CN113591508B - Bar code decoding method and device based on artificial intelligence target recognition and storage medium - Google Patents
Bar code decoding method and device based on artificial intelligence target recognition and storage medium Download PDFInfo
- Publication number
- CN113591508B CN113591508B CN202111146903.7A CN202111146903A CN113591508B CN 113591508 B CN113591508 B CN 113591508B CN 202111146903 A CN202111146903 A CN 202111146903A CN 113591508 B CN113591508 B CN 113591508B
- Authority
- CN
- China
- Prior art keywords
- bar code
- artificial intelligence
- barcode
- decoding
- unit
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K7/00—Methods or arrangements for sensing record carriers, e.g. for reading patterns
- G06K7/10—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation
- G06K7/14—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation using light without selection of wavelength, e.g. sensing reflected white light
- G06K7/1404—Methods for optical code recognition
- G06K7/1408—Methods for optical code recognition the method being specifically adapted for the type of code
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K7/00—Methods or arrangements for sensing record carriers, e.g. for reading patterns
- G06K7/10—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation
- G06K7/14—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation using light without selection of wavelength, e.g. sensing reflected white light
- G06K7/1404—Methods for optical code recognition
- G06K7/1439—Methods for optical code recognition including a method step for retrieval of the optical code
- G06K7/1443—Methods for optical code recognition including a method step for retrieval of the optical code locating of the code in an image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K7/00—Methods or arrangements for sensing record carriers, e.g. for reading patterns
- G06K7/10—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation
- G06K7/14—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation using light without selection of wavelength, e.g. sensing reflected white light
- G06K7/1404—Methods for optical code recognition
- G06K7/146—Methods for optical code recognition the method including quality enhancement steps
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Toxicology (AREA)
- General Health & Medical Sciences (AREA)
- Electromagnetism (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a bar code decoding method, a device and a storage medium based on artificial intelligence target identification, wherein the bar code decoding method comprises the steps of establishing an artificial intelligence identification model; and carrying out bar code identification based on the artificial intelligence identification model. Firstly, an artificial intelligence recognition model is constructed and trained, then the bar codes are positioned and recognized based on the trained artificial intelligence recognition model, finally the recognized bar codes are decoded, and for the logistics recognition bar code scene, the bar codes are not required to be superposed or switched under different conditions in the inference process, so that the scene efficiency is improved, and the bar codes at random positions are quickly captured; for a multi-code recognition scene, a plurality of bar codes can be simultaneously positioned and recognized, and the method is not limited by the factors of position/background/type difference of the bar codes and defective bar codes.
Description
Technical Field
The invention relates to the technical field of bar code decoding, in particular to a bar code decoding method and device based on artificial intelligence target identification and a storage medium.
Background
The method has the advantages that the realization of complete product traceability is the target of the current production intelligent development, the key link production traceability is required to realize that materials related to the product are integrated one by one, the materials are merged step by step in the production process to complete final product coding, the product coding can be used for quality traceability of subsequent sales, and the market supervision is convenient and is also a powerful guarantee for the product quality. And big data generated in the coding process can be used for evaluating the production efficiency and is a means for improving the quality of enterprises on the whole, so that the method is developed rapidly. The bar code information attached to the material entity is required to be acquired in the product code assigning process, the used front-end equipment is a bar code reader, the bar code reader has the function that a camera shoots bar codes to form pictures, then an internal arithmetic unit processes the pictures, locates the bar codes, solves the digital information contained in the bar codes and transmits the digital information to a computer on a production site. However, with the development of industry, bar code reading is widely and deeply, wherein multi-code identification and logistics identification are difficult points in industry application, the traditional bar code decoding method searches bar code characteristics in an image line by line, one mark is searched out every time, and in a logistics code reading scene, as a large amount of computing power is consumed in searching bar codes, the higher the demand speed is, the larger the consumption computing power is. In a plurality of barcode distribution scenes, particularly, position/background/type differences exist at the same time, and defective barcodes exist, more calculation power is needed to be consumed for realizing multi-barcode search, so that the existing barcode decoding method has low positioning and identifying efficiency no matter in a multi-barcode identification scene or a logistics identification barcode scene.
