CN109919060A - A kind of identity card content identifying system and method based on characteristic matching - Google Patents
A kind of identity card content identifying system and method based on characteristic matching Download PDFInfo
- Publication number
- CN109919060A CN109919060A CN201910143343.6A CN201910143343A CN109919060A CN 109919060 A CN109919060 A CN 109919060A CN 201910143343 A CN201910143343 A CN 201910143343A CN 109919060 A CN109919060 A CN 109919060A
- Authority
- CN
- China
- Prior art keywords
- image
- identity card
- recognition result
- card
- 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.)
- Pending
Links
Abstract
The embodiment of the present invention proposes a kind of identity card content identifying system and method based on characteristic matching, is related to technical field of image processing.The system comprises image pre-processing units, are successively classified to image to be detected comprising ID Card Image, are obtained and correction process;Image detecting element determines the region of character image in identity card in the ID Card Image after acquisition and correction;Image identification unit identifies the text in identity card in text graphics field, obtains intermediate recognition result;Image processing unit matches recognition result and is corrected, and generates final recognition result;Data set unit is stored with for trained training set of images and for matched address data set.Have the advantages that speed is fast, identification is accurate and high-efficient.
Description
Technical field
The present invention relates to technical field of image processing, in particular to a kind of identity card content based on characteristic matching
Identifying system and method.
Background technique
Deep learning is the main stream approach of current natural scene text detection and identification.And use the method pair of deep learning
Identity card carries out text detection and recognition effect is poor always.This is because identity card there are front and back sides, it is based on this particularity, body
Text region algorithm on part card has several difficult points:
1. the picture of identity card is often in a complicated background, and there are various illumination, the problem of angle.How from again
The front and back sides of identity card are determined in miscellaneous background, and it is a very big difficult point that it is plucked out from original image and is corrected
2. the character script on identity card is relatively special, there are some erroneous detections, missing inspection in existing detection method
3. the character background on identity card is relatively special, there are some identification mistakes in existing recognition methods
4. the address part on identity card is successively divided generally according to province, city and region, therefore can be carried out to recognition result certain
Correction, current existing certain methods often have ignored this point.
The art solutions of Patent No. CN200510029791.1 are as follows: including second generation identity card card reader, camera shooting
Equipment, recognition of face device, identifying data information bank, comprising the following steps: the first step is read by second generation identity card reader
Information in identity card;Information read-out in the first step is transferred in recognition of face device by second step;Third step, camera shooting are set
It is standby that real human face is imaged, and real human face image is passed in recognition of face device;4th step, recognition of face device is according to
Two steps read the information in identity card by second generation identity card reader and export the corresponding letter provided in identifying data information bank
Breath;5th step identifies the true and false of the second generation identity card, if personal identification card document is really to carry out the 6th step again;6th step, is distinguished
Whether the second generation identity card is not held in person.The present invention can be used to distinguish whether the second generation identity card is held in person
Have.
It only identifies facial image, does not account for the identification to text information in identity card.
Summary of the invention
The purpose of the present invention is to provide a kind of identity card content identifying system and method based on characteristic matching has speed
The advantage that degree is fast, identification is accurate and high-efficient.
To achieve the goals above, technical solution used in the embodiment of the present invention is as follows:
In a first aspect, the embodiment of the invention provides a kind of identity card content identifying system based on characteristic matching, described
System includes:
Image pre-processing unit, successively classified to image to be detected comprising ID Card Image, obtain and correction at
Reason;
Image detecting element determines the region of character image in identity card in the ID Card Image after acquisition and correction;
Image identification unit identifies the text in identity card in text graphics field, obtains intermediate recognition result;
Image processing unit matches recognition result and is corrected, and generates final recognition result;
Data set unit is stored with for trained training set of images and for matched address data set.
Further, described image pretreatment unit includes:
Image classification unit is classified to image to be detected comprising ID Card Image, is determined in image to be detected
ID Card Image part is identity card front or reverse side;
Image obtains and correction unit obtains ID Card Image in determining sorted ID Card Image to be detected
Region, and the region of the ID Card Image of acquisition is corrected.
