CN109697440A - A kind of ID card information extracting method - Google Patents
A kind of ID card information extracting method Download PDFInfo
- Publication number
- CN109697440A CN109697440A CN201811501998.8A CN201811501998A CN109697440A CN 109697440 A CN109697440 A CN 109697440A CN 201811501998 A CN201811501998 A CN 201811501998A CN 109697440 A CN109697440 A CN 109697440A
- Authority
- CN
- China
- Prior art keywords
- picture
- text
- card
- card information
- follows
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/22—Image preprocessing by selection of a specific region containing or referencing a pattern; Locating or processing of specific regions to guide the detection or recognition
-
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
Abstract
A kind of ID card information extracting method, comprising the following steps: S1: enter ID Card Recognition System, obtain the true face picture of identity card；S2: by VGG16 sorter network, gray processing, normalization, spline interpolation, order filters calculate identity card angle in the picture and correction；S3: being detected by yolo3 and screens effective text information in picture；S4: text information is identified using crnn network end-to-end, and exports result；S5: output result is verified；S6: ID card information extraction system is exited.ID card information extractive technique of the invention can solve the prior art and calibrate inaccuracy to picture angle, and Character segmentation inaccuracy, poor to the generalization ability of varying environment, recognition accuracy is low, the excessively complicated problem of identification process.
Description
Technical field
The present invention relates to information extraction technology field, especially a kind of ID card information extracting method.
Background technique
With the rapid development of information age, as the effective management tool of population information --- identity card,
It has been deep into different social sectors.Identity card is the symbol of China's residential identity.Identity card registration at present mostly uses greatly
The mode of manual entry.Under this is not only time-consuming, and efficiency is very low.Therefore, how utilize computer technology, high speed, effectively,
Completely typing personal information, and managed and verified accordingly as urgent problem in many information systems.
ID card information extraction system has very big demand in service-type bussiness, traffic and public security system, it can accelerate
It completes identity card number and personal information quickly and effectively inputs, and inquired by corresponding information management system, verified
Deng operation, while it can also network and information is uploaded into the Ministry of Public Security, convenient for grasping the trend of floating population, further statistical query
And management.Everything is all beneficial to the Office Information and the networking that push relevant departments.
In ID card information extraction process, how to extract the text position of identity card and identifies that text therein is key,
Common feature extracting method, which has, passes through density, text profile etc. based on center of gravity, coarse grid, projection, stroke, but these are extracted
There is poor anti jamming capability in method, insensitive to lopsided shift transformation.
In ID card information extractive technique, the pretreatment for picture is a considerable step, is influenced final
Recognition effect, pre-processing most important process is the rotational correction to ID Card Image, corrects incorrect can largely effect below
Text Feature Extraction and identification.
In ID card information extractive technique, to extract it is text filed carry out know method for distinguishing be usually the region into
The segmentation of row individual character, and then identify single character, but will appear the character being partitioned into list cutting procedure and increase or reduce, it leads
Cause subsequent text identification result inaccuracy.
Summary of the invention
In view of this, can solve in the prior art the purpose of the invention is to provide a kind of identity card recognition method
The problem of picture rotation inaccuracy, individual character divides not high success rate and poor anti jamming capability.
The technical solution of the present invention is as follows:
A kind of ID card information extracting method, includes the following steps:
S1: enter ID Card Recognition System, obtain the true face picture of identity card；
S2: by VGG16 sorter network, gray processing, normalization, spline interpolation, sequenc-ing cluster calculating identity card is in image
In angle and correction；
S3: being detected by yolo3 and screens effective text information in picture；
S4: text information is identified using crnn network end-to-end, and exports result；
S5: output result is verified；
S6: ID card information extraction system is exited.
Further, the operating process of the step S1 are as follows:
S11: the front picture of camera shooting identity card is used；
S12: picture is uploaded to cloud interface by network.
