CN106934408A - Identity card picture sorting technique based on convolutional neural networks - Google Patents
Identity card picture sorting technique based on convolutional neural networks Download PDFInfo
- Publication number
- CN106934408A CN106934408A CN201511009006.6A CN201511009006A CN106934408A CN 106934408 A CN106934408 A CN 106934408A CN 201511009006 A CN201511009006 A CN 201511009006A CN 106934408 A CN106934408 A CN 106934408A
- Authority
- CN
- China
- Prior art keywords
- identity card
- neural networks
- convolutional neural
- pixels
- photo
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computational Linguistics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Evolutionary Biology (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The present invention provides a kind of identity card picture sorting technique based on convolutional neural networks, including step：Multifarious identity card picture sample set is obtained, and is divided into training sample set, checking sample set and test sample collection；Set up convolutional neural networks；The samples pictures that training sample is concentrated are pre-processed；The convolutional neural networks are input into using pretreated picture as training data, the study for having supervision is carried out, the parameter of each layer of the convolutional neural networks after being trained；Using the parameter of each layer of the convolutional neural networks after training, initialisation structures identical convolutional neural networks obtain the image recognition network with identity card picture classification feature.The present invention utilizes multifarious samples pictures, convolutional neural networks is targetedly trained, is adjusted, and can reach classification accuracy higher.
Description
Technical field
The present invention relates to a kind of identity card picture sorting technique based on convolutional neural networks, belong to pattern-recognition
Technical field.
Background technology
Identity card is the unique certificate for proving holder's identity, is typically necessary when particular transaction is handled
Show identity card or Copy of ID Card as proof of identification, various mechanisms, such as bank, business hall, examine
Test system etc. is equipped with ID Card Recognition System, the true and false for recognizing identity card, and recognition methods is usually
First shift to an earlier date the characteristic information on identity card original paper or identity card picture, the characteristic information that then will be extracted with
Corresponding characteristic information is compared in the property data base known, is verified if matching, is otherwise verified and do not pass through,
This method more takes, and the accuracy rate of identification is not high.
The content of the invention
In view of the foregoing, it is an object of the invention to provide a kind of identity card photograph based on convolutional neural networks
Convolutional neural networks are targetedly trained, adjusted by piece sorting technique using multifarious samples pictures
It is whole, photo classification accuracy rate higher can be reached.
To achieve the above object, the present invention uses following technical scheme：
A kind of identity card picture sorting technique based on convolutional neural networks, including：
S1：Multifarious identity card picture sample set is obtained, and is divided into training sample set, checking sample set
And test sample collection；
S2：Set up convolutional neural networks；
S3：The samples pictures that training sample is concentrated are pre-processed；
S4：The convolutional neural networks are input into using pretreated picture as training data, carry out having supervision
Study, the parameter of each layer of the convolutional neural networks after being trained；
S5：Using the parameter of each layer of the convolutional neural networks after training, described in initialization and step S2
Convolutional neural networks structure identical convolutional neural networks, obtain the image with identity card picture classification feature
Identification network.
The multifarious identity card picture sample set includes that identity card original paper photo, Copy of ID Card shine
The identity card original paper shown on the mobile phone that piece, the identity card picture crossed through software processing, mobile phone camera are obtained shines
The photo of the identity card original paper photo shown on the computer display that the photo of piece, mobile phone camera are obtained.