Further, the data set unit includes:
Training set of images unit is stored with for trained training set of images;
Address database collection unit, is stored with the address data set for matching and correcting.
Further, described image detection unit includes:
Feature extraction unit extracts characteristic pattern from several images in training set of images;
Feature synthesis unit merges the characteristic pattern of extraction;
Output unit, detection model of the output for detection;
Detection unit detects the ID Card Image after acquisition and correction according to detection model, determines identity
The region of character image in card.
Further, described image recognition unit includes:
Convolution unit extracts characteristic sequence from several images in training set of images;
Cycling element predicts the label distribution of the characteristic sequence extracted from convolution unit;
The label distribution that cycling element obtains is converted into intermediate identification knot by operations such as duplicate removal, integration by transcriptional units
Fruit.
Further, described image processing unit includes:
Distracter in intermediate recognition result is filtered by filter element according to intermediate recognition result, is obtained filtering and is known
Other result;
Fuzzy matching unit carries out fuzzy matching to filtering recognition result according to address data set, obtains fuzzy matching knot
Fruit;
Secondary correction unit: secondary correction is carried out to fuzzy matching result, generates final recognition result.
Second aspect, the embodiment of the invention also provides a kind of identity card content identification method based on characteristic matching, institute
The method of stating includes:
A method of the identity card content identifying system based on characteristic matching, which comprises
Step 1: successively being classified to image to be detected comprising ID Card Image, obtained and correction process;
Step 2: in the ID Card Image after acquisition and correction, determining the region of character image in identity card;
Step 3: the text in identification identity card in text graphics field obtains intermediate recognition result;
Step 4: recognition result being matched and is corrected, and generates final recognition result.
Further, in the step 1, successively classified to image to be detected comprising ID Card Image, obtain and
The method of correction process includes: according to identity card front and back sides template picture, to the spy of image to be detected comprising ID Card Image
Sign point is detected, and is then carried out the matching of characteristic point and then is determined the front and back sides of identity card, then from picture to be identified really
Determine the region of identity card and is corrected.
Further, in the step 2, in the ID Card Image after acquisition and correction, text figure in identity card is determined
The method in the region of picture includes: to collect a large amount of identity card template picture, and carry out mark filtering, constitutes training set of images;
Then training set of images training detection model is utilized;By the detection model to the ID Card Image after acquisition and correction at
Reason, determines the region of character image in identity card.
Further, in the step 4, recognition result is matched and is corrected, and generate the side of final recognition result
Method includes: to summarize address field composition rule, constructs corresponding regular expression according to the address field composition rule, then sharp
Intermediate recognition result is handled with these regular expressions, filters out obvious incoherent distracter, obtains filtering identification
As a result, the address date for filtering recognition result and address data set storage is then carried out fuzzy matching, finally to fuzzy
Secondary correction is carried out with obtained result, to obtain final recognition result.
A kind of identity card content identifying system one kind based on characteristic matching provided in an embodiment of the present invention is based on feature
The identity card content identifying system and method matched, have the advantages that image to be detected comprising ID Card Image according to
It is secondary to be classified, obtained and correction process, improve the positive and negative region feature detection of identity card accuracy and with picture to be identified
Matched speed.Great amount of samples picture is acquired as image data collection, recycles the data set to be trained, effectively solves existing
The problem of erroneous detection existing for detection method and missing inspection, and can rapidly and accurately capture the character area on identity card.
Great amount of samples picture is acquired as image data collection, the data set is recycled to be trained, existing recognizer is effectively solved and knows
The not low problem of rate;Scalping is carried out using summarize 200 a plurality of regularities, carries out mould followed by 100,000 a plurality of address bases
Paste matching, finally carries out secondary correction, improves the accuracy rate of address field recognition result well.
To enable the above objects, features and advantages of the present invention to be clearer and more comprehensible, preferred embodiment is cited below particularly, and cooperate
Appended attached drawing, is described in detail below.