Disclosure of Invention
The technical problem to be solved by the invention is to provide a barcode decoding method and device based on artificial intelligence target identification, which can realize the concurrent positioning of a plurality of barcodes and improve the barcode searching and positioning efficiency.
The first invention of the invention is to provide a bar code decoding method based on artificial intelligence target recognition, comprising:
establishing an artificial intelligence recognition model, wherein the operation steps of establishing the artificial intelligence recognition model are as follows:
acquiring a barcode scene image to form a training set and a verification set;
constructing an artificial intelligence recognition model, and training the artificial intelligence recognition model by using a training set;
the verification set is sorted by using a decoding algorithm and used for verifying the artificial intelligence algorithm, the operation result of the artificial intelligence algorithm is supervised by taking the decoding algorithm as a reference, the two running unmatched results are put into a training set for retraining and the process is repeated, and the verification set is distilled into a miniature training set; and
performing bar code recognition based on the artificial intelligence recognition model, wherein the operation steps of performing bar code recognition based on the artificial intelligence recognition model are as follows:
inputting the image with the bar code into a trained artificial intelligence recognition model to position the bar code to obtain a bar code area;
and decoding the bar code area to obtain bar code information.
Further, the verification set is sorted by using a decoding algorithm and used for verifying the artificial intelligence algorithm. And monitoring the operation result of the artificial intelligence algorithm by taking the decoding algorithm as a reference, putting the decoding algorithm and the artificial intelligence algorithm into a training set when the decoding algorithm and the artificial intelligence algorithm are not matched in operation, training again and repeating the process, wherein the operation steps of distilling the verification set into a miniature training set are as follows:
decoding and testing the barcode scene images in the verification set by using a decoding algorithm to obtain and store corresponding intermediate variables, wherein the intermediate variables comprise calculation method identification values and corresponding time values;
classifying the verification set according to the intermediate variable, wherein the classification identification is a primary type, and the secondary corresponds to the quality;
loading an artificial intelligence recognition model by an artificial intelligence recognition algorithm, and recognizing the classified verification set to obtain corresponding recognition result data, wherein the recognition result data comprises recall rate and accuracy rate;
and screening out corresponding barcode scene images from the identification result data and the actual identification result data to form a new training set, retraining the artificial intelligent identification model from the new training set, and circulating the steps to finally form a training set with stable micro-scale.
Further, the operation steps of inputting the image with the bar code into a trained artificial intelligence recognition model to position the bar code and obtaining the bar code area are as follows:
the method comprises the following operation steps of inputting an image with a bar code into a trained artificial intelligent recognition model to position the bar code to obtain a bar code area:
extracting image features, and predicting a position and a category probability value according to the image features;
combining the prediction result to output a code position and a confidence value;
and establishing coordinates, intercepting the bar code according to the coordinates, and determining the category and quality of the bar code according to the category probability value and the confidence value corresponding to the bar code.
Further, the operation steps of decoding the barcode region to obtain barcode information are as follows:
performing corresponding binarization processing according to the category and quality of the bar code, and extracting a bar code characteristic image;
and decoding according to the bar code characteristic image to obtain bar code information.
Further, the establishing of the coordinates and the intercepting of the bar code according to the coordinates, the determining of the category and the quality of the bar code according to the category probability value and the confidence value corresponding to the bar code further comprise image preprocessing of the intercepted bar code, wherein the image preprocessing comprises but is not limited to denoising, blurring, sharpening, color-to-black, scaling and normalization.