Further, the operating process of the step S2 are as follows:
S21: photographic model is corrected roughly by training tetra- disaggregated model of VGG16, angle classification: 0 °, 90 °, 180 °,
270°；
S22: gray processing is carried out to the RGB image of input and obtains grayscale image, calculation formula is as follows:
Gray (i, j)=/ 3 (1) [R (i, j)+G (i, j)+B (i, j)]
S23: the pixel of gray processing picture is normalized using Min-max, Xnew=(Xold- min)/(max-min), wherein
XnewIndicate the data after normalization, XoldIndicate the data before normalization, min is indicated in a column feature of the data most
Small value, max indicate the maximum value in a column feature；
S24: carrying out the scaling of corresponding proportion by spline interpolation to image, can be before not losing picture pixels feature
Put the size for reducing picture；
S25: extracting background pixel to picture by order filters, subtracts background area by the subtraction of pixel, most
Only remaining text pixel in picture afterwards；
S26: each integer degree value within the scope of -15 °~15 ° is rotated, at this angle respectively to postrotational
Two-dimensional picture matrix is averaged using the second dimension as axis, obtains one-dimensional mean value array, asks its variance to obtain one-dimensional mean value
Last value records the corresponding variance yields of each angle, and maximum variance is exactly the angle rotated required for picture；
Two-dimensional array is averaged, and formula is as follows, and A [m] [n] is the array of m row n column, and B [m] [1] is the array that m row 1 arranges:
B [m] [1]=mean (A [m] [n]) (2)
One-dimension array asks formula of variance as follows, and B [m] [1] is the array that m row 1 arranges, and C [1] is the constant of 1 row 1 column:
C [1]=var (B [m] [1]) (3).
Further, the operating process of the step S3 are as follows:
S31: by online collection picture, and the text information for using CTPN algorithm to detect trains yolo3 as training set
Text detection model；
S32: by yolo3 model can with the text position in coarse localization picture, obtain text filed location information with
Reliability information；
S33: setting confidence interval threshold is deleted and is lower than the text filed of this threshold value, and duplicate text filed, merging is filtered
The adjacent line of text in left and right, finally filtering is text filed lower than minimum text size, exports remaining text results.
Preferably, the operating process of the step S4 are as follows:
S41: collecting text library required for identity card identification, as training label；
S42: for the textual image under online collection varying environment as training set, the text that training obtains yolo3 is end-to-end
Identification model；
S43: the text that input S4 is obtained obtains output result.
In the present invention, this identity card information extracting system can be fine using the object detection method detection text of yolo3
Ground solves the problems, such as poor anti jamming capability, insensitive to lopsided shift transformation.
In the ID card information extraction system, the thick correction of angle is carried out by VGG16 model, then pass through gray processing,
Normalization, spline interpolation, sequenc-ing cluster accurately correct ID Card Image, and can well solve can not correct picture
The not high problem with correction accuracy.
In the ID card information extraction system, end may be implemented using model of the convolutional network in conjunction with Recursive Networks and arrive
The identification character at end, reduces the operation of separating character, improves accuracy rate.
Compared with prior art, a kind of ID card information extracting method provided by the invention, has the following beneficial effects:
(1) thick angle correction is carried out by tetra- disaggregated model of VGG16, gray processing is carried out to the picture slightly corrected, is normalized,
The operation of spline interpolation, sequenc-ing cluster can carry out accurately angle correction, can solution well can not be to crooked picture
The problem of being corrected and extracting information.
(2) information area that can extract identity card under complex background by yolo3 model solves character position positioning
Poor anti jamming capability, the low problem of universality；
(3) the step of being reduced by crnn model to text filed progress individual character segmentation improves text identification result
Accuracy rate.
Detailed description of the invention
Fig. 1 is flow chart of the method for the present invention.
Specific embodiment
With reference to the accompanying drawings and detailed description, the present invention will be further described.