Detailed description of the invention
In order to illustrate the technical solution of the embodiments of the present invention more clearly, below will be to needed in the embodiment attached
Figure is briefly described, it should be understood that the following drawings illustrates only certain embodiments of the present invention, therefore is not construed as pair
The restriction of range for those of ordinary skill in the art without creative efforts, can also be according to this
A little attached drawings obtain other relevant attached drawings.
Fig. 1 shows the system structure of the identity card content identifying system provided in an embodiment of the present invention based on characteristic matching
Schematic diagram.
The pretreatment list of Fig. 2 shows the provided in an embodiment of the present invention identity card content identifying system based on characteristic matching
Meta structure schematic diagram.
Fig. 3 shows the data set list of the identity card content identifying system provided in an embodiment of the present invention based on characteristic matching
Meta structure schematic diagram.
Fig. 4 shows the image detection of the identity card content identifying system provided in an embodiment of the present invention based on characteristic matching
Cellular construction schematic diagram.
Fig. 5 shows the image recognition of the identity card content identifying system provided in an embodiment of the present invention based on characteristic matching
Cellular construction schematic diagram.
Fig. 6 shows the image procossing of the identity card content identifying system provided in an embodiment of the present invention based on characteristic matching
Cellular construction schematic diagram.
Fig. 7 shows the method flow of the identity card content identification method provided in an embodiment of the present invention based on characteristic matching
Schematic diagram.
Specific embodiment
Below in conjunction with attached drawing in the embodiment of the present invention, technical solution in the embodiment of the present invention carries out clear, complete
Ground description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.Usually exist
The component of the embodiment of the present invention described and illustrated in attached drawing can be arranged and be designed with a variety of different configurations herein.Cause
This, is not intended to limit claimed invention to the detailed description of the embodiment of the present invention provided in the accompanying drawings below
Range, but it is merely representative of selected embodiment of the invention.Based on the embodiment of the present invention, those skilled in the art are not doing
Every other embodiment obtained under the premise of creative work out, shall fall within the protection scope of the present invention.
It should also be noted that similar label and letter indicate similar terms in following attached drawing, therefore, once a certain Xiang Yi
It is defined in a attached drawing, does not then need that it is further defined and explained in subsequent attached drawing.Meanwhile of the invention
In description, term " first ", " second " etc. are only used for distinguishing description, are not understood to indicate or imply relative importance.
Embodiment 1:
As shown in Figure 1, a kind of identity card content identifying system based on characteristic matching, the system comprises:
Image pre-processing unit, successively classified to image to be detected comprising ID Card Image, obtain and correction at
Reason;
Image detecting element determines the region of character image in identity card in the ID Card Image after acquisition and correction;
Image identification unit identifies the text in identity card in text graphics field, obtains intermediate recognition result;
Image processing unit matches recognition result and is corrected, and generates final recognition result;
Data set unit is stored with for trained training set of images and for matched address data set.
Further, described image pretreatment unit includes:
Image classification unit is classified to image to be detected comprising ID Card Image, is determined in image to be detected
ID Card Image part is identity card front or reverse side;
Image obtains and correction unit obtains ID Card Image in determining sorted ID Card Image to be detected
Region, and the region of the ID Card Image of acquisition is corrected.
Image pre-processing unit can be in such a way that SURF algorithm and FLANN adaptation combine.The combination of the two can be with
Solve the problems, such as the front and back sides of accurate judgement identity card and positioning from complex background and illumination.Body is carried out using SURF algorithm
Then the characteristic point detection of part card front and back sides template picture and picture to be identified carries out the matching of characteristic point and then determining identity
The front and back sides of card.Then, the region of identity card is plucked out from picture to be identified using FLANN adaptation and is corrected.
Further, the data set unit includes:
Training set of images unit is stored with for trained training set of images;
Address database collection unit, is stored with the address data set for matching and correcting.
Further, described image detection unit includes:
Feature extraction unit extracts characteristic pattern from several images in training set of images;
Feature synthesis unit merges the characteristic pattern of extraction;
Output unit, detection model of the output for detection;
Detection unit detects the ID Card Image after acquisition and correction according to detection model, determines identity
The region of character image in card.