A second object of the present invention is to provide a barcode decoding apparatus based on artificial intelligence target recognition, comprising:
the system comprises an acquisition unit, a storage unit and a verification unit, wherein the acquisition unit is used for acquiring a barcode scene image and dividing the barcode scene image into a training set and a verification set, the training set is used for training the artificial intelligence recognition model, and the verification set is used for verifying the artificial intelligence recognition model;
the model building and training unit is used for building an artificial intelligence recognition model and training the artificial intelligence recognition model through the training set;
the verification unit is used for verifying the artificial intelligence recognition model through the verification set, monitoring the operation result of the artificial intelligence algorithm by taking the decoding algorithm as a reference, putting the decoding algorithm and the artificial intelligence recognition model into a training set for retraining and repeating the process, and distilling the verification set into a miniature training set;
the bar code positioning unit is used for inputting the acquired image with the bar code into a trained artificial intelligence recognition model and positioning the bar code;
and the bar code decoding unit is used for decoding the bar code to acquire bar code information.
Furthermore, the barcode positioning unit comprises a feature processing unit, a barcode prediction unit and a barcode processing unit, wherein the feature processing unit is used for extracting and processing image features, and the barcode prediction unit is used for predicting the barcode position and the category probability value according to the image features and outputting the barcode position and the confidence value; the bar code processing unit is used for establishing coordinates according to the bar code position and intercepting the bar code according to the coordinates.
Further, the bar code decoding unit comprises a binarization unit and a decoding unit, wherein the binarization unit is used for carrying out binarization processing on the category and the quality of the positioned bar code and extracting a bar code characteristic image; and the decoding unit decodes the bar code characteristic image.
A third object of the present invention is to provide a computer-readable storage medium storing computer instructions for causing the computer to execute the barcode decoding method based on artificial intelligence target recognition as described above.
Has the advantages that: firstly, constructing an artificial intelligence recognition model and training the artificial intelligence recognition model, and then based on the trained artificial intelligence recognition model, recognizing bar codes; finally, decoding the bar code area to obtain bar code information, identifying a bar code scene for logistics based on a trained artificial intelligence identification model, wherein different conditions do not need to be superposed or switched in the inference process, and the bar code at a random position is quickly captured in the face of improving scene efficiency; for a multi-code recognition scene, a plurality of bar codes can be simultaneously positioned and recognized, and the method is not limited by the factors of position/background/type difference of the bar codes and defective bar codes.
Detailed Description
Example 1
The bar code decoding method based on artificial intelligence target recognition comprises the following steps:
establishing an artificial intelligence recognition model, wherein the operation steps of establishing the artificial intelligence recognition model are as follows:
acquiring a barcode scene image to form a training set and a verification set;
constructing an artificial intelligence recognition model, and training the artificial intelligence recognition model by using a training set;
and sorting the verification set by using a decoding algorithm for verifying the artificial intelligence algorithm. Monitoring the operation result of the artificial intelligence algorithm by taking the decoding algorithm as a reference, putting the decoding algorithm and the artificial intelligence algorithm into a training set when the decoding algorithm and the artificial intelligence algorithm are not matched in operation, then training the training set, repeating the process, and distilling the verification set into a miniature training set; and
performing bar code recognition based on the artificial intelligence recognition model, wherein the operation steps of performing bar code recognition based on the artificial intelligence recognition model are as follows:
inputting the image with the bar code into a trained artificial intelligence recognition model to position the bar code to obtain a bar code area;
and decoding the bar code area to obtain bar code information.
And arranging the verification set by using a decoding algorithm for verifying the artificial intelligence algorithm. And (3) supervising the operation result of the artificial intelligence algorithm by taking the decoding algorithm as a reference, putting the decoding algorithm and the artificial intelligence algorithm into a training set to train again when the decoding algorithm and the artificial intelligence algorithm are not matched in operation, and repeating the process, wherein the operation steps of distilling the verification set into a miniature training set are as follows:
decoding and testing the barcode scene images in the verification set by using a decoding algorithm to obtain and store corresponding intermediate variables, wherein the intermediate variables comprise calculation method identification values and corresponding time values;
classifying the verification set according to the intermediate variable, wherein the classification identification is a primary type, and the secondary corresponds to the quality;
loading an artificial intelligence recognition model by an artificial intelligence recognition algorithm, and recognizing the classified verification set to obtain corresponding recognition result data, wherein the recognition result data comprises recall rate and accuracy rate;
and screening out corresponding barcode scene images from the identification result data and the actual identification result data to form a new training set, retraining the artificial intelligent identification model from the new training set, and circulating the steps to finally form a training set with stable micro-scale.