Convolutional neural networks in the step S2 include five convolutional layers, three down-sampling layers, three it is complete
Articulamentum and softmax graders, wherein,
The size of the first convolutional layer wave filter is 11 × 11 pixels, and step-length is 4 pixels, and characteristic pattern is 96
Individual, the size of the first down-sampling layer wave filter is 3 × 3 pixels, and step-length is 2 pixels；Second convolutional layer
The size of wave filter is 5 × 5 pixels, and characteristic pattern is 256, and the size of the second down-sampling layer wave filter is 3
× 3 pixels, step-length is 2 pixels；The size of the 3rd convolutional layer wave filter is 3 × 3 pixels, and characteristic pattern is
384；The size of Volume Four lamination wave filter is 3 × 3 pixels, and characteristic pattern is 384；5th convolutional layer
The size of wave filter is 3 × 3 pixels, and characteristic pattern is 256, and the size of the 3rd down-sampling layer wave filter is 3
× 3 pixels, step-length is 2 pixels；Full articulamentum and softmax graders use the dropout layers of company of reduction
Number is connect, softmax graders export the target of five types：Identity card original paper photo, Copy of ID Card
The identity card original paper shown on the mobile phone that photo, the identity card picture crossed through software processing, mobile phone camera are obtained
The photo of the identity card original paper photo shown on the computer display that the photo of photo, mobile phone camera are obtained.
In the step S3, carrying out pretreatment to the samples pictures that training sample is concentrated includes：Subtract each
The corresponding average value of pixel, carries out horizontal mirror image processing, carries out cutting treatment.
It is an advantage of the invention that：
1st, using multifarious samples pictures, convolutional neural networks are targetedly trained, are adjusted,
Classification accuracy higher can be reached；
2nd, using the method for Machine self-learning, it is to avoid manual intervention, the convolutional neural networks are enable to learn to arrive
The complete feature of each class identity card picture, the generalization ability and strong adaptability of convolutional neural networks.
Brief description of the drawings
Fig. 1 is the structural representation of convolutional neural networks of the invention.
Fig. 2 is flow chart of the method for the present invention.
Specific embodiment
As shown in Figure 1, 2, the identity card picture sorting technique based on convolutional neural networks disclosed by the invention,
Comprise the following steps：
S1：Sample set is obtained, and sample set is divided into training sample set, checking sample set and test sample
Collection；
Gather substantial amounts of identity card original paper photo, Copy of ID Card photo, through software (e.g., PhotoShop)
The photo of the identity card original paper photo shown on the mobile phone that treated identity card picture, mobile phone camera are obtained,
The photo of the identity card original paper photo shown on the computer display that mobile phone camera is obtained；
Above-mentioned all photos are zoomed to the picture of 256 × 256 pixel sizes as samples pictures, to all
Samples pictures carry out mean value calculation in each pixel, and add label in all samples pictures, for example,
0 is added in identity card original paper picture, the identity card shown on the computer display that mobile phone camera is obtained
1 is added in the picture of original paper photo, the identity card original paper photo shown on the mobile phone that mobile phone camera is obtained
2 are added in picture, 3 are added in Copy of ID Card picture, in the identity card picture that software processing is crossed
Addition 4；
All samples pictures are divided into training sample set (accounting for the 85% of total sample), checking sample set (to account for
Total sample 10%) and test sample collection (accounting for the 5% of total sample).
S2：Set up convolutional neural networks；
As shown in figure 1, the convolutional neural networks that the present invention sets up include five convolutional layers, three down-sampling layers,
Three full articulamentums and softmax graders.Wherein, the size of the first convolutional layer wave filter is 11 × 11
Pixel, step-length is 4 pixels, and characteristic pattern is 96, and the size of the first down-sampling layer wave filter is 3 × 3
Pixel, step-length is 2 pixels；The size of the second convolutional layer wave filter is 5 × 5 pixels, and characteristic pattern is 256
Individual, the size of the second down-sampling layer wave filter is 3 × 3 pixels, and step-length is 2 pixels；3rd convolutional layer
The size of wave filter is 3 × 3 pixels, and characteristic pattern is 384；The size of Volume Four lamination wave filter is 3
× 3 pixels, characteristic pattern is 384；The size of the 5th convolutional layer wave filter is 3 × 3 pixels, and characteristic pattern is
256, the size of the 3rd down-sampling layer wave filter is 3 × 3 pixels, and step-length is 2 pixels；Full articulamentum
And softmax graders reduce connection number using dropout layers, the ratio of dropout is 0.5, softmax
Grader exports the target of five types：Identity card original paper photo, Copy of ID Card photo, through software at
Photo, the hand of the identity card original paper photo shown on the mobile phone that the identity card picture managed, mobile phone camera are obtained
The photo of the identity card original paper photo shown on the computer display that machine camera is obtained.