Image detecting element mainly solves the problems, such as erroneous detection, missing inspection.It collects and simulates a large amount of identity card picture, as picture
Training set, and carry out mark filtering.Then our detection model of these image datas training is utilized.What detection model was selected
It is improved EAST model, input is the picture of 512*512 pixel, is obtained after then stopping by four of ResNet network
Characteristics map cascaded, finally obtain the characteristics map of a 128*128*3 pixel, then this feature map returned
Return and classifies.
Further, described image recognition unit includes:
Convolution unit extracts characteristic sequence from several images in training set of images;
Cycling element predicts the label distribution of the characteristic sequence extracted from convolution unit;
The label distribution that cycling element obtains is converted into intermediate identification knot by operations such as duplicate removal, integration by transcriptional units
Fruit.
Image identification unit mainly solves the problems, such as identification mistake.For identity card background the characteristics of, simulation generates big
The training picture of amount constitutes image data training set, and the accuracy rate of identification model is then promoted using image data training set.Know
The part CNN is changed to DenseNet network, exported as the characteristics map of 32*280 pixel, so by the CRNN model of other model refinement
The extraction that two-way LSTM network carries out sequence signature is inputted afterwards, and last we obtain final recognition result using CTC layers.
Further, described image processing unit includes:
Distracter in intermediate recognition result is filtered by filter element according to intermediate recognition result, is obtained filtering and is known
Other result;
Fuzzy matching unit carries out fuzzy matching to filtering recognition result according to address data set, obtains fuzzy matching knot
Fruit;
Secondary correction unit: secondary correction is carried out to fuzzy matching result, generates final recognition result.
Image processing unit mainly solves the problems, such as address field identification mistake.Firstly, summarizing a address about more than 200
Field composition rule, then we filter according to the corresponding regular expression of rule construct followed by these regular expressions
Fall obvious incoherent distracter.According to the composition rule of identity card address field, building one includes 100,000 a plurality of records
The filtered result of regular expression and the content of address base are carried out fuzzy matching, finally obtained to fuzzy matching by address base
Result carry out secondary correction, to obtain final recognition result.
Embodiment 2:
The embodiment of the invention also provides a kind of identity card content identification method based on characteristic matching, the method packet
It includes:
Step 1: successively being classified to image to be detected comprising ID Card Image, obtained and correction process;
Step 2: in the ID Card Image after acquisition and correction, determining the region of character image in identity card;
Step 3: the text in identification identity card in text graphics field obtains intermediate recognition result;
Step 4: recognition result being matched and is corrected, and generates final recognition result.
Further, in the step 1, successively classified to image to be detected comprising ID Card Image, obtain and
The method of correction process includes: according to identity card front and back sides template picture, to the spy of image to be detected comprising ID Card Image
Sign point is detected, and is then carried out the matching of characteristic point and then is determined the front and back sides of identity card, then from picture to be identified really
Determine the region of identity card and is corrected.
The detection and description that characteristic point is carried out first with SURF algorithm carry out characteristic point followed by FLANN adaptation
Matching, finally carry out match point correction.SURF algorithm has preferable robustness to situations such as rotation, illumination of object, and
Faster compared with calculating speed for SIFT algorithm.Every width figure is set and detects 2000 characteristic points.FLANN is quick estimation arest neighbors
Library.It contains some optimization algorithms for search quick neighbour and high dimensional feature in large data sets.It large data sets when
It waits faster than BFMatcher.2 are set by the optimal match point of each descriptor.
Further, in the step 2, in the ID Card Image after acquisition and correction, text figure in identity card is determined
The method in the region of picture includes: to collect a large amount of identity card template picture, and carry out mark filtering, constitutes training set of images;
Then training set of images training detection model is utilized;By the detection model to the ID Card Image after acquisition and correction at
Reason, determines the region of character image in identity card.