In the scheme, the decoding algorithm is used as the teacher role, the artificial intelligence recognition algorithm in the artificial intelligence recognition algorithm model is used as the student role, the classified verification set is recognized, the corresponding recall rate and accuracy after recognition are obtained, corresponding images are screened out according to the recognized actual data effect to form a new training set, the artificial intelligence recognition algorithm model training is carried out again, the whole process is recycled, and the training set is gradually decreased along with the convergence of the result. In the macroscopic effect, the increment of the verification set can be offset, and finally a training set with stable scale can be formed. For the situations of many barcode recognition scenes, many changes and large data volume, the existing method generally combines the newly added pictures with the original picture library, then re-identifies the training, and the workload is increased gradually along with the lapse of time. Compared with the prior art, the method and the device have the advantages that the knowledge distillation effect can be realized, the process of repeated learning and memory is equivalent, the concentrated knowledge base is formed, the automatic verification is realized, the training set is automatically generated, the good application effect can be realized on the scenes with many barcode recognition scenes, many changes and large data volume, and the method and the device can be used for the scenes with many barcode recognition scenes and large data volume
The method comprises the following steps of inputting an image with a bar code into a trained artificial intelligence recognition model to position the bar code, and obtaining a bar code area:
extracting image features, and predicting a position and a category probability value according to the image features;
combining the prediction result to output a code position and a confidence value;
and establishing coordinates, intercepting the bar code according to the coordinates, and determining the category and quality of the bar code according to the category probability value and the confidence value corresponding to the bar code.
The operation steps of decoding the bar code area to obtain bar code information are as follows:
performing corresponding binarization processing according to the category and quality of the bar code;
and decoding according to the bar code characteristic image to obtain bar code information.
In the embodiment, the method comprises the steps of establishing coordinates, intercepting the bar code according to the coordinates, determining the category and the quality of the bar code according to the category probability value and the confidence value corresponding to the bar code, and performing image preprocessing on the intercepted bar code, wherein the image preprocessing comprises but is not limited to denoising, blurring, sharpening, color-to-black conversion, scaling and normalization.
In summary, the invention idea of the application is that an artificial intelligence recognition model and a decoding algorithm are combined, the decoding algorithm can position and solve a two-dimensional code, and the artificial intelligence recognition model can realize quick positioning and quick bar code classification to reduce the judgment process of the decoding algorithm, so that the decoding algorithm is used as a core to educate the artificial intelligence recognition algorithm, thereby not only reducing the work of artificial identification, but also realizing the knowledge distillation effect, being equivalent to a process of repeated learning and memory, forming a concentrated knowledge base, realizing automatic verification, automatically generating a training set, and having good application effects on the conditions of many bar code recognition scenes, many changes and large data volume. Firstly, constructing an artificial intelligence recognition model and training the artificial intelligence recognition model, and then based on the trained artificial intelligence recognition model, recognizing bar codes; finally, decoding the bar code area to obtain bar code information, and identifying the bar codes for logistics based on a trained artificial intelligence identification model without overlapping or switching different conditions in the reasoning process, so that the efficiency of a scene is improved, and the bar codes at random positions are quickly captured; for a multi-code recognition scene, a plurality of bar codes can be simultaneously positioned and recognized, and the method is not limited by the factors of position/background/type difference of the bar codes and defective bar codes.
