption result is obtained, wherein the first encryption result and the second encryption result until the Kth encryption result are information provided by K different factories, and K is a natural number greater than or equal to 2;
step S850: and updating the first initial model by using the first encryption parameter to obtain the first decoration quality evaluation model.
Specifically, a first initial model is trained and constructed by using a plurality of sets of the first appearance feature subset, the first composition feature subset and the first image position identification information, and relevant parameters of the first initial model can be obtained, namely the first model parameters. And further downloading an encryption mode from the first cooperation system for encrypting the parameters of the first model, and uploading the encrypted result to the first cooperation system. And collecting and sorting parameter information provided by K factories, and sequentially carrying out the same encryption processing to obtain the first encryption result, the second encryption result and the Kth encryption result. Wherein K is a natural number of 2 or more. And aggregating the first encryption result, the second encryption result and the Kth encryption result in the first cooperation system to finally obtain model parameter information, namely the first encryption parameter. And updating the first initial model through the first encryption parameter to finally obtain the first decoration quality evaluation model.
The model parameters corresponding to each factory are obtained by aggregating parameter data provided by a plurality of factories, one model parameter is finally determined based on the plurality of model parameters, so that the model constructed based on the final model parameter is more reasonable, the evaluation result is more accurate and effective, and meanwhile, the model parameters provided by each factory are encrypted, thereby ensuring that the model parameters of each factory are not leaked.
In summary, the packaging decoration quality inspection method based on machine vision provided by the embodiment of the application has the following technical effects:
1. acquiring a first product decoration image through a first image sensor, inputting the first product decoration image into a first preprocessing module, and acquiring a first image set, wherein the first image set is a multi-dimensional acquisition result; sequentially performing feature extraction on the first image set to obtain a first image feature set, wherein the first image feature set comprises a first appearance feature subset and a first composition feature subset; obtaining a first display surface according to the first product, wherein the first display surface is a display area of the first image set; constructing a first plane coordinate system based on the first display surface, and carrying out position identification on the first image set to obtain first image position identification information; inputting the first appearance feature subset, the first composition feature subset and the first image position identification information into a first decoration quality evaluation model to obtain a first evaluation result; obtaining a first evaluation threshold value, and judging whether the first evaluation result meets the first evaluation threshold value; when the first evaluation result satisfies the first evaluation threshold, first qualified label information is identified for the first product decoration image. The quality of the product packaging material and the quality of the printing decoration are intelligently monitored in real time based on a computer vision technology, so that the unqualified packaging decoration is identified and timely adjusted, the packaging decoration quality inspection efficiency and the quality inspection accuracy are further improved, and the technical effect that the product packaging decoration meets the requirements is finally ensured.
2. And intelligently analyzing points of the product packaging decoration which do not meet the related requirements of the first evaluation threshold, and adaptively adjusting the equipment printing process parameters based on the corresponding process steps, so that the standard level of the subsequent product packaging decoration in the corresponding printing equipment is guaranteed, and the technical effect of intelligently adjusting the product packaging decoration process parameters is achieved.
3. The model parameters corresponding to each factory are obtained by aggregating parameter data provided by a plurality of factories, one model parameter is finally determined based on the plurality of model parameters, so that the model constructed based on the final model parameter is more reasonable, the evaluation result is more accurate and effective, and meanwhile, the model parameters provided by each factory are encrypted, thereby ensuring that the model parameters of each factory are not leaked.