Using ResNet50 network as backbone network, cascade merging then is carried out to feature, final output returns frame.Mould
What type inputted is the picture of 512*512, and the characteristics map output after stopping by ResNet first is 256*256*64, warp
Cross second, ResNet network stop after characteristics map output be 64*64*256, by ResNet network third blocking
Characteristics map output later is 32*32*512, and the characteristics map output after stopping by the 4th, ResNet network is 16*
16*1024 then to every layer of characteristics map carries out that pond operation is taken then to cascade respectively at upper one layer of output, final
Characteristics map to a 128*128*3 is returned and is classified.
In step 3, characteristics map is extracted first with DenseNet121 network, it is special followed by BiLstm abstraction sequence
Sign, the output finally predicted using CTC layers.The input of model is the picture of 512*512, after Dense Block1
The size of output is 256*256, is 128*128 by the size exported after Dense Block2, by Dense Block3
The size exported later is 64*64, is input in LSTM layers two-way, and the size of final output is 32*128, Transcription
The final determination of Layer progress recognition result.
Further, in the step 4, recognition result is matched and is corrected, and generate the side of final recognition result
Method includes: to summarize address field composition rule, constructs corresponding regular expression according to the address field composition rule, then sharp
Intermediate recognition result is handled with these regular expressions, filters out obvious incoherent distracter, obtains filtering identification
As a result, the address date for filtering recognition result and address data set storage is then carried out fuzzy matching, finally to fuzzy
Secondary correction is carried out with obtained result, to obtain final recognition result.
Firstly, we summarize a address field composition rule about more than 200, then we are corresponding just according to rule construct
Then expression formula is filtered followed by the result that these regular expressions provide identification network, is weeded out obvious uncorrelated
Distracter.According to the composition rule of identity card address field, we construct the address base comprising 100,000 a plurality of records, will
The filtered result of regular expression and the content of address base carry out fuzzy matching, finally carry out to the result that fuzzy matching obtains
Secondary correction, to obtain final recognition result.
In several embodiments provided herein, it should be understood that disclosed device and method can also pass through
Other modes are realized.The apparatus embodiments described above are merely exemplary, for example, flow chart and block diagram in attached drawing
Show the device of multiple embodiments according to the present invention, the architectural framework in the cards of method and computer program product,
Function and operation.In this regard, each box in flowchart or block diagram can represent the one of a unit, program segment or code
Part, a part of the unit, program segment or code, which includes that one or more is for implementing the specified logical function, to be held
Row instruction.It should also be noted that function marked in the box can also be to be different from some implementations as replacement
The sequence marked in attached drawing occurs.For example, two continuous boxes can actually be basically executed in parallel, they are sometimes
It can execute in the opposite order, this depends on the function involved.It is also noted that every in block diagram and or flow chart
The combination of box in a box and block diagram and or flow chart can use the dedicated base for executing defined function or movement
It realizes, or can realize using a combination of dedicated hardware and computer instructions in the system of hardware.
In addition, each functional unit in each embodiment of the present invention can integrate one independent portion of formation together
Point, it is also possible to each unit individualism, an independent part can also be integrated to form with two or more units.
It, can be with if the function is realized in the form of SFU software functional unit and when sold or used as an independent product
It is stored in a computer readable storage medium.Based on this understanding, technical solution of the present invention is substantially in other words
The part of the part that contributes to existing technology or the technical solution can be embodied in the form of software products, the meter
Calculation machine software product is stored in a storage medium, including some instructions are used so that a computer equipment (can be a
People's computer, server or network equipment etc.) it performs all or part of the steps of the method described in the various embodiments of the present invention.
And storage medium above-mentioned includes: that USB flash disk, mobile hard disk, read-only memory (ROM, Read-Onl8Memor8), arbitrary access are deposited
The various media that can store program code such as reservoir (RAM, Random Access Memor8), magnetic or disk.