Example 2
The embodiment provides a bar code decoding device based on artificial intelligence target recognition, including:
the system comprises an acquisition unit, a storage unit and a verification unit, wherein the acquisition unit is used for acquiring a barcode scene image and dividing the barcode scene image into a training set and a verification set, the training set is used for training the artificial intelligence recognition model, and the verification set is used for verifying the artificial intelligence recognition model;
the model building and training unit is used for building an artificial intelligence recognition model and training the artificial intelligence recognition model through the training set;
the verification unit is used for verifying the artificial intelligence recognition model through the verification set, monitoring the operation result of the artificial intelligence algorithm by taking the decoding algorithm as a reference, putting the decoding algorithm and the artificial intelligence recognition model into a training set for retraining and repeating the process, and distilling the verification set into a miniature training set;
the bar code positioning unit is used for inputting the acquired image with the bar code into a trained artificial intelligence recognition model and positioning the bar code;
and the bar code decoding unit is used for decoding the bar code to acquire bar code information.
The bar code positioning unit comprises a feature processing unit, a bar code prediction unit and a bar code processing unit, wherein the feature processing unit is used for extracting and processing image features, and the bar code prediction unit is used for predicting the bar code position and the category probability value according to the image features and outputting the bar code position and the confidence value; the bar code processing unit is used for establishing coordinates according to the bar code position and intercepting the bar code according to the coordinates.
Specifically, the barcode decoding unit comprises a binarization unit and a decoding unit, wherein the binarization unit is used for carrying out binarization processing on the category and quality of the positioned barcode and extracting a barcode feature image; and the decoding unit decodes the bar code characteristic image.
Example 3
The present embodiment provides a computer-readable storage medium storing computer instructions for causing the computer to execute the barcode decoding method based on artificial intelligence object recognition described in embodiment 1.
In the description herein, references to the description of the terms "one embodiment," "certain embodiments," "an illustrative embodiment," "an example," "a specific example," or "some examples," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
While there have been shown and described what are at present considered the fundamental principles and essential features of the invention and its advantages, it will be apparent to those skilled in the art that the invention is not limited to the details of the foregoing exemplary embodiments, but is capable of other specific forms without departing from the spirit or essential characteristics thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein.
Furthermore, it should be understood that although the present description refers to embodiments, not every embodiment may contain only a single embodiment, and such description is for clarity only, and those skilled in the art should integrate the description, and the embodiments may be combined as appropriate to form other embodiments understood by those skilled in the art.
Claims (8)
1. The bar code decoding method based on artificial intelligence target recognition is characterized by comprising the following steps:
establishing an artificial intelligence recognition model, wherein the operation steps of establishing the artificial intelligence recognition model are as follows:
acquiring a barcode scene image to form a training set and a verification set;
the verification set is sorted by using a decoding algorithm and used for verifying the artificial intelligence algorithm, the operation result of the artificial intelligence algorithm is supervised by taking the decoding algorithm as a reference, the two running unmatched results are put into a training set for retraining and the process is repeated, and the verification set is distilled into a miniature training set; the operation steps are as follows: decoding and testing the barcode scene images in the verification set by using a decoding algorithm to obtain and store corresponding intermediate variables, wherein the intermediate variables comprise calculation method identification values and corresponding time values; classifying the verification set according to the intermediate variable, wherein the classification identification is a primary type, and the secondary corresponds to the quality; loading an artificial intelligence recognition model by an artificial intelligence recognition algorithm, and recognizing the classified verification set to obtain corresponding recognition result data, wherein the recognition result data comprises recall rate and accuracy rate; screening out corresponding barcode scene images from the identification result data and the actual identification result data to form a new training set, retraining the artificial intelligent identification model from the new training set, and circulating the steps to finally form a training set with stable micro-scale;
and carrying out bar code identification based on the artificial intelligence identification model, wherein the operation steps of carrying out bar code identification based on the artificial intelligence identification model are as follows:
inputting the image with the bar code into a trained artificial intelligence recognition model to position the bar code to obtain a bar code area;
and decoding the bar code area to obtain bar code information.
