CN107451994A - Object detecting method and device based on generation confrontation network - Google Patents
Object detecting method and device based on generation confrontation network Download PDFInfo
- Publication number
- CN107451994A CN107451994A CN201710614277.7A CN201710614277A CN107451994A CN 107451994 A CN107451994 A CN 107451994A CN 201710614277 A CN201710614277 A CN 201710614277A CN 107451994 A CN107451994 A CN 107451994A
- Authority
- CN
- China
- Prior art keywords
- image
- maker
- confrontation network
- discriminator
- generation confrontation
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
Landscapes
- Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of object detecting method and device based on generation confrontation network, generation confrontation network includes maker and discriminator, and methods described comprises the following steps：The original image of input is subjected to size change over and obtains the first image, and denoising is filtered to described first image and obtains the second image；Second image is input in generation confrontation network；The maker is trained to second image, and the Object representation text of corresponding second image of generation simultaneously sends it to the discriminator；The discriminator differentiates whether the Object representation text is real data, and identification result is sent into the maker, and training is adjusted to the maker.Implement the object detecting method and device based on generation confrontation network of the present invention, have the advantages that：Object detection efficiency and discrimination are higher.
Description
Technical field
The present invention relates to target object detection field, more particularly to a kind of object detecting method based on generation confrontation network
And device.
Background technology
In the prior art, when picture is identified, existing method has two kinds, and one kind is to use to be based on picture segmentation
Method, using picture segmentation, feature extraction is identified；Another kind is the method based on artificial neural network CNN, passes through people
Artificial neural networks CNN to carry out convolutional calculation detection object to picture.First method is that first picture is split and then carried
The feature of setting is taken, this method efficiency comparison is low, and discrimination is not high, while needs to preset the characteristic species of extraction
Class, thus it is very poorly efficient.Second method is that convolution detection object is carried out to picture, is computed repeatedly during CNN, and need
Candidate frame is set, and the feature of candidate frame is extracted, efficiency comparison is poorly efficient.
The content of the invention
The technical problem to be solved in the present invention is, for the drawbacks described above of prior art, there is provided a kind of object detection effect
Rate and discrimination higher object detecting method and device based on generation confrontation network.
The technical solution adopted for the present invention to solve the technical problems is：Construct a kind of object based on generation confrontation network
Detection method, the generation confrontation network include maker and discriminator, and methods described comprises the following steps：
A the original image of input) is subjected to size change over and obtains the first image, and described first image is filtered
Processing of making an uproar obtains the second image；
B) second image is input in generation confrontation network；
C) maker is trained to second image, the Object representation text of corresponding second image of generation
And send it to the discriminator；
D) discriminator differentiates whether the Object representation text is real data, and identification result is sent into institute
Maker is stated, training is adjusted to the maker.
Of the present invention based in the object detecting method of generation confrontation network, the internal structure of the maker is
Convolutional neural networks, the structure of the discriminator is BLSTM structural models.
In the object detecting method of the present invention that network is resisted based on generation, described first image 256*256
Image.
The invention further relates to a kind of device for realizing the above-mentioned object detecting method based on generation confrontation network, the generation
Confrontation network includes maker and discriminator, and described device includes：
Image transforming unit：The first image is obtained for the original image of input to be carried out into size change over, and to described the
One image is filtered denoising and obtains the second image；
Image input units：For second image to be input in generation confrontation network；
Training unit：For making the maker be trained second image, corresponding second image of generation
Object representation text and send it to the discriminator；
Text judging unit：For making the discriminator differentiate whether the Object representation text is real data, and
Identification result is sent to the maker, training is adjusted to the maker.
In device of the present invention, the internal structure of the maker is convolutional neural networks, the discriminator
Structure is BLSTM structural models.
In device of the present invention, described first image is 256*256 image.
Implement the object detecting method and device based on generation confrontation network of the present invention, have the advantages that：By
The first image is obtained in the original image of input is carried out into size change over, and denoising is filtered to the first image and obtains the
Two images；Second image is input in generation confrontation network；Maker is trained to the second image, generates corresponding second figure
The Object representation text of picture simultaneously sends it to discriminator；Discriminator differentiates whether description text is real data, and will mirror
Other result is sent to maker, and training is adjusted to maker, the object detecting method of network is resisted by generating, it is not only
Accuracy rate can be improved, and detection speed is also very fast, and object detection efficiency and discrimination are higher.
Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
There is the required accompanying drawing used in technology description to be briefly described, it should be apparent that, drawings in the following description are only this
Some embodiments of invention, for those of ordinary skill in the art, on the premise of not paying creative work, can be with
Other accompanying drawings are obtained according to these accompanying drawings.