It should be noted that, in this document, relational terms such as first and second and the like are used merely to a reality
Body or operation are distinguished with another entity or operation, are deposited without necessarily requiring or implying between these entities or operation
In any actual relationship or order or sequence.Moreover, the terms "include", "comprise" or its any other variant are intended to
Non-exclusive inclusion, so that the process, method, article or equipment including a series of elements is not only wanted including those
Element, but also including other elements that are not explicitly listed, or further include for this process, method, article or equipment
Intrinsic element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that
There is also other identical elements in process, method, article or equipment including the element.
The foregoing is only a preferred embodiment of the present invention, is not intended to restrict the invention, for the skill of this field
For art personnel, the invention may be variously modified and varied.All within the spirits and principles of the present invention, made any to repair
Change, equivalent replacement, improvement etc., should all be included in the protection scope of the present invention.It should also be noted that similar label and letter exist
Similar terms are indicated in following attached drawing, therefore, once being defined in a certain Xiang Yi attached drawing, are then not required in subsequent attached drawing
It is further defined and explained.
Claims (10)
1. a kind of identity card content identifying system based on characteristic matching, which is characterized in that the system comprises:
Image pre-processing unit is successively classified to image to be detected comprising ID Card Image, is obtained and correction process;
Image detecting element determines the region of character image in identity card in the ID Card Image after acquisition and correction;
Image identification unit identifies the text in identity card in text graphics field, obtains intermediate recognition result;
Image processing unit matches recognition result and is corrected, and generates final recognition result;
Data set unit is stored with for trained training set of images and for matched address data set.
2. the identity card content identifying system based on characteristic matching as described in claim 1, which is characterized in that described image is pre-
Processing unit includes:
Image classification unit classifies to image to be detected comprising ID Card Image, determines the identity in image to be detected
Card image section is identity card front or reverse side;
Image obtains and correction unit obtains the region of ID Card Image in determining sorted ID Card Image to be detected,
And the region of the ID Card Image of acquisition is corrected.
3. the identity card content identifying system based on characteristic matching as described in claim 1, which is characterized in that the data set
Unit includes:
Training set of images unit is stored with for trained training set of images;
Address database collection unit, is stored with the address data set for matching and correcting.
4. the identity card content identifying system based on characteristic matching as claimed in claim 3, which is characterized in that described image inspection
Surveying unit includes:
Feature extraction unit extracts characteristic pattern from several images in training set of images;
Feature synthesis unit merges the characteristic pattern of extraction;
Output unit, detection model of the output for detection;
Detection unit detects the ID Card Image after acquisition and correction according to detection model, determines in identity card
The region of character image.
5. the identity card content identifying system based on characteristic matching as claimed in claim 4, which is characterized in that described image is known
Other unit includes:
Convolution unit extracts characteristic sequence from several images in training set of images;
Cycling element predicts the label distribution of the characteristic sequence extracted from convolution unit;
Transcriptional units are distributed the label that cycling element obtains and are converted into intermediate recognition result by operations such as duplicate removal, integration.
6. the identity card content identifying system based on characteristic matching as claimed in claim 5, which is characterized in that at described image
Managing unit includes:
Distracter in intermediate recognition result is filtered by filter element according to intermediate recognition result, obtains filtering identification knot
Fruit;
Fuzzy matching unit carries out fuzzy matching to filtering recognition result according to address data set, obtains fuzzy matching result;
Secondary correction unit: secondary correction is carried out to fuzzy matching result, generates final recognition result.
7. a kind of method based on the identity card content identifying system described in one of claim 1 to 6 based on characteristic matching,
It is characterized in that, which comprises
Step 1: successively being classified to image to be detected comprising ID Card Image, obtained and correction process;
Step 2: in the ID Card Image after acquisition and correction, determining the region of character image in identity card;
Step 3: the text in identification identity card in text graphics field obtains intermediate recognition result;
Step 4: recognition result being matched and is corrected, and generates final recognition result.
8. the identity card content identification method based on characteristic matching as claimed in claim 7, which is characterized in that the step 1
In, successively classified to image to be detected comprising ID Card Image, obtained and the method for correction process includes: according to identity
Front and back sides template picture is demonstrate,proved, the characteristic point of image to be detected comprising ID Card Image is detected, characteristic point is then carried out
Matching so that determine identity card front and back sides, then from picture to be identified determine identity card region and be corrected.