S3：The samples pictures that training sample is concentrated are pre-processed；
Carrying out pretreatment to samples pictures includes：The corresponding average value of each pixel is subtracted, horizon glass is carried out
As treatment, cut (random cropping) treatment and obtained the picture that resolution ratio is 227 × 227 pixels.
S4：Training sample is concentrated pretreated samples pictures be input into convolutional Neural net as training data
Network, carries out the study for having supervision of tape label, the parameter of each layer of the convolutional neural networks after being trained；
It is each using the parameter of each layer in stochastic gradient descent method adjustment convolutional neural networks in training process
Loss value of the convolutional neural networks on training sample set, and convolutional Neural net are calculated after the completion of secondary iteration
Classification accuracy of the network on checking sample set, the calculation of Loss values is least-mean-square error algorithm, i.e.,
Output according to convolutional neural networks is verifying sample set to the error calculation being input into according to convolutional neural networks
On accuracy rate and loss values, adjust convolutional neural networks learning rate, it is ensured that convolutional neural networks training
Restrained on sample set and reach classification accuracy higher on checking sample set, when convolutional neural networks convergence
Afterwards (threshold value and loss values of rate of accuracy reached to setting no longer decline), each layer in preservation convolutional neural networks
Parameter.
S5：Using the parameter of each layer of the convolutional neural networks after training, the convolution god of same structure is initialized
Through network, the image recognition network with identity card picture classification feature is obtained.
It is follow-up test sample collection to be tested using the image recognition network, is classified.
In a specific embodiment, the training sample set size for using is 17000 identity card pictures, checking
The size of sample set is 2000 identity card pictures, and the size of test sample collection is 1000 identity card pictures,
Before test, first the samples pictures that test sample is concentrated are pre-processed, then input picture identification network,
The final classification accuracy of the image recognition network in test sample collection has reached 92%, due in sample
Identity card picture has diversity (including identity card original paper photo, Copy of ID Card photo, software processing
Photo, the mobile phone of the identity card original paper photo shown on the mobile phone that the identity card picture crossed, mobile phone camera are obtained
The photo of the identity card original paper photo shown on the computer display that camera is obtained), thus identity card picture
Classification accuracy it is higher.
The above is presently preferred embodiments of the present invention and its know-why used, for the skill of this area
It is without departing from the spirit and scope of the present invention, any based on the technology of the present invention side for art personnel
Equivalent transformation on the basis of case, it is simple replace etc. it is obvious change, belong to the scope of the present invention it
It is interior.
Claims (4)
1. the identity card picture sorting technique of convolutional neural networks is based on, it is characterised in that including：
S1：Multifarious identity card picture sample set is obtained, and is divided into training sample set, checking sample set
And test sample collection；
S2：Set up convolutional neural networks；
S3：The samples pictures that training sample is concentrated are pre-processed；
S4：The convolutional neural networks are input into using pretreated picture as training data, carry out having supervision
Study, the parameter of each layer of the convolutional neural networks after being trained；
S5：Using the parameter of each layer of the convolutional neural networks after training, described in initialization and step S2
Convolutional neural networks structure identical convolutional neural networks, obtain the image with identity card picture classification feature
Identification network.
2. the identity card picture sorting technique based on convolutional neural networks according to claim 1, its
It is characterised by, the multifarious identity card picture sample set includes that identity card original paper photo, identity card are duplicated
The identity card shown on the mobile phone that part photo, the identity card picture crossed through software processing, mobile phone camera are obtained is former
The photograph of the identity card original paper photo shown on the computer display that the photo of part photo, mobile phone camera are obtained
Piece.