2. The method for decoding the bar code based on the artificial intelligence target recognition according to claim 1, wherein the step of inputting the image with the bar code into the trained artificial intelligence recognition model to locate the bar code and obtaining the bar code area comprises the following steps:
extracting image features, and predicting a position and a category probability value according to the image features;
combining the prediction result to output a code position and a confidence value;
and establishing coordinates, intercepting the bar code according to the coordinates, and determining the category and quality of the bar code according to the category probability value and the confidence value corresponding to the bar code.
3. The method for decoding the barcode based on the artificial intelligence target recognition according to claim 2, wherein the operation of decoding the barcode region to obtain the barcode information comprises:
performing corresponding binarization processing according to the category and quality of the bar code, and extracting a bar code characteristic image;
and decoding according to the bar code characteristic image to obtain bar code information.
4. The method of claim 2, wherein the establishing coordinates and intercepting the barcode according to the coordinates, the determining the category and quality of the barcode according to the category probability value and the confidence value corresponding to the barcode further comprises pre-processing the intercepted barcode, the pre-processing including but not limited to de-noising, blurring, sharpening, color-to-black, scaling, and normalizing.
5. Bar code decoding device based on artificial intelligence target identification, its characterized in that includes:
the system comprises an acquisition unit, a storage unit and a verification unit, wherein the acquisition unit is used for acquiring a barcode scene image and dividing the barcode scene image into a training set and a verification set, the training set is used for training the artificial intelligence recognition model, and the verification set is used for verifying the artificial intelligence recognition model;
the model building and training unit is used for building an artificial intelligence recognition model and training the artificial intelligence recognition model through the training set;
the verification unit is used for verifying the artificial intelligence recognition model through the verification set, monitoring the operation result of the artificial intelligence algorithm by taking the decoding algorithm as a reference, putting the decoding algorithm and the artificial intelligence recognition model into a training set for retraining and repeating the process, and distilling the verification set into a miniature training set;
the bar code positioning unit is used for inputting the acquired image with the bar code into a trained artificial intelligence recognition model and positioning the bar code;
and the bar code decoding unit is used for decoding the bar code to acquire bar code information.
6. The barcode decoding device based on artificial intelligence target identification of claim 5, wherein the barcode positioning unit comprises a feature processing unit, a barcode prediction unit and a barcode processing unit, the feature processing unit is used for extracting and processing image features, the barcode prediction unit is used for predicting the barcode position and the class probability value according to the image features and outputting the barcode position and the confidence value; the bar code processing unit is used for establishing coordinates according to the bar code position and intercepting the bar code according to the coordinates.
7. The barcode decoding device based on artificial intelligence target identification as claimed in claim 6, wherein the barcode decoding unit comprises a binarization unit and a decoding unit, the binarization unit is used for performing binarization processing on the category and quality of the located barcode and extracting a barcode feature image; and the decoding unit decodes the bar code characteristic image.