Example two
Based on the same inventive concept as the packaging decoration quality inspection method based on machine vision in the foregoing embodiment, the present invention further provides a packaging decoration quality inspection system based on machine vision, please refer to fig. 5, where the system includes:
the first obtaining unit 11: the first obtaining unit 11 is configured to collect a first product decoration image through a first image sensor, input the first product decoration image into a first preprocessing module, and obtain a first image set, where the first image set is a multi-dimensional collection result;
the second obtaining unit 12: the second obtaining unit 12 is configured to sequentially perform feature extraction on the first image set to obtain a first image feature set, where the first image feature set includes a first outline feature subset and a first composition feature subset;
the third obtaining unit 13: the third obtaining unit 13 is configured to obtain a first display surface according to the first product, where the first display surface is a display area of the first image set;
the fourth obtaining unit 14: the fourth obtaining unit 14 is configured to construct a first planar coordinate system based on the first display surface, perform position identification on the first image set, and obtain first image position identification information;
the fifth obtaining unit 15: the fifth obtaining unit 15 is configured to input the first appearance feature subset, the first composition feature subset, and the first image position identification information into a first decoration quality evaluation model, so as to obtain a first evaluation result;
the first judgment unit 16: the first judging unit 16 is configured to obtain a first evaluation threshold, and judge whether the first evaluation result satisfies the first evaluation threshold;
the first identification unit 17: the first identification unit 17 is configured to identify first qualified label information for the first product decoration image when the first evaluation result satisfies the first evaluation threshold.
Further, the system further comprises:
a second identification unit configured to identify first non-qualified label information for the first product decoration image when the first evaluation result does not satisfy the first evaluation threshold;
a sixth obtaining unit, configured to obtain first error information according to the first non-qualified tag information, where the first error information includes first characteristic error information or/and first position error information;
a first matching unit, configured to match the first printing device with the first error information, where the first printing device is a device that generates an error;
a seventh obtaining unit, configured to obtain first parameter adjustment information through the first characteristic error information or/and the first position error information;
a first adjusting unit configured to adjust the first printing apparatus by the first parameter adjustment information.
Further, the system further comprises:
an eighth obtaining unit, configured to construct a first spatial coordinate system based on the first planar coordinate system, perform position identification on the first printing device, and obtain first device operation position identification information, where the first device position identification information includes position identification information of a plurality of operation elements of the first printing device;
the first determining unit is used for obtaining first product decoration standard information and determining first parameter adjustment direction information of the plurality of operating elements according to the first product decoration standard information;
a ninth obtaining unit, configured to obtain first parameter adjustment scalar information of the plurality of operating elements through the first feature error information or/and the first position error information, where the first parameter adjustment scalar information and the first parameter adjustment direction information are in one-to-one correspondence;
a first setting unit configured to set position identification information of the plurality of operation elements as first parameter adjustment start point information;
a second setting unit configured to set the first parameter adjustment direction information, the first parameter adjustment scalar information, and the first parameter adjustment start point information as the first parameter adjustment information.
Further, the system further comprises:
a tenth obtaining unit, configured to obtain a first environmental noise, and set the first environmental noise as a primary image noise, where the first environmental noise is an influence factor of an acquisition environment of the first product decoration image on image quality;
an eleventh obtaining unit, configured to obtain a first transmission noise, and set the first transmission noise as a secondary image noise, where the first transmission noise is an influence factor of a transmission process of the decoration image of the first product on image quality;
a twelfth obtaining unit, configured to input the secondary image noise and the first product decoration image into the first preprocessing module to match the first denoising channel, so as to obtain a primary denoising image set;
a thirteenth obtaining unit, configured to input the primary image noise and the primary noise-reduced image set into the first preprocessing module to match a second noise-reduction channel, so as to obtain the first image set.
Further, the system further comprises:
a fourteenth obtaining unit configured to obtain a first appearance quality evaluation result, a first composition quality evaluation result, and a first image position quality evaluation result, according to the first evaluation result;
a second judging unit, configured to respectively judge whether the first appearance quality assessment result, the first composition quality assessment result, and the first image position quality assessment result satisfy the first assessment threshold, and obtain a first judgment result, a second judgment result, and a third judgment result;
a second determination unit configured to determine that the first evaluation result satisfies the first evaluation threshold if the first determination result, the second determination result, and the third determination result are satisfied at the same time.
Further, the system further comprises:
a fifteenth obtaining unit, configured to perform feature extraction on the first display surface to obtain first texture feature information;
a sixteenth obtaining unit, configured to input the first texture feature information into the first decoration quality assessment model, and obtain a first texture quality assessment result;
a third determining unit, configured to determine whether the first texture quality evaluation result satisfies the first evaluation threshold, and obtain a fourth determination result.