9. the identity card content identification method based on characteristic matching as claimed in claim 8, which is characterized in that the step 2
In, in the ID Card Image after acquisition and correction, the method for determining the region of character image in identity card includes: to collect largely
Identity card template picture, and carry out mark filtering, constitute training set of images;Then training set of images training detection mould is utilized
Type;The ID Card Image after acquisition and correction is handled by the detection model, determines the area of character image in identity card
Domain.
10. the identity card content identification method based on characteristic matching as claimed in claim 9, which is characterized in that the step 4
In, the method for being matched and being corrected to recognition result, and generate final recognition result includes: to summarize address field to constitute rule
Then, corresponding regular expression is constructed according to the address field composition rule, centre is identified followed by these regular expressions
As a result it is handled, filters out obvious incoherent distracter, obtain filtering recognition result, it then will filtering recognition result and ground
Data centrally stored address date in location carries out fuzzy matching, finally carries out secondary correction to the result that fuzzy matching obtains, from
And obtain final recognition result.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910143343.6A CN109919060A (en) | 2019-02-26 | 2019-02-26 | A kind of identity card content identifying system and method based on characteristic matching |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910143343.6A CN109919060A (en) | 2019-02-26 | 2019-02-26 | A kind of identity card content identifying system and method based on characteristic matching |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109919060A true CN109919060A (en) | 2019-06-21 |
Family
ID=66962495
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910143343.6A Pending CN109919060A (en) | 2019-02-26 | 2019-02-26 | A kind of identity card content identifying system and method based on characteristic matching |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109919060A (en) |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110399845A (en) * | 2019-07-29 | 2019-11-01 | 上海海事大学 | Continuously at section text detection and recognition methods in a kind of image |
CN110889365A (en) * | 2019-11-21 | 2020-03-17 | 上海烟草集团有限责任公司 | Label information obtaining method and device |
CN110991440A (en) * | 2019-12-11 | 2020-04-10 | 易诚高科(大连)科技有限公司 | Pixel-driven mobile phone operation interface text detection method |
CN111046883A (en) * | 2019-12-05 | 2020-04-21 | 吉林大学 | Intelligent evaluation method and system based on ancient coin image |
CN111353492A (en) * | 2020-03-12 | 2020-06-30 | 上海合合信息科技发展有限公司 | Image identification and information extraction method and device for standardized document |
CN111401249A (en) * | 2020-03-17 | 2020-07-10 | 中国科学院计算技术研究所厦门数据智能研究院 | Object re-recognition method based on different granularity feature matching consistency |
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 |
CN111783757A (en) * | 2020-06-01 | 2020-10-16 | 成都科大极智科技有限公司 | OCR technology-based identification card recognition method in complex scene |
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 |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103488983A (en) * | 2013-09-13 | 2014-01-01 | 复旦大学 | Business card OCR data correction method and system based on knowledge base |
CN107247950A (en) * | 2017-06-06 | 2017-10-13 | 电子科技大学 | A kind of ID Card Image text recognition method based on machine learning |
CN108304838A (en) * | 2017-05-11 | 2018-07-20 | 腾讯科技(深圳)有限公司 | A kind of pictorial information recognition methods and terminal |
-
2019
- 2019-02-26 CN CN201910143343.6A patent/CN109919060A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103488983A (en) * | 2013-09-13 | 2014-01-01 | 复旦大学 | Business card OCR data correction method and system based on knowledge base |
CN108304838A (en) * | 2017-05-11 | 2018-07-20 | 腾讯科技(深圳)有限公司 | A kind of pictorial information recognition methods and terminal |
CN107247950A (en) * | 2017-06-06 | 2017-10-13 | 电子科技大学 | A kind of ID Card Image text recognition method based on machine learning |
Non-Patent Citations (2)
Title |
---|