8. A computer-readable storage medium storing computer instructions for causing a computer to execute the artificial intelligence target recognition-based barcode decoding method according to any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111146903.7A CN113591508B (en) | 2021-09-29 | 2021-09-29 | Bar code decoding method and device based on artificial intelligence target recognition and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111146903.7A CN113591508B (en) | 2021-09-29 | 2021-09-29 | Bar code decoding method and device based on artificial intelligence target recognition and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113591508A CN113591508A (en) | 2021-11-02 |
| CN113591508B true CN113591508B (en) | 2022-02-18 |
Family
ID=78242473
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111146903.7A Active CN113591508B (en) | 2021-09-29 | 2021-09-29 | Bar code decoding method and device based on artificial intelligence target recognition and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113591508B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116432676B (en) * | 2023-06-12 | 2023-11-07 | 北京紫光青藤微系统有限公司 | Method and device for decoding bar code and electronic equipment |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111144151A (en) * | 2019-12-25 | 2020-05-12 | 山东财经大学 | High-speed dynamic bar code real-time detection method based on image recognition |
| CN112651289A (en) * | 2020-10-19 | 2021-04-13 | 广东工业大学 | Intelligent identification and verification system and method for value-added tax common invoice |
| CN112989863A (en) * | 2021-03-10 | 2021-06-18 | 北京骑胜科技有限公司 | Two-dimensional code state identification method and device |
| CN113379255A (en) * | 2021-06-15 | 2021-09-10 | 广东鸿土规划设计有限公司 | Method and system for managing irrigation and water conservancy facilities by adopting two-dimensional code tiles |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10055626B2 (en) * | 2016-12-06 | 2018-08-21 | Datalogic Usa, Inc. | Data reading system and method with user feedback for improved exception handling and item modeling |
-
2021
- 2021-09-29 CN CN202111146903.7A patent/CN113591508B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111144151A (en) * | 2019-12-25 | 2020-05-12 | 山东财经大学 | High-speed dynamic bar code real-time detection method based on image recognition |
| CN112651289A (en) * | 2020-10-19 | 2021-04-13 | 广东工业大学 | Intelligent identification and verification system and method for value-added tax common invoice |
| CN112989863A (en) * | 2021-03-10 | 2021-06-18 | 北京骑胜科技有限公司 | Two-dimensional code state identification method and device |
| CN113379255A (en) * | 2021-06-15 | 2021-09-10 | 广东鸿土规划设计有限公司 | Method and system for managing irrigation and water conservancy facilities by adopting two-dimensional code tiles |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113591508A (en) | 2021-11-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110738207A (en) | character detection method for fusing character area edge information in character image | |
| CN107679250A (en) | A kind of multitask layered image search method based on depth own coding convolutional neural networks | |
| Zheng et al. | Research on detecting bearing-cover defects based on improved YOLOv3 | |
| Hu et al. | Defect identification method for poplar veneer based on progressive growing generated adversarial network and MASK R-CNN Model | |
| CN111709311A (en) | Pedestrian re-identification method based on multi-scale convolution feature fusion | |
| CN110969166A (en) | Small target identification method and system in inspection scene | |
| CN111598860A (en) | Lithium battery defect detection method based on yolov3 network embedded in self-attention door module | |
| Luo et al. | CAT-EDNet: Cross-attention transformer-based encoder–decoder network for salient defect detection of strip steel surface | |
| Zhang et al. | Robust facial landmark detection via heatmap-offset regression | |
| CN113591508B (en) | Bar code decoding method and device based on artificial intelligence target recognition and storage medium | |
| Wan et al. | Mixed local channel attention for object detection | |
| CN114219753A (en) | Power equipment surface defect detection method based on deep learning and terminal | |
| Yin et al. | Road Damage Detection and Classification based on Multi-level Feature Pyramids. | |
| Fu et al. | Complementarity-aware Local-global Feature Fusion Network for Building Extraction in Remote Sensing Images | |
| CN116666676A (en) | Intelligent repair method for waste vanadium battery electrolyte | |
| Calefati et al. | Reading meter numbers in the wild | |
| CN116052110A (en) | Intelligent positioning method and system for pavement marking defects | |
| Castillo et al. | Object detection in digital documents based on machine learning algorithms | |
| Bi et al. | CASA-Net: a context-aware correlation convolutional network for scale-adaptive crack detection | |
| Wang et al. | Conditional TransGAN-Based Data Augmentation for PCB Electronic Component Inspection | |
| CN113269052A (en) | Price tag identification method, terminal and storage device | |
| Shukla et al. | Mixed-Type Wafer Classification For Low Memory Devices Using Knowledge Distillation | |
| Gan et al. | Multi-region two-stream deep architecture for visual power monitoring systems | |
| Wu et al. | Express parcel detection based on improved faster regions with CNN features | |
| Saberironaghi | Deep learning models for defect and anomaly detection on industrial surfaces |
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 |