CN114417904A - Bar code identification method based on deep learning and book retrieval system - Google Patents
Bar code identification method based on deep learning and book retrieval system Download PDFInfo
- Publication number
- CN114417904A CN114417904A CN202210052226.0A CN202210052226A CN114417904A CN 114417904 A CN114417904 A CN 114417904A CN 202210052226 A CN202210052226 A CN 202210052226A CN 114417904 A CN114417904 A CN 114417904A
- Authority
- CN
- China
- Prior art keywords
- bar code
- image
- character
- probability
- identification
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 39
- 238000013135 deep learning Methods 0.000 title claims abstract description 17
- 238000013528 artificial neural network Methods 0.000 claims abstract description 39
- 238000012545 processing Methods 0.000 claims abstract description 25
- 238000005457 optimization Methods 0.000 claims abstract description 10
- 230000009466 transformation Effects 0.000 claims abstract description 9
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 7
- 238000001514 detection method Methods 0.000 claims abstract description 7
- 238000012549 training Methods 0.000 claims description 13
- 230000008859 change Effects 0.000 claims description 7
- 230000000306 recurrent effect Effects 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000005070 sampling Methods 0.000 claims description 3
- 230000011218 segmentation Effects 0.000 claims description 3
- 238000013527 convolutional neural network Methods 0.000 claims description 2
- 238000004590 computer program Methods 0.000 description 9
- 238000010586 diagram Methods 0.000 description 8
- 230000006870 function Effects 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 238000003860 storage Methods 0.000 description 3
- 239000003086 colorant Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 108091026890 Coding region Proteins 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000004806 packaging method and process Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 239000000047 product Substances 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
- 230000037303 wrinkles Effects 0.000 description 1
Images
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
- G06K7/146—Methods for optical code recognition the method including quality enhancement steps
- G06K7/1482—Methods for optical code recognition the method including quality enhancement steps using fuzzy logic or natural solvers, such as neural networks, genetic algorithms and simulated annealing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
-
- 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
- G06N3/045—Combinations of networks
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Health & Medical Sciences (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Probability & Statistics with Applications (AREA)
- Automation & Control Theory (AREA)
- Fuzzy Systems (AREA)
- Quality & Reliability (AREA)
- Electromagnetism (AREA)
- Toxicology (AREA)
- Character Discrimination (AREA)
Abstract
The invention discloses a bar code identification method and a book retrieval system based on deep learning in the field of bar code identification, which are used for extracting the characteristics of a bar code image in a detection image; carrying out binarization processing on the characteristics of the bar code image; correcting the inclined bar code of the bar code by utilizing Hough transformation to obtain a bar code image to be identified; extracting an identification sequence from a bar code image to be identified by adopting a trained full convolution neural network, and inputting the identification sequence into an RNN (neural network) to obtain a character element sequence identified by each channel; obtaining probability result of character classification according to correlation among characters in the character sequence; acquiring the total probability of the characters corresponding to each character element sequence by adopting a global optimization algorithm, and outputting the character with the maximum probability to finish accurately identifying the bar code; clear bar code images to be identified are obtained through an image processing mode, and the identification precision of the bar codes is effectively improved through deep learning.
Description
Technical Field
The invention belongs to the field of bar code identification, and particularly relates to a bar code identification method and a book retrieval system based on deep learning.
Background
Barcodes were first proposed by Woodland in 1949, and in recent years, barcode technology has been widely used in many aspects of life with the development and popularization of computer technology. Barcodes follow specific coding rules: each black and white stripe represents a character and respectively represents different character codes according to different widths; several groups of black and white alternate stripes, namely a plurality of characters, form a character, different codes follow a specific character coding sequence, and the character comprises a plurality of data characters from a starting character, and finally comprises a check code character and an ending character; according to the character sequence in the bar code, the original information can be decoded bit by referring to the codebook.
The traditional digital image processing method splits the relation between the adjacent characters and characters in the bar code, loses the favorable context information, is limited in adaptability, accuracy and the like, is easily influenced by the bar code quality, stain wrinkles, scanning distance and the like, has generally high rejection rate on the one hand, and needs manual repeated scanning or even manual input for supplement; on the other hand, code recognition errors caused by insufficient accuracy also cause troubles for subsequent control and digital management. Because bar code technology is not stable and reliable enough, it is necessary to use a more flat and standard carton surface rather than soft plastic packaging, which brings high cost overhead.
Disclosure of Invention
The invention aims to provide a barcode identification method based on deep learning, which is used for acquiring a clear barcode image to be identified in an image processing mode and effectively improving the identification precision of a barcode through the deep learning.
In order to achieve the purpose, the technical scheme adopted by the invention is as follows:
the invention provides a bar code identification method based on deep learning on one hand, which comprises the following steps:
extracting the characteristics of the bar code image according to the gray scale change in the vertical direction and the horizontal direction in the detected image;
determining a threshold value of the bar code image characteristic binarization processing by adopting a maximum inter-class variance method, and performing binarization processing on the bar code image characteristic;
correcting the inclined bar code of the bar code by utilizing Hough transformation to obtain a bar code image to be identified;
extracting an identification sequence from a barcode image to be identified by adopting a trained full convolution neural network, inputting the identification sequence into a pre-trained RNN (neural network), obtaining character element sequences identified by each channel of the RNN through the RNN neural network, obtaining probability results of character element classification according to the correlation among character elements in the character element sequences, obtaining the full probability of characters corresponding to each character element sequence by adopting a global optimization algorithm, and outputting the characters with the maximum probability to finish accurate barcode identification.
Preferably, the barcode image features are extracted according to the gray scale changes in the vertical and horizontal directions in the detection image; the method comprises the following steps:
and acquiring a transverse differential gray value and a longitudinal differential gray value in the detected image through a bar code reader, taking the ratio of the transverse differential gray value to the longitudinal differential gray value as a characteristic parameter, and extracting an image area with the characteristic parameter larger than a set value as the bar code image characteristic.
Preferably, the method for determining the threshold value of the binarization processing of the barcode image features by using the maximum inter-class variance method comprises the following steps:
setting a gray value k as a threshold, wherein the class A and the class B respectively consist of pixel points with gray values from 0 to k-1 and from k to 255, and the variance between the classes is as follows:
σ2(κ)=ωA(μA-μ)2+ωB(μB-μ)2
μ=μAωA+μBωB
where μ is the mean gray value of the barcode image feature, ωA,ωBIs a summary of the appearance of class A and class BRate, muA,μBIs the mean gray value of class a and class B;
to sigma2And (kappa) carrying out optimization to obtain the maximum value of the gray value k as the threshold value of the binary processing segmentation.
Preferably, the Hough transform is used for correcting the oblique bar code of the bar code to obtain a bar code image to be identified, and the method comprises the following steps:
and dividing the characteristics of the bar code image into a plurality of bar code areas according to the longitudinal differential gray value, respectively calculating the inclination angle, and respectively correcting each bar code area.
Preferably, the trained full convolution neural network is adopted to extract the identification sequence of the barcode image to be identified, and the method comprises the following steps:
adopting a trained full convolution neural network to set the height sampling of the barcode image to be identified as 1, and acquiring the change of the gray value in the width direction of the barcode image to be identified as an identification sequence;
the training of the full convolutional neural network comprises: acquiring a historical barcode image from a database as a training data set; the full convolution neural network is trained through a training data set.
Preferably, the RNN recurrent neural network is processed by softmax to obtain a probability result of classifying the neurons.
Preferably, a global optimization algorithm is adopted to obtain the total probability of the characters corresponding to each character element sequence, and the characters with the maximum probability are output to finish accurate bar code recognition, wherein the method comprises the following steps:
converting the character element probability into the form of information entropy, and summing the information entropy corresponding to each character element sequence, wherein the calculation formula is as follows:
wherein, PnTo representProbability of each class of each character, Sn=logPnThe information entropy of each character element classification probability transformation is represented, and n is the number of character element classifications;
and acquiring the character element sequence of the maximum information entropy sum value and outputting corresponding characters to finish accurately identifying the bar code.
The invention provides a book retrieval system on the other hand, which comprises a bar code reader and a bar code identification device; the bar code reader is electrically connected with the bar code recognition device; the bar code reader is used for collecting a detection image containing a bar code on a book; the bar code identification device realizes the bar code identification method to obtain a bar code identification result.
Compared with the prior art, the invention has the beneficial effects that:
according to the invention, according to the correlation among all characters in the character sequence, the probability result of character classification is firstly obtained, the total probability of the characters corresponding to all the character sequences is deduced, then the characters with the maximum probability are output to finish the accurate bar code identification, and the algorithm of global optimization is provided in the network according to the structural characteristics of bar code coding design, so that the acceptance rate and the accuracy of the bar code can be effectively improved.
The invention extracts the characteristics of the bar code image in the detection image by utilizing the gray scale change in the vertical direction and the horizontal direction; determining a threshold value of the bar code image characteristic binarization processing by adopting a maximum inter-class variance method, and performing binarization processing on the bar code image characteristic; correcting the inclined bar code of the bar code by utilizing Hough transformation to obtain a bar code image to be identified; interference of other colors to identification can be effectively avoided through binarization processing, and identification errors caused by bar code distortion can be prevented through correction of the bar code, so that the identification accuracy is improved through an image processing mode.
Drawings
Fig. 1 is a flowchart of a barcode identification method based on deep learning according to the present invention.
Detailed Description
The invention is further described below with reference to the accompanying drawings. The following examples are only for illustrating the technical solutions of the present invention more clearly, and the protection scope of the present invention is not limited thereby.
Example one
As shown in fig. 1, a barcode identification method based on deep learning includes:
extracting the characteristics of the bar code image according to the gray scale change in the vertical direction and the horizontal direction in the detected image; and acquiring a transverse differential gray value and a longitudinal differential gray value in the detected image through a bar code reader, taking the ratio of the transverse differential gray value to the longitudinal differential gray value as a characteristic parameter, and extracting an image area with the characteristic parameter larger than a set value as the bar code image characteristic.
Determining a threshold value of the bar code image characteristic binarization processing by adopting a maximum inter-class variance method, and performing binarization processing on the bar code image characteristic, wherein the method comprises the following steps:
setting a gray value k as a threshold, wherein the class A and the class B respectively consist of pixel points with gray values from 0 to k-1 and from k to 255, and the variance between the classes is as follows:
σ2(κ)=ωA(μA-μ)2+ωB(μB-μ)2
μ=μAωA+μBωB
where μ is the mean gray value of the barcode image feature, ωA,ωBIs the probability of occurrence of class A and class B, μA,μBIs the mean gray value of class a and class B;
to sigma2(k) Carrying out optimization solution to obtain the maximum value of the gray value k as a threshold value for binarization processing segmentation; the invention can effectively avoid the interference of other colors to the identification through the binarization processing.
Dividing the characteristics of the bar code image into a plurality of bar code areas according to the longitudinal differential gray value, and respectively calculating the inclination angle, thereby utilizing Hough transformation to correct the inclined bar code of the bar code to obtain the bar code image to be recognized, and preventing the recognition error caused by the bar code distortion by correcting the inclined bar code of the bar code
Acquiring a historical barcode image from a database as a training data set; training the full convolution neural network through a training data set; adopting a trained full convolution neural network to set the height sampling of a barcode image to be recognized as 1, acquiring the change of gray value in the width direction of the barcode image to be recognized as a recognition sequence, inputting the recognition sequence into a pre-trained RNN (neural network) and acquiring character element sequences recognized by four channels of the RNN neural network; and according to the correlation among all the characters in the character sequence, the RNN recurrent neural network obtains a probability result of classifying the characters through softmax processing.
Wherein the RNN recurrent neural network training comprises: and training the RNN recurrent neural network through a training data set, wherein the RNN recurrent neural network acquires character sequences and correlation among characters from the training data set.
Adopting a global optimization algorithm to obtain the total probability of the characters corresponding to each character element sequence, wherein the method comprises the following steps:
the character sequence is composed of 7 characters, and the joint probability is obtained by calculating the product of the probabilities of 7 characters, so as to { p }1,p2,p3,p4Denotes the probability value of character four classification, by xi=logpiConverting the probability into the form of information entropy, converting the character element probability into the form of information entropy, and adding the information entropy corresponding to each character element sequence, wherein the calculation formula is as follows:
Pnrepresenting the probability, S, of each class of each charactern=logPnThe information entropy of each character element classification probability transformation is represented, and n is the number of character element classifications;
and acquiring the character element sequence of the maximum information entropy sum value and outputting corresponding characters to finish accurately identifying the bar code.
Example two
As shown in fig. 1, a book retrieval system includes a bar code reader and a bar code recognition device; the bar code reader is electrically connected with the bar code recognition device; the bar code reader is used for collecting a detection image containing a bar code on a book; the bar code identification device stores a computer program; when the computer program is executed by the bar code identification device, the bar code identification method of the embodiment one is realized to obtain a bar code identification result; when the database has no related character data, the book retrieval system outputs rejection, and when the database has no related character data, the book retrieval system outputs related information corresponding to the book, thereby assisting the user in returning the book.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
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 instruction means 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.
The above description is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, several modifications and variations can be made without departing from the technical principle of the present invention, and these modifications and variations should also be regarded as the protection scope of the present invention.
Claims (8)
1. A barcode identification method based on deep learning is characterized by comprising the following steps:
extracting the characteristics of the bar code image according to the gray scale change in the vertical direction and the horizontal direction in the detected image;
determining a threshold value of the bar code image characteristic binarization processing by adopting a maximum inter-class variance method, and performing binarization processing on the bar code image characteristic;
correcting the inclined bar code of the bar code by utilizing Hough transformation to obtain a bar code image to be identified;
extracting an identification sequence from a barcode image to be identified by adopting a trained full convolution neural network, inputting the identification sequence into a pre-trained RNN (neural network), obtaining character element sequences identified by each channel of the RNN through the RNN neural network, obtaining probability results of character element classification according to the correlation among character elements in the character element sequences, obtaining the full probability of characters corresponding to each character element sequence by adopting a global optimization algorithm, and outputting the characters with the maximum probability to finish accurate barcode identification.
2. The barcode recognition method based on deep learning of claim 1, wherein the barcode image features are extracted according to vertical and horizontal gray scale changes in the detection image, and the method comprises:
and acquiring a transverse differential gray value and a longitudinal differential gray value in the detected image through a bar code reader, taking the ratio of the transverse differential gray value to the longitudinal differential gray value as a characteristic parameter, and extracting an image area with the characteristic parameter larger than a set value as the bar code image characteristic.
3. The barcode recognition method based on deep learning of claim 2, wherein a threshold value of the binarization processing of the barcode image features is determined by a maximum inter-class variance method, and the method comprises:
setting a gray value k as a threshold, wherein the class A and the class B respectively consist of pixel points with gray values from 0 to k-1 and from k to 255, and the variance between the classes is as follows:
σ2(κ)=ωA(μA-μ)2+ωB(μB-μ)2
μ=μAωA+μBωB
where μ is the average gray scale value of the barcode image feature, ωA,ωBIs the probability of occurrence of class A and class B, μA,μBIs the mean gray value of class a and class B;
to sigma2And (kappa) carrying out optimization to obtain the maximum value of the gray value k as the threshold value of the binary processing segmentation.
4. The deep learning-based bar code identification method according to claim 3, wherein the inclined bar code of the bar code is corrected by Hough transformation to obtain a bar code image to be identified, and the method comprises the following steps:
and dividing the characteristics of the bar code image into a plurality of bar code areas according to the longitudinal differential gray value, respectively calculating the inclination angle, and respectively correcting each bar code area.
5. The deep learning-based barcode recognition method according to claim 1 or 4, wherein a trained full convolution neural network is adopted to extract a recognition sequence for a barcode image to be recognized, and the method comprises the following steps:
adopting a trained full convolution neural network to set the height sampling of the barcode image to be identified as 1, and acquiring the change of the gray value in the width direction of the barcode image to be identified as an identification sequence;
the training of the full convolutional neural network comprises: acquiring a historical barcode image from a database as a training data set; the full convolution neural network is trained through a training data set.
6. The deep learning-based bar code identification method according to claim 5, wherein the RNN recurrent neural network is subjected to softmax processing to obtain probability results of classifying the character elements.
7. The deep learning-based bar code recognition method according to claim 6, wherein a global optimization algorithm is used to obtain the total probability of the characters corresponding to each character element sequence, and the character with the highest output probability completes accurate bar code recognition, and the method comprises:
converting the character element probability into the form of information entropy, and summing the information entropy corresponding to each character element sequence, wherein the calculation formula is as follows:
wherein, PnRepresenting the probability, S, of each class of each charactern=logPnThe information entropy of each character element classification probability transformation is represented, and n is the number of character element classifications;
and acquiring the character element sequence of the maximum information entropy sum value and outputting corresponding characters to finish accurately identifying the bar code.
8. A book retrieval system comprises a bar code reader and a bar code identification device; the bar code reader is electrically connected with the bar code recognition device; the bar code reader is used for collecting a detection image containing a bar code on a book; the bar code recognition device is characterized in that the bar code recognition device realizes the bar code recognition method of any one of claims 1 to 7 to obtain a bar code recognition result.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210052226.0A CN114417904A (en) | 2022-01-18 | 2022-01-18 | Bar code identification method based on deep learning and book retrieval system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210052226.0A CN114417904A (en) | 2022-01-18 | 2022-01-18 | Bar code identification method based on deep learning and book retrieval system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114417904A true CN114417904A (en) | 2022-04-29 |
Family
ID=81273170
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210052226.0A Pending CN114417904A (en) | 2022-01-18 | 2022-01-18 | Bar code identification method based on deep learning and book retrieval system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114417904A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115578464A (en) * | 2022-12-07 | 2023-01-06 | 深圳思谋信息科技有限公司 | Bar code identification method and device, computer equipment and readable storage medium |
CN116110037A (en) * | 2023-04-11 | 2023-05-12 | 深圳市华图测控系统有限公司 | Book checking method and device based on visual identification and terminal equipment |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101916357A (en) * | 2010-09-03 | 2010-12-15 | 西安富立叶微电子有限责任公司 | Laser barcode reading device and barcode reading method |
US20180181785A1 (en) * | 2015-09-07 | 2018-06-28 | Fujian Landi Commercial Equipment Co., Ltd. | Detection Method and System for Characteristic Patterns of Han Xin Codes |
CN111368576A (en) * | 2020-03-12 | 2020-07-03 | 成都信息工程大学 | Code128 bar Code automatic reading method based on global optimization |
CN113112503A (en) * | 2021-05-10 | 2021-07-13 | 上海贝德尔生物科技有限公司 | Method for realizing automatic detection of medicine label based on machine vision |
-
2022
- 2022-01-18 CN CN202210052226.0A patent/CN114417904A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101916357A (en) * | 2010-09-03 | 2010-12-15 | 西安富立叶微电子有限责任公司 | Laser barcode reading device and barcode reading method |
US20180181785A1 (en) * | 2015-09-07 | 2018-06-28 | Fujian Landi Commercial Equipment Co., Ltd. | Detection Method and System for Characteristic Patterns of Han Xin Codes |
CN111368576A (en) * | 2020-03-12 | 2020-07-03 | 成都信息工程大学 | Code128 bar Code automatic reading method based on global optimization |
CN113112503A (en) * | 2021-05-10 | 2021-07-13 | 上海贝德尔生物科技有限公司 | Method for realizing automatic detection of medicine label based on machine vision |
Non-Patent Citations (1)
Title |
---|
曾欣科 等: "基于全局优化与深度学习的条形码识别方法", 《计算机应用》 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115578464A (en) * | 2022-12-07 | 2023-01-06 | 深圳思谋信息科技有限公司 | Bar code identification method and device, computer equipment and readable storage medium |
CN116110037A (en) * | 2023-04-11 | 2023-05-12 | 深圳市华图测控系统有限公司 | Book checking method and device based on visual identification and terminal equipment |
CN116110037B (en) * | 2023-04-11 | 2023-06-23 | 深圳市华图测控系统有限公司 | Book checking method and device based on visual identification and terminal equipment |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
EP2943911B1 (en) | Process of handwriting recognition and related apparatus | |
CN110619274A (en) | Identity verification method and device based on seal and signature and computer equipment | |
US8587685B2 (en) | Method and apparatus for retrieving label | |
CN112651289B (en) | Value-added tax common invoice intelligent recognition and verification system and method thereof | |
CN114417904A (en) | Bar code identification method based on deep learning and book retrieval system | |
CN111046859B (en) | Character recognition method and device | |
CN111476109A (en) | Bill processing method, bill processing apparatus, and computer-readable storage medium | |
CN111639646A (en) | Test paper handwritten English character recognition method and system based on deep learning | |
CN111652233B (en) | Text verification code automatic identification method aiming at complex background | |
CN112307919B (en) | Improved YOLOv 3-based digital information area identification method in document image | |
CN110598581B (en) | Optical music score recognition method based on convolutional neural network | |
CN116704209B (en) | Quick flange contour extraction method and system | |
CN114333338A (en) | License plate identification system and method | |
CN115620312A (en) | Cross-modal character handwriting verification method, system, equipment and storage medium | |
CN115116074A (en) | Handwritten character recognition and model training method and device | |
CN113657377B (en) | Structured recognition method for mechanical bill image | |
CN117195928B (en) | Bar code positioning detection method based on deep learning | |
JP2000331120A (en) | Device and method for recognizing character and recording medium stored with control program therefor | |
CN107679505B (en) | Method for realizing rejection of handwritten character | |
CN111368576B (en) | Global optimization-based Code128 bar Code automatic reading method | |
CN114495135A (en) | Bill identification method and device | |
CN114610940A (en) | Hash image retrieval method based on local random sensitivity self-encoder | |
CN109871910B (en) | Handwritten character recognition method and device | |
Nelson et al. | A modified GA classifier for offline Tamil handwritten character recognition | |
CN114821134B (en) | Method for identifying print style number of publication based on template matching |
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 |