Referring to Fig.1, a kind of ID card information extracting method carries out ID card information with identity card public security department and is extracted as
Example, includes the following steps:
S1: enter ID Card Recognition System, obtain the true face picture of identity card；
The wherein operating process of step S1 are as follows:
S11: the front picture of camera shooting identity card is used；
S12: picture is uploaded to cloud interface by network；
S2: by VGG16 sorter network, gray processing, normalization, spline interpolation, order filters, which calculate identity card, is scheming
Angle and correction as in；
The wherein operating process of step S2 are as follows:
S21: photographic model is corrected roughly by training tetra- disaggregated model of VGG16, angle classification: 0 °, 90 °, 180 °,
270°；
S22: gray processing is carried out to the RGB image of input and obtains grayscale image, calculation formula is as follows:
Gray (i, j)=/ 3 (1) [R (i, j)+G (i, j)+B (i, j)]
S23: the pixel of gray processing picture is normalized using Min-max, Xnew=(Xold- min)/(max-min), wherein
XnewIndicate the data after normalization, XoldIndicate the data before normalization, min is indicated in a column feature of the data most
Small value, max indicate the maximum value in a column feature；
S24: carrying out the scaling of corresponding proportion by spline interpolation to image, can be before not losing picture pixels feature
Put the size for reducing picture；
S25: extracting background pixel to picture by order filters, subtracts background area by the subtraction of pixel, most
Only remaining text pixel in picture afterwards；
S26: each integer degree value within the scope of -15 °~15 ° is rotated, at this angle respectively to postrotational
Two-dimensional picture matrix is averaged using the second dimension as axis, obtains one-dimensional mean value array, asks its variance to obtain one-dimensional mean value
Last value records the corresponding variance yields of each angle, and maximum variance is exactly the angle rotated required for picture；
Two-dimensional array is averaged, and formula is as follows, and A [m] [n] is the array of m row n column, and B [m] [1] is the array that m row 1 arranges:
B [m] [1]=mean (A [m] [n]) (2)
One-dimension array asks formula of variance as follows, and B [m] [1] is the array that m row 1 arranges, and C [1] is the constant of 1 row 1 column:
C [1]=var (B [m] [1]) (3)
S3: being detected by yolo3 and screens effective text information in picture；
The wherein operating process of step S3 are as follows:
S31: by online collection picture, and the text information for using CTPN algorithm to detect trains yolo3 as training set
Text detection model；
S32: by yolo3 model can with the text position in coarse localization picture, obtain text filed location information with
Reliability information；
S33: setting confidence interval threshold is deleted and is lower than the text filed of this threshold value, and duplicate text filed, merging is filtered
The adjacent line of text in left and right, finally filtering is text filed lower than minimum text size, exports remaining text results；
S4: text information is identified using crnn network end-to-end, and exports result；
The wherein operating process of step S4 are as follows:
S41: collecting text library required for identity card identification, as training label；
S42: for the textual image under online collection varying environment as training set, the text that training obtains yolo3 is end-to-end
Identification model；
S43: the text that input S4 is obtained obtains output result；
S5: output result is verified；
S6: ID card information extraction system is exited.
The present invention provides a kind of ID card information extracting method, shoots identity card front picture by camera, is uploaded to
Cloud carries out the thick correction of angle by tetra- disaggregated model of VGG16 to picture, then uses gray processing, normalization, and batten is inserted
Value, the method for sequenc-ing cluster carry out accurately angle correction to picture, efficiently solve picture correction is unsuccessful or correction essence
Spend low problem.The position of text can be accurately positioned out by yolo3 model, also have for complicated background picture very high
Accuracy rate can solve the poor anti jamming capability of text position extraction, the low problem of universality.It may be implemented by crnn model
Character identification function end to end removes the step of individual character is cut from, can solve since it is desired that identifying knot caused by individual character is cut
The problem of fruit declines.Last output result is adjusted and exported, the program improves ID card information extraction well
Accuracy rate.
The above description is only an embodiment of the present invention, is not intended to limit the scope of the invention, all to utilize this hair
Equivalent structure or equivalent flow shift made by bright specification and accompanying drawing content is applied directly or indirectly in other relevant skills
Art field, is included within the scope of the present invention.
Claims (5)
1. a kind of ID card information extracting method, which is characterized in that the described method comprises the following steps:
S1: enter ID Card Recognition System, obtain the true face picture of identity card；
S2: by VGG16 sorter network, gray processing, normalization, spline interpolation, sequenc-ing cluster calculates identity card in the picture
Angle is simultaneously corrected；
S3: being detected by yolo3 and screens effective text information in picture；
S4: text information is identified using crnn network end-to-end, and exports result；
S5: output result is verified；
S6: ID card information extraction system is exited.