Further, the system further comprises:
a first downloading unit, configured to download a first encryption scheme from a first collaboration system;
a seventeenth obtaining unit, configured to train and construct a first initial model using the plurality of sets of the first appearance feature subset, the first composition feature subset, and the first image position identification information, and obtain a first model parameter;
an eighteenth obtaining unit, configured to encrypt the first model parameter by using the first encryption method, and obtain a first encryption result;
a nineteenth obtaining unit, configured to upload the first encryption result to the first collaboration system, where the first collaboration system obtains the first encryption parameter by aggregating the first encryption result and the second encryption result until a kth encryption result, where the first encryption result, the second encryption result until the kth encryption result are information provided by K different plants, and K is a natural number greater than or equal to 2;
a twentieth obtaining unit, configured to update the first initial model using the first encryption parameter, and obtain the first decoration quality evaluation model.
The embodiments in the present specification are described in a progressive manner, and each embodiment focuses on the difference from the other embodiments, and the aforementioned machine vision-based packaging decoration quality inspection method and the specific example in the first embodiment in fig. 1 are also applicable to the machine vision-based packaging decoration quality inspection system in the present embodiment, and through the foregoing detailed description of the machine vision-based packaging decoration quality inspection method, a machine vision-based packaging decoration quality inspection system in the present embodiment is clearly known to those skilled in the art, and therefore, for the brevity of the description, detailed description is not repeated here. The device disclosed by the embodiment corresponds to the method disclosed by the embodiment, so that the description is simple, and the relevant points can be referred to the method part for description.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
The electronic apparatus of the embodiment of the present application is described below with reference to fig. 6.
Fig. 6 illustrates a schematic structural diagram of an electronic device according to an embodiment of the present application.
Based on the inventive concept of a machine vision-based packaging decoration quality inspection method in the foregoing embodiments, the present invention also provides a machine vision-based packaging decoration quality inspection system, on which a computer program is stored, which when executed by a processor implements the steps of any one of the foregoing machine vision-based packaging decoration quality inspection methods.
Where in fig. 6 a bus architecture (represented by bus 300), bus 300 may include any number of interconnected buses and bridges, bus 300 linking together various circuits including one or more processors, represented by processor 302, and memory, represented by memory 304. The bus 300 may also link together various other circuits such as peripherals, voltage regulators, power management circuits, and the like, which are well known in the art, and therefore, will not be described any further herein. A bus interface 305 provides an interface between the bus 300 and the receiver 301 and transmitter 303. The receiver 301 and the transmitter 303 may be the same element, i.e., a transceiver, providing a means for communicating with various other apparatus over a transmission medium.
The processor 302 is responsible for managing the bus 300 and general processing, and the memory 304 may be used for storing data used by the processor 302 in performing operations.
The application provides a machine vision-based packaging decoration quality inspection method, which is applied to a machine vision-based packaging decoration quality inspection system, wherein the method comprises the following steps: acquiring a first product decoration image through a first image sensor, inputting the first product decoration image into a first preprocessing module, and acquiring a first image set; sequentially performing feature extraction on the first image set to obtain a first image feature set; obtaining a first display surface according to the first product; constructing a first plane coordinate system based on the first display surface to obtain first image position identification information; obtaining a first evaluation result based on the first decoration quality evaluation model; obtaining a first evaluation threshold value, and judging whether the first evaluation result meets the first evaluation threshold value; when the first evaluation result satisfies the first evaluation threshold, first qualified label information is identified for the first product decoration image. The technical problems of low efficiency, wrong detection and missed detection of the product packaging materials and the printing and decorating quality in the prior art are solved. The quality of the product packaging material and the quality of the printing decoration are intelligently monitored in real time based on a computer vision technology, so that the unqualified packaging decoration is identified and timely adjusted, the packaging decoration quality inspection efficiency and the quality inspection accuracy are further improved, and the technical effect that the product packaging decoration meets the requirements is finally ensured.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, apparatus, or computer program product. Accordingly, the present application may take the form of an entirely software embodiment, an entirely hardware embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present application is in the form of a computer program product that may be embodied on one or more computer-usable storage media having computer-usable program code embodied therewith. These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including an instruction system which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks. While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all such alterations and modifications as fall within the scope of the invention.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.