3. the identity card picture sorting technique based on convolutional neural networks according to claim 2, its
Be characterised by, the convolutional neural networks in the step S2 include five convolutional layers, three down-sampling layers,
Three full articulamentums and softmax graders, wherein,
The size of the first convolutional layer wave filter is 11 × 11 pixels, and step-length is 4 pixels, and characteristic pattern is 96
Individual, the size of the first down-sampling layer wave filter is 3 × 3 pixels, and step-length is 2 pixels；Second convolutional layer
The size of wave filter is 5 × 5 pixels, and characteristic pattern is 256, and the size of the second down-sampling layer wave filter is 3
× 3 pixels, step-length is 2 pixels；The size of the 3rd convolutional layer wave filter is 3 × 3 pixels, and characteristic pattern is
384；The size of Volume Four lamination wave filter is 3 × 3 pixels, and characteristic pattern is 384；5th convolutional layer
The size of wave filter is 3 × 3 pixels, and characteristic pattern is 256, and the size of the 3rd down-sampling layer wave filter is 3
× 3 pixels, step-length is 2 pixels；Full articulamentum and softmax graders use the dropout layers of company of reduction
Number is connect, softmax graders export the target of five types：Identity card original paper photo, Copy of ID Card
The identity card original paper shown on the mobile phone that photo, the identity card picture crossed through software processing, mobile phone camera are obtained
The photo of the identity card original paper photo shown on the computer display that the photo of photo, mobile phone camera are obtained.
4. the identity card picture sorting technique based on convolutional neural networks according to claim 2, its
It is characterised by, in the step S3, carrying out pretreatment to the samples pictures that training sample is concentrated includes：Subtract
The corresponding average value of each pixel is removed, horizontal mirror image processing is carried out, cutting treatment is carried out.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201511009006.6A CN106934408A (en) | 2015-12-29 | 2015-12-29 | Identity card picture sorting technique based on convolutional neural networks |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201511009006.6A CN106934408A (en) | 2015-12-29 | 2015-12-29 | Identity card picture sorting technique based on convolutional neural networks |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106934408A true CN106934408A (en) | 2017-07-07 |
Family
ID=59459227
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201511009006.6A Pending CN106934408A (en) | 2015-12-29 | 2015-12-29 | Identity card picture sorting technique based on convolutional neural networks |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106934408A (en) |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107292353A (en) * | 2017-08-09 | 2017-10-24 | 广东工业大学 | A kind of fruit tree classification method and system |
| CN107292749A (en) * | 2017-08-04 | 2017-10-24 | 平安科技（深圳）有限公司 | Car damages sorting technique, system and the readable storage medium storing program for executing of certificate photograph |
| CN107491766A (en) * | 2017-08-31 | 2017-12-19 | 四川长虹电器股份有限公司 | Photo classification method based on image recognition |
| CN108229298A (en) * | 2017-09-30 | 2018-06-29 | 北京市商汤科技开发有限公司 | The training of neural network and face identification method and device, equipment, storage medium |
| CN109165674A (en) * | 2018-07-19 | 2019-01-08 | 南京富士通南大软件技术有限公司 | A kind of certificate photo classification method based on multi-tag depth convolutional network |
| CN109409421A (en) * | 2018-10-09 | 2019-03-01 | 杭州诚道科技股份有限公司 | Motor vehicle, driver's archival image recognition methods based on convolutional neural networks |
| CN109696698A (en) * | 2019-03-05 | 2019-04-30 | 湖南国科微电子股份有限公司 | Navigator fix prediction technique, device, electronic equipment and storage medium |