四川省土地学会: "《四川省土地学会2013年学术交流年会论文集》", 31 August 2014 * |
张艺玮 等: "结合密集神经网络与长短时记忆模型的中文识别", 《计算机系统应用》 * |
Cited By (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110399845A (en) * | 2019-07-29 | 2019-11-01 | 上海海事大学 | Continuously at section text detection and recognition methods in a kind of image |
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 |
CN110889365A (en) * | 2019-11-21 | 2020-03-17 | 上海烟草集团有限责任公司 | Label information obtaining method and device |
CN111046883B (en) * | 2019-12-05 | 2022-08-23 | 吉林大学 | Intelligent assessment method and system based on ancient coin image |
CN111046883A (en) * | 2019-12-05 | 2020-04-21 | 吉林大学 | Intelligent evaluation method and system based on ancient coin image |
CN110991440A (en) * | 2019-12-11 | 2020-04-10 | 易诚高科(大连)科技有限公司 | Pixel-driven mobile phone operation interface text detection method |
CN110991440B (en) * | 2019-12-11 | 2023-10-13 | 易诚高科(大连)科技有限公司 | Pixel-driven mobile phone operation interface text detection method |
CN111353492A (en) * | 2020-03-12 | 2020-06-30 | 上海合合信息科技发展有限公司 | Image identification and information extraction method and device for standardized document |
CN111353492B (en) * | 2020-03-12 | 2023-10-13 | 上海合合信息科技股份有限公司 | Image recognition and information extraction method and device for standardized document |
CN111401249A (en) * | 2020-03-17 | 2020-07-10 | 中国科学院计算技术研究所厦门数据智能研究院 | Object re-recognition method based on different granularity feature matching consistency |
CN111401249B (en) * | 2020-03-17 | 2023-08-08 | 中科(厦门)数据智能研究院 | Object re-identification method based on matching consistency of features with different granularities |
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 |
CN111783757A (en) * | 2020-06-01 | 2020-10-16 | 成都科大极智科技有限公司 | OCR technology-based identification card recognition method in complex scene |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109919060A (en) | A kind of identity card content identifying system and method based on characteristic matching | |
CN110263774B (en) | A kind of method for detecting human face | |
CN108334848B (en) | Tiny face recognition method based on generation countermeasure network | |
CN110008909B (en) | Real-name system business real-time auditing system based on AI | |
CN106599907A (en) | Multi-feature fusion-based dynamic scene classification method and apparatus | |
CN106228129A (en) | A kind of human face in-vivo detection method based on MATV feature | |
CN107145862A (en) | A kind of multiple features matching multi-object tracking method based on Hough forest | |
CN110070090A (en) | A kind of logistic label information detecting method and system based on handwriting identification | |
CN106295532B (en) | A kind of human motion recognition method in video image | |
CN108764269A (en) | A kind of cross datasets pedestrian recognition methods again based on space-time restriction incremental learning | |
CN111046886A (en) | Automatic identification method, device and equipment for number plate and computer readable storage medium | |
CN108197644A (en) | A kind of image-recognizing method and device | |
CN104809451B (en) | A kind of person's handwriting identification system based on stroke curvature measuring | |
CN107622489A (en) | A kind of distorted image detection method and device | |
CN104573711B (en) | The image understanding method of object and scene based on text objects scene relation | |
CN111445459A (en) | Image defect detection method and system based on depth twin network | |
Zhu et al. | Deep residual text detection network for scene text | |
CN109902223A (en) | A kind of harmful content filter method based on multi-modal information feature | |
CN108154132A (en) | A kind of identity card text extraction method, system and equipment and storage medium | |
CN106355154A (en) | Method for detecting frequent pedestrian passing in surveillance video | |
CN106326451B (en) | A kind of webpage heat transfer agent block decision method of view-based access control model feature extraction | |
Sapkota et al. | Large scale unconstrained open set face database | |
CN115062186B (en) | Video content retrieval method, device, equipment and storage medium | |
Bajić et al. | Chart classification using simplified VGG model | |
CN113963399A (en) | Personnel trajectory retrieval method and device based on multi-algorithm fusion application |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190621 |