Claims (9)
1. A machine vision based packaging decoration quality inspection method, wherein the method is applied to a machine vision based packaging decoration quality inspection system, and the method comprises the following steps:
acquiring a first product decoration image through a first image sensor, inputting the first product decoration image into a first preprocessing module, and acquiring a first image set, wherein the first image set is a multi-dimensional acquisition result;
sequentially performing feature extraction on the first image set to obtain a first image feature set, wherein the first image feature set comprises a first appearance feature subset and a first composition feature subset;
obtaining a first display surface according to the first product, wherein the first display surface is a display area of the first image set;
constructing a first plane coordinate system based on the first display surface, and carrying out position identification on the first image set to obtain first image position identification information;
inputting the first appearance feature subset, the first composition feature subset and the first image position identification information into a first decoration quality evaluation model to obtain a first evaluation result;
obtaining a first evaluation threshold value, and judging whether the first evaluation result meets the first evaluation threshold value;
when the first evaluation result satisfies the first evaluation threshold, first qualified label information is identified for the first product decoration image.
2. The method of claim 1, wherein the method further comprises:
when the first evaluation result does not satisfy the first evaluation threshold, identifying first non-qualified label information for the first product decoration image;
obtaining first error information according to the first non-qualified tag information, wherein the first error information comprises first characteristic error information or/and first position error information;
matching a first printing device through the first error information, wherein the first printing device is a device generating errors;
obtaining first parameter adjustment information through the first characteristic error information or/and the first position error information;
and adjusting the first printing equipment through the first parameter adjustment information.
3. The method of claim 2, wherein the obtaining first parameter adjustment information from the first characteristic error information or/and the first position error information comprises:
constructing a first space coordinate system based on the first plane coordinate system, and carrying out position identification on the first printing equipment to obtain first equipment operation position identification information, wherein the first equipment operation position identification information comprises position identification information of a plurality of operation elements of the first printing equipment;
obtaining first product decoration standard information, and determining first parameter adjustment direction information of the plurality of operation elements according to the first product decoration standard information;
obtaining first parameter adjustment scalar information of the plurality of operating elements through the first characteristic error information or/and the first position error information, wherein the first parameter adjustment scalar information and the first parameter adjustment direction information are in one-to-one correspondence;
using the position identification information of the plurality of operation elements as first parameter adjustment start point information;
setting the first parameter adjustment direction information, the first parameter adjustment scalar information, and the first parameter adjustment start point information as the first parameter adjustment information.
4. The method of claim 1, wherein said capturing a first product decoration image by a first image sensor, inputting into a first pre-processing module, obtaining a first set of images, comprises:
obtaining first environment noise, and setting the first environment noise as primary image noise, wherein the first environment noise is an influence factor of the acquisition environment of the decoration image of the first product on the image quality;
obtaining first transmission noise, and setting the first transmission noise as secondary image noise, wherein the first transmission noise is an influence factor of the transmission process of the decoration image of the first product on the image quality;
inputting the secondary image noise and the first product decoration image into the first preprocessing module to be matched with a first noise reduction channel, and obtaining a primary noise reduction image set;
and inputting the primary image noise and the primary noise reduction image set into the first preprocessing module to match with a second noise reduction channel to obtain the first image set.
5. The method of claim 1, wherein the obtaining a first evaluation threshold and determining whether the first evaluation result satisfies the first evaluation threshold comprises:
obtaining a first appearance quality evaluation result, a first composition quality evaluation result and a first image position quality evaluation result according to the first evaluation result;
respectively judging whether the first appearance quality evaluation result, the first composition quality evaluation result and the first image position quality evaluation result meet the first evaluation threshold value, and obtaining a first judgment result, a second judgment result and a third judgment result;
when the first determination result, the second determination result, and the third determination result are simultaneously satisfied, the first evaluation result satisfies the first evaluation threshold.