| CN110222736A (en) * | 2019-05-20 | 2019-09-10 | 北京字节跳动网络技术有限公司 | Method, apparatus, electronic equipment and the computer readable storage medium of training classifier |
| CN110262653A (en) * | 2018-03-12 | 2019-09-20 | 东南大学 | A kind of millimeter wave sensor gesture identification method based on convolutional neural networks |
| CN110432892A (en) * | 2019-08-05 | 2019-11-12 | 苏州米特希赛尔人工智能有限公司 | Machine learning ECG Automatic Diagnosis System |
| CN110472664A (en) * | 2019-07-17 | 2019-11-19 | 杭州有盾网络科技有限公司 | A kind of certificate image identification method, device and equipment based on deep learning |
| CN111062338A (en) * | 2019-12-19 | 2020-04-24 | 厦门商集网络科技有限责任公司 | Certificate portrait consistency comparison method and system |
| CN111324874A (en) * | 2020-01-21 | 2020-06-23 | 支付宝实验室(新加坡)有限公司 | Certificate authenticity identification method and device |
| CN112686237A (en) * | 2020-12-21 | 2021-04-20 | 福建新大陆软件工程有限公司 | Certificate OCR recognition method |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2176406A1 (en) * | 1995-05-25 | 1996-11-26 | John Stewart Denker | Capture and use of a personal identification feature at a service terminal |
| CN103488998A (en) * | 2013-09-11 | 2014-01-01 | 东华大学 | Identity card recognition method based on neural network and image processing technology |
| CN105117739A (en) * | 2015-07-29 | 2015-12-02 | 南京信息工程大学 | Clothes classifying method based on convolutional neural network |
| CN105118048A (en) * | 2015-07-17 | 2015-12-02 | 北京旷视科技有限公司 | Method and device for identifying copying certificate image |
-
2015
- 2015-12-29 CN CN201511009006.6A patent/CN106934408A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2176406A1 (en) * | 1995-05-25 | 1996-11-26 | John Stewart Denker | Capture and use of a personal identification feature at a service terminal |
| CN103488998A (en) * | 2013-09-11 | 2014-01-01 | 东华大学 | Identity card recognition method based on neural network and image processing technology |
| CN105118048A (en) * | 2015-07-17 | 2015-12-02 | 北京旷视科技有限公司 | Method and device for identifying copying certificate image |
| CN105117739A (en) * | 2015-07-29 | 2015-12-02 | 南京信息工程大学 | Clothes classifying method based on convolutional neural network |
Non-Patent Citations (7)
| Title |
|---|
| K UEKI ET AL: "Multi-layer feature extractions for image classification — Knowledge from deep CNNs", 《 2015 INTERNATIONAL CONFERENCE ON SYSTEMS, SIGNALS AND IMAGE PROCESSING (IWSSIP)》 * |
| ZHE DONG ET AL: "Multi-level photo quality assessment with multi-view features", 《NEUROCOMPUTING》 * |
| 余平德: "《毛纺生产技术275问》", 31 October 2007 * |
| 李卫: "深度学习在图像识别中的研究及应用", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 田志梅: "《CORELDRAW X5从入门到精通中文版》", 31 May 2015 * |
| 赵伟达等: "人脸检测中基于自适应ICA的特征提取算法", 《计算机仿真》 * |
| 郑永森: "基于卷积神经网络的身份证号码识别研究与实现", 《计算机光盘软件与应用》 * |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107292749A (en) * | 2017-08-04 | 2017-10-24 | 平安科技（深圳）有限公司 | Car damages sorting technique, system and the readable storage medium storing program for executing of certificate photograph |
| WO2019024234A1 (en) * | 2017-08-04 | 2019-02-07 | 平安科技（深圳）有限公司 | Vehicle loss-related identification photo classification method and system, electronic device, and readable storage medium |
| CN107292353A (en) * | 2017-08-09 | 2017-10-24 | 广东工业大学 | A kind of fruit tree classification method and system |
| CN107491766A (en) * | 2017-08-31 | 2017-12-19 | 四川长虹电器股份有限公司 | Photo classification method based on image recognition |
| CN108229298A (en) * | 2017-09-30 | 2018-06-29 | 北京市商汤科技开发有限公司 | The training of neural network and face identification method and device, equipment, storage medium |
| CN110262653A (en) * | 2018-03-12 | 2019-09-20 | 东南大学 | A kind of millimeter wave sensor gesture identification method based on convolutional neural networks |
| CN109165674A (en) * | 2018-07-19 | 2019-01-08 | 南京富士通南大软件技术有限公司 | A kind of certificate photo classification method based on multi-tag depth convolutional network |