2. a kind of ID card information extracting method as described in claim 1, which is characterized in that the operating process of the step S1
Are as follows:
S11: the front picture of camera shooting identity card is used；
S12: picture is uploaded to cloud interface by network.
3. a kind of ID card information extracting method as claimed in claim 1 or 2, which is characterized in that the operation of the step S2
Process are as follows:
S21: photographic model, angle classification: 0 °, 90 °, 180 °, 270 ° are corrected roughly by training tetra- disaggregated model of VGG16；
S22: gray processing is carried out to the RGB image of input and obtains grayscale image, calculation formula is as follows:
Gray (i, j)=/ 3 (1) [R (i, j)+G (i, j)+B (i, j)]
S23: the pixel of gray processing picture is normalized using Min-max, Xnew=(Xold- min)/(max-min), wherein Xnew
Indicate the data after normalization, XoldIndicate the data before normalization, min indicates the minimum in a column feature of the data
Value, max indicate the maximum value in a column feature；
S24: carrying out the scaling of corresponding proportion by spline interpolation to image, can be under the premise of not losing picture pixels feature
Reduce the size of picture；
S25: extracting background pixel to picture by order filters, subtracts background area by the subtraction of pixel, finally schemes
Only remaining text pixel in piece；
S26: each integer degree value within the scope of -15 °~15 ° is rotated, at this angle respectively to postrotational two dimension
Picture matrix average using the second dimension as axis, obtain one-dimensional mean value array, ask its variance to obtain to the end one-dimensional mean value
Value, record the corresponding variance yields of each angle, maximum variance is exactly the angle rotated required for picture；
Two-dimensional array is averaged, and formula is as follows, and A [m] [n] is the array of m row n column, and B [m] [1] is the array that m row 1 arranges:
B [m] [1]=mean (A [m] [n]) (2)
One-dimension array asks formula of variance as follows, and B [m] [1] is the array that m row 1 arranges, and C [1] is the constant of 1 row 1 column:
C [1]=var (B [m] [1]) (3).
4. a kind of ID card information extracting method as claimed in claim 1 or 2, which is characterized in that the operation of the step S3
Process are as follows:
S31: by online collection picture, and the text information for using CTPN algorithm to detect trains yolo3 text as training set
Detection model；
S32: by yolo3 model can with the text position in coarse localization picture, obtain text filed location information with it is credible
Spend information；
S33: setting confidence interval threshold is deleted and is lower than the text filed of this threshold value, filters duplicate text filed, merging left and right
Adjacent line of text, finally filtering is text filed lower than minimum text size, exports remaining text results.
5. a kind of ID card information extracting method as claimed in claim 1 or 2, which is characterized in that the operation of the step S4
Process are as follows:
S41: collecting text library required for identity card identification, as training label
S42: for the textual image under online collection varying environment as training set, training obtains the end-to-end identification of text of yolo3
Model；
S43: the text that input S4 is obtained obtains output result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811501998.8A CN109697440A (en) | 2018-12-10 | 2018-12-10 | A kind of ID card information extracting method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811501998.8A CN109697440A (en) | 2018-12-10 | 2018-12-10 | A kind of ID card information extracting method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109697440A true CN109697440A (en) | 2019-04-30 |
Family
ID=66230490
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811501998.8A Pending CN109697440A (en) | 2018-12-10 | 2018-12-10 | A kind of ID card information extracting method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109697440A (en) |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110276253A (en) * | 2019-05-15 | 2019-09-24 | 中国科学院信息工程研究所 | A kind of fuzzy literal detection recognition method based on deep learning |