6. The method of claim 1, wherein the method further comprises:
extracting features of the first display surface to obtain first texture feature information;
inputting the first texture feature information into the first decoration quality evaluation model to obtain a first texture quality evaluation result;
and judging whether the first texture quality evaluation result meets the first evaluation threshold value or not, and obtaining a fourth judgment result.
7. The method of claim 1, wherein the method comprises:
downloading a first encryption mode from a first collaboration system;
training and constructing a first initial model by using a plurality of groups of the first appearance feature subsets, the first composition feature subsets and the first image position identification information to obtain first model parameters;
encrypting the first model parameter by using the first encryption mode to obtain a first encryption result;
uploading the first encryption result to the first cooperation system, and acquiring a first encryption parameter by the first cooperation system by aggregating the first encryption result and the second encryption result until a Kth encryption result, wherein the first encryption result and the second encryption result until the Kth encryption result are information provided by K different factories, and K is a natural number greater than or equal to 2;
and updating the first initial model by using the first encryption parameter to obtain the first decoration quality evaluation model.
8. A machine vision based packaging decoration quality inspection system, wherein the system comprises:
a first obtaining unit: the first obtaining unit is used for collecting a first product decoration image through a first image sensor, inputting the first product decoration image into a first preprocessing module, and obtaining a first image set, wherein the first image set is a multi-dimensional collecting result;
a second obtaining unit: the second obtaining unit is used for sequentially performing feature extraction on the first image set to obtain a first image feature set, wherein the first image feature set comprises a first appearance feature subset and a first composition feature subset;
a third obtaining unit: the third obtaining unit is configured to obtain a first display surface according to the first product, where the first display surface is a display area of the first image set;
a fourth obtaining unit: the fourth obtaining unit is configured to construct a first planar coordinate system based on the first display surface, perform position identification on the first image set, and obtain first image position identification information;
a fifth obtaining unit: the fifth obtaining unit is used for inputting the first appearance feature subset, the first composition feature subset and the first image position identification information into a first decoration quality evaluation model to obtain a first evaluation result;
a first judgment unit: the first judging unit is used for obtaining a first evaluation threshold value and judging whether the first evaluation result meets the first evaluation threshold value;
a first identification unit: the first identification unit is used for identifying first qualified label information for the first product decoration image when the first evaluation result meets the first evaluation threshold value.
9. A machine vision based packaging decoration quality inspection system comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor when executing the program implements the steps of the method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111177798.3A CN113610851B (en) | 2021-10-09 | 2021-10-09 | Packaging decoration quality inspection method and system based on machine vision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111177798.3A CN113610851B (en) | 2021-10-09 | 2021-10-09 | Packaging decoration quality inspection method and system based on machine vision |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113610851A true CN113610851A (en) | 2021-11-05 |
| CN113610851B CN113610851B (en) | 2021-12-31 |
Family
ID=78343437
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111177798.3A Active CN113610851B (en) | 2021-10-09 | 2021-10-09 | Packaging decoration quality inspection method and system based on machine vision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113610851B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114889971A (en) * | 2022-04-26 | 2022-08-12 | 浙江复禾降解新材料有限公司 | Degradable express packaging bag and preparation method thereof |
| CN117351292A (en) * | 2023-12-04 | 2024-01-05 | 安徽省农业科学院烟草研究所 | Tobacco production facility management method and system based on Internet of things |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6118893A (en) * | 1997-07-16 | 2000-09-12 | Cognex Corporation | Analysis of an image of a pattern of discrete objects |
| CN106290383A (en) * | 2015-06-26 | 2017-01-04 | Jvm有限公司 | Medicated bag checks device and medicated bag inspection method |
| CN109377485A (en) * | 2018-10-12 | 2019-02-22 | 龙口味美思环保科技有限公司 | A kind of instant noodles packaging defect machine vision detection method |
| CN111242900A (en) * | 2019-12-31 | 2020-06-05 | 电子科技大学中山学院 | Product qualification determination method and device, electronic equipment and storage medium |
| CN112977974A (en) * | 2021-03-18 | 2021-06-18 | 厦门烟草工业有限责任公司 | Cigarette packet appearance quality detection device and method and cigarette packet packaging machine |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113247387B (en) * | 2021-05-18 | 2023-04-25 | 淄博职业学院 | Food green packaging method and device |
-
2021
- 2021-10-09 CN CN202111177798.3A patent/CN113610851B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6118893A (en) * | 1997-07-16 | 2000-09-12 | Cognex Corporation | Analysis of an image of a pattern of discrete objects |
| CN106290383A (en) * | 2015-06-26 | 2017-01-04 | Jvm有限公司 | Medicated bag checks device and medicated bag inspection method |
| CN109377485A (en) * | 2018-10-12 | 2019-02-22 | 龙口味美思环保科技有限公司 | A kind of instant noodles packaging defect machine vision detection method |
| CN111242900A (en) * | 2019-12-31 | 2020-06-05 | 电子科技大学中山学院 | Product qualification determination method and device, electronic equipment and storage medium |
| CN112977974A (en) * | 2021-03-18 | 2021-06-18 | 厦门烟草工业有限责任公司 | Cigarette packet appearance quality detection device and method and cigarette packet packaging machine |
Non-Patent Citations (2)
| Title |
|---|
| XIA NIAN-JIONG，ET.AL: "A machine vision system for ball grid array package inspection", 《JOURNAL OF SHANGHAI JIAOTONG UNIVERSITY》 * |
| 杨小艳: "基于图像处理技术的包装表面缺陷检测", 《微型电脑应用》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114889971A (en) * | 2022-04-26 | 2022-08-12 | 浙江复禾降解新材料有限公司 | Degradable express packaging bag and preparation method thereof |
| CN117351292A (en) * | 2023-12-04 | 2024-01-05 | 安徽省农业科学院烟草研究所 | Tobacco production facility management method and system based on Internet of things |
| CN117351292B (en) * | 2023-12-04 | 2024-03-08 | 安徽省农业科学院烟草研究所 | Tobacco production facility management method and system based on Internet of things |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113610851B (en) | 2021-12-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113610851B (en) | Packaging decoration quality inspection method and system based on machine vision | |
| US11240441B2 (en) | Method, device, system and computer-program product for setting lighting condition and storage medium | |
| US20210390678A1 (en) | Method for monitoring manufacture of assembly units | |
| CN106770332A (en) | A kind of electronic die blank defects detection implementation method based on machine vision | |
| CN102521560B (en) | Instrument pointer image identification method of high-robustness rod | |
| US11435297B2 (en) | Image capture method and image capture device | |
| CN103034838B (en) | A kind of special vehicle instrument type identification based on characteristics of image and scaling method | |
| CN104913797A (en) | Pointer type instrument number reading and recognition method and system | |
| CN110865077B (en) | Visual inspection system for appearance defects in RFID antenna production | |
| CA2525594A1 (en) | A system and method for monitoring and visualizing the output of a production process | |
| CN111950396A (en) | Instrument reading neural network identification method | |
| CN109863530A (en) | Determine the grain weight of grain ear | |
| CN112862757A (en) | Weight evaluation system based on computer vision technology and implementation method | |
| CN114758249A (en) | Target object monitoring method, device, equipment and medium based on field night environment | |
| CN111242057A (en) | Product sorting system, method, computer device and storage medium | |
| CN115619787A (en) | UV adhesive defect detection method, system, equipment and medium | |
| CN109035666B (en) | Fire and smoke detection method and device and terminal equipment | |
| CN109934231A (en) | Bar section character picture recognition methods based on multi-direction gray level co-occurrence matrixes | |
| CN116243623B (en) | Robot scene simulation method applied to digital robot industrial chain | |
| CN111597898A (en) | Image recognition and visual analysis method and system for power inspection robot | |
| CN115311447A (en) | Pointer instrument indicating number identification method based on deep convolutional neural network | |
| CN114187494A (en) | Perception information fusion method of multiple sensors | |
| CN113781013A (en) | Coating color point monitoring method, system, medium and terminal | |
| CN117798654B (en) | Intelligent adjusting system for center of steam turbine shafting | |
| CN111709620B (en) | Mobile portable online detection system for structural parameters of woven fabric |
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 |