| CN109409421B (en) * | 2018-10-09 | 2021-12-07 | 杭州诚道科技股份有限公司 | Motor vehicle and driver archive image identification method based on convolutional neural network |
| CN109409421A (en) * | 2018-10-09 | 2019-03-01 | 杭州诚道科技股份有限公司 | Motor vehicle, driver's archival image recognition methods based on convolutional neural networks |
| CN109696698A (en) * | 2019-03-05 | 2019-04-30 | 湖南国科微电子股份有限公司 | Navigator fix prediction technique, device, electronic equipment and storage medium |
| CN110222736A (en) * | 2019-05-20 | 2019-09-10 | 北京字节跳动网络技术有限公司 | Method, apparatus, electronic equipment and the computer readable storage medium of training classifier |
| CN110472664A (en) * | 2019-07-17 | 2019-11-19 | 杭州有盾网络科技有限公司 | A kind of certificate image identification method, device and equipment based on deep learning |
| CN110432892A (en) * | 2019-08-05 | 2019-11-12 | 苏州米特希赛尔人工智能有限公司 | Machine learning ECG Automatic Diagnosis System |
| CN111062338A (en) * | 2019-12-19 | 2020-04-24 | 厦门商集网络科技有限责任公司 | Certificate portrait consistency comparison method and system |
| CN111062338B (en) * | 2019-12-19 | 2023-11-17 | 厦门商集网络科技有限责任公司 | License and portrait consistency comparison method and system |
| CN111324874A (en) * | 2020-01-21 | 2020-06-23 | 支付宝实验室(新加坡)有限公司 | Certificate authenticity identification method and device |
| CN112686237A (en) * | 2020-12-21 | 2021-04-20 | 福建新大陆软件工程有限公司 | Certificate OCR recognition method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106934408A (en) | Identity card picture sorting technique based on convolutional neural networks | |
| Bayar et al. | Design principles of convolutional neural networks for multimedia forensics | |
| CN107563999A (en) | A kind of chip defect recognition methods based on convolutional neural networks | |
| CN107316295A (en) | A kind of fabric defects detection method based on deep neural network | |
| CN105303193B (en) | A kind of passenger number statistical system based on single-frame images processing | |
| CN111339839B (en) | Intensive target detection metering method | |
| WO2019031503A1 (en) | Tire image recognition method and tire image recognition device | |
| CN107657249A (en) | Method, apparatus, storage medium and the processor that Analysis On Multi-scale Features pedestrian identifies again | |
| CN109657716A (en) | A kind of vehicle appearance damnification recognition method based on deep learning | |
| CN106650740B (en) | A kind of licence plate recognition method and terminal | |
| CN109859171A (en) | A kind of flooring defect automatic testing method based on computer vision and deep learning | |
| CN107784654A (en) | Image partition method, device and full convolutional network system | |
| CN107729854A (en) | A kind of gesture identification method of robot, system and robot | |
| CN106372648A (en) | Multi-feature-fusion-convolutional-neural-network-based plankton image classification method | |
| CN104778702A (en) | Image stego-detection method on basis of deep learning | |
| CN106548182A (en) | Based on deep learning and the causal analytic pavement distress survey method and device of master | |
| CN111046774A (en) | Chinese signature handwriting identification method based on convolutional neural network | |
| CN108960261A (en) | A kind of obvious object detection method based on attention mechanism | |
| CN106934319A (en) | People's car objective classification method in monitor video based on convolutional neural networks | |
| CN105426917A (en) | Element classification method and device | |
| CN110009622B (en) | Display panel appearance defect detection network and defect detection method thereof | |
| CN111209858A (en) | Real-time license plate detection method based on deep convolutional neural network | |
| CN110490214A (en) | The recognition methods and system of image, storage medium and processor | |
| CN107067015A (en) | A kind of vehicle checking method and device based on multiple features deep learning | |
| CN110349167A (en) | A kind of image instance dividing method and device |
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: 20170707 |