| CN110298343A (en) * | 2019-07-02 | 2019-10-01 | 哈尔滨理工大学 | A kind of hand-written blackboard writing on the blackboard recognition methods |
| CN110458162A (en) * | 2019-07-25 | 2019-11-15 | 上海兑观信息科技技术有限公司 | A kind of method of intelligent extraction pictograph information |
| CN110569839A (en) * | 2019-08-09 | 2019-12-13 | 河海大学常州校区 | Bank card number identification method based on CTPN and CRNN |
| CN110598566A (en) * | 2019-08-16 | 2019-12-20 | 深圳中兴网信科技有限公司 | Image processing method, device, terminal and computer readable storage medium |
| CN110738238A (en) * | 2019-09-18 | 2020-01-31 | 平安科技（深圳）有限公司 | certificate information classification positioning method and device |
| CN110807455A (en) * | 2019-09-19 | 2020-02-18 | 平安科技（深圳）有限公司 | Bill detection method, device and equipment based on deep learning and storage medium |
| CN110969154A (en) * | 2019-11-29 | 2020-04-07 | 上海眼控科技股份有限公司 | Text recognition method and device, computer equipment and storage medium |
| CN111738979A (en) * | 2020-04-29 | 2020-10-02 | 北京易道博识科技有限公司 | Automatic certificate image quality inspection method and system |
| CN111860522A (en) * | 2020-07-23 | 2020-10-30 | 中国平安人寿保险股份有限公司 | Identity card picture processing method and device, terminal and storage medium |
| CN111950554A (en) * | 2020-08-17 | 2020-11-17 | 深圳市丰巢网络技术有限公司 | Identification card identification method, device, equipment and storage medium |
| CN112016547A (en) * | 2020-08-20 | 2020-12-01 | 上海天壤智能科技有限公司 | Image character recognition method, system and medium based on deep learning |
| CN112560850A (en) * | 2019-09-26 | 2021-03-26 | 中电万维信息技术有限责任公司 | Automatic identity card information extraction and authenticity verification method based on custom template |
| CN113011409A (en) * | 2021-04-02 | 2021-06-22 | 北京世纪好未来教育科技有限公司 | Image identification method and device, electronic equipment and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106991421A (en) * | 2017-03-22 | 2017-07-28 | 湖南联信科技有限公司 | A kind of ID card information extraction system |
| CN107665354A (en) * | 2017-09-19 | 2018-02-06 | 北京小米移动软件有限公司 | Identify the method and device of identity card |
| CN108446699A (en) * | 2018-02-08 | 2018-08-24 | 东华大学 | Identity card pictorial information identifying system under a kind of complex scene |
-
2018
- 2018-12-10 CN CN201811501998.8A patent/CN109697440A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106991421A (en) * | 2017-03-22 | 2017-07-28 | 湖南联信科技有限公司 | A kind of ID card information extraction system |
| CN107665354A (en) * | 2017-09-19 | 2018-02-06 | 北京小米移动软件有限公司 | Identify the method and device of identity card |
| CN108446699A (en) * | 2018-02-08 | 2018-08-24 | 东华大学 | Identity card pictorial information identifying system under a kind of complex scene |
Non-Patent Citations (2)
| Title |
|---|
| WENLIHAOYU: "chineseocr", 《HTTPS://GITHUB.COM/CHINESEOCR/CHINESEOCR/TREE/716CE957F38FD2B3E5FDF6C9E26661ACF7482268》 * |
| 汪一文: "深度卷积神经网络在OCR问题中的应用研究", 《中国优秀硕士学位论文全文数据库》 * |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110276253A (en) * | 2019-05-15 | 2019-09-24 | 中国科学院信息工程研究所 | A kind of fuzzy literal detection recognition method based on deep learning |
| CN110298343A (en) * | 2019-07-02 | 2019-10-01 | 哈尔滨理工大学 | A kind of hand-written blackboard writing on the blackboard recognition methods |
| CN110458162A (en) * | 2019-07-25 | 2019-11-15 | 上海兑观信息科技技术有限公司 | A kind of method of intelligent extraction pictograph information |
| CN110569839A (en) * | 2019-08-09 | 2019-12-13 | 河海大学常州校区 | Bank card number identification method based on CTPN and CRNN |
| CN110569839B (en) * | 2019-08-09 | 2023-05-16 | 河海大学常州校区 | Bank card number identification method based on CTPN and CRNN |
| CN110598566A (en) * | 2019-08-16 | 2019-12-20 | 深圳中兴网信科技有限公司 | Image processing method, device, terminal and computer readable storage medium |
| WO2021051553A1 (en) * | 2019-09-18 | 2021-03-25 | 平安科技（深圳）有限公司 | Certificate information classification and positioning method and apparatus |
| CN110738238A (en) * | 2019-09-18 | 2020-01-31 | 平安科技（深圳）有限公司 | certificate information classification positioning method and device |