Fig. 1 is the flow of method in object detecting method and device one embodiment of the present invention based on generation confrontation network
Figure；
Fig. 2 is the structural representation of generation confrontation network in the embodiment；
Fig. 3 is the internal structure schematic diagram of maker in the embodiment；
Fig. 4 is the structural representation of discriminator in the embodiment；
Fig. 5 is the structural representation of device in the embodiment.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Site preparation describes, it is clear that described embodiment is only part of the embodiment of the present invention, rather than whole embodiments.It is based on
Embodiment in the present invention, those of ordinary skill in the art are obtained every other under the premise of creative work is not made
Embodiment, belong to the scope of protection of the invention.
In object detecting method of the present invention based on generation confrontation network and device embodiment, it is based on generation confrontation net
The flow chart of the object detecting method of network is as shown in Figure 1.In the present embodiment, generation confrontation network includes maker and discriminator.
In Fig. 1, it should be comprised the following steps based on the object detecting method of generation confrontation network：
The original image of input is carried out size change over and obtains the first image by step S01, and the first image is filtered
Denoising obtains the second image：In this step, the original image of input is subjected to size change over and obtains the first image, it is then right
First image obtains the second image after being filtered denoising.It is noted that the first image is 256*256 figure
Picture.That is, after the original image to input carries out size change over, the image that size is 256*256, the image can be obtained
It is exactly the first image.Certainly, under the certain situation of the present embodiment, the size of the first image can carry out phase as the case may be
It should adjust.
Second image is input in generation confrontation network by step S02：It is defeated using the second image as input in this step
Enter to generation and resist in network.
Step S03 makers are trained to the second image, generate corresponding second image Object representation text and by its
It is sent to discriminator：In this step, maker is trained to the second image, generate to should the second image Object representation text
This, and the Object representation text of the image of correspondence second is sent to discriminator.
Step S04 discriminators differentiate whether Object representation text is real data, and identification result is sent into generation
Device, training is adjusted to maker：In this step, discriminator differentiates to above-mentioned Object representation text, judges above-mentioned thing
Body describes whether text is real data, and identification result is sent into maker, and training is adjusted to maker.Finally
Purpose be maker export during to discriminator be difficult judge be it is real or forge data, that is, maximize discriminator, most
Small metaplasia is grown up to be a useful person, and its object function is：
In above formula, x~Pdata (x) represents that x comes from Pdata (x) distribution；Z~Pz (z) represents points of the z from Pz (z)
Cloth；EX~Pdata (x)Represent x expectation；EZ~Pz (z)Represent z expectation；D (x | y) represents that the input of discriminator is：In given y feelings
Under condition, x input；G (z | y) represents that the input of maker is：In the case of given y, z input；D (G (z | y)) represent mirror
In the case that the input of other device is maker G, and current producer G input is given y, z input.That is G (z |
Y) input of the output as discriminator D.
The object detecting method based on generation confrontation network of the present invention, network is resisted by generating to the picture of input
After the Maker model for training generation, you can the object included by output picture.
Fig. 2 is that the structural representation for resisting network is generated in the present embodiment, and in Fig. 2, G is maker, and x represents true number
The label information for Object representation is represented according to, y, and z represents noise, z Gaussian distributeds or is uniformly distributed.Pass through generation pair
The object detecting method of anti-network, it can not only improve accuracy rate, and detection speed is also very fast, object detection efficiency and knowledge
Not rate is higher.
Fig. 3 is the internal structure schematic diagram of maker in the present embodiment, and in the present embodiment, maker G internal structure is
Convolutional neural networks, wherein, Maxpooling represents the pond layer for taking maximum.Fig. 4 is the structure of discriminator in the present embodiment
Schematic diagram, in the present embodiment, discriminator D structure is BLSTM structural models, wherein object tags vector for one-hone to
Amount, average represent for input take one-hot vectors be output to after average.
The present embodiment further relates to a kind of device for realizing the above-mentioned object detecting method based on generation confrontation network, its structure
Schematic diagram is as shown in Figure 5.In the present embodiment, generation confrontation network includes maker and discriminator.In Fig. 5, the device includes
Image transforming unit 1, image input units 2, training unit 3 and text judging unit 4；Wherein, image transforming unit 1 is used to incite somebody to action
The original image of input carries out size change over and obtains the first image, and is filtered denoising to the first image and obtains the second figure
Picture；It is noted that the first image is 256*256 image.That is, when the original image to input carries out size change
After changing, the image that size is 256*256 can be obtained, the image is exactly the first image.Certainly, in the certain situation of the present embodiment
Under, the size of the first image can adjust accordingly as the case may be.
Image input units 2 are used to the second image being input in generation confrontation network；Training unit 3 is used to make maker
Second image is trained, the Object representation text of corresponding second image is generated