| CN110738238B (en) * | 2019-09-18 | 2023-05-26 | 平安科技（深圳）有限公司 | Classification positioning method and device for certificate information |
| CN110807455A (en) * | 2019-09-19 | 2020-02-18 | 平安科技（深圳）有限公司 | Bill detection method, device and equipment based on deep learning and storage medium |
| CN112560850A (en) * | 2019-09-26 | 2021-03-26 | 中电万维信息技术有限责任公司 | Automatic identity card information extraction and authenticity verification method based on custom template |
| CN110969154A (en) * | 2019-11-29 | 2020-04-07 | 上海眼控科技股份有限公司 | Text recognition method and device, computer equipment and storage medium |
| CN111738979A (en) * | 2020-04-29 | 2020-10-02 | 北京易道博识科技有限公司 | Automatic certificate image quality inspection method and system |
| CN111738979B (en) * | 2020-04-29 | 2024-01-19 | 北京易道博识科技有限公司 | Certificate image quality automatic checking method and system |
| CN111860522A (en) * | 2020-07-23 | 2020-10-30 | 中国平安人寿保险股份有限公司 | Identity card picture processing method and device, terminal and storage medium |
| CN111860522B (en) * | 2020-07-23 | 2024-02-02 | 中国平安人寿保险股份有限公司 | Identity card picture processing method, device, terminal and storage medium |
| CN111950554A (en) * | 2020-08-17 | 2020-11-17 | 深圳市丰巢网络技术有限公司 | Identification card identification method, device, equipment and storage medium |
| CN112016547A (en) * | 2020-08-20 | 2020-12-01 | 上海天壤智能科技有限公司 | Image character recognition method, system and medium based on deep learning |
| CN113011409A (en) * | 2021-04-02 | 2021-06-22 | 北京世纪好未来教育科技有限公司 | Image identification method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109697440A (en) | A kind of ID card information extracting method | |
| CN109241894B (en) | Bill content identification system and method based on form positioning and deep learning | |
| US10510152B2 (en) | Systems, methods, and devices for image matching and object recognition in images using textures | |
| CN107067006B (en) | Verification code identification method and system serving for data acquisition | |
| CN107153817B (en) | Pedestrian re-identification data labeling method and device | |
| CN107944427B (en) | Dynamic face recognition method and computer readable storage medium | |
| CN114119676B (en) | Target detection tracking identification method and system based on multi-feature information fusion | |
| CN103679147A (en) | Method and device for identifying model of mobile phone | |
| CN107481374B (en) | Intelligent terminal fingerprint unblock door opener | |
| CN111310662B (en) | Flame detection and identification method and system based on integrated deep network | |
| CN103488983A (en) | Business card OCR data correction method and system based on knowledge base | |
| CN103778409A (en) | Human face identification method based on human face characteristic data mining and device | |
| CN111968152B (en) | Dynamic identity recognition method and device | |
| CN111783757A (en) | OCR technology-based identification card recognition method in complex scene | |
| CN116030396B (en) | Accurate segmentation method for video structured extraction | |
| CN112784834A (en) | Automatic license plate identification method in natural scene | |
| CN110490185A (en) | One kind identifying improved method based on repeatedly comparison correction OCR card information | |
| TW202215296A (en) | System and method for license plate recognition and method for extracting character segments in a license plate | |
| CN111160107B (en) | Dynamic region detection method based on feature matching | |
| CN114359538A (en) | Water meter reading positioning and identifying method | |
| CN112258545A (en) | Tobacco leaf image online background processing system and online background processing method | |
| CN116524725B (en) | Intelligent driving traffic sign image data identification system | |
| CN110826390A (en) | Video data processing method based on face vector characteristics | |
| Mi et al. | A novel video text extraction approach based on multiple frames | |
| CN111832497A (en) | Text detection post-processing method based on geometric features |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190430 |
|
| RJ01 | Rejection of invention patent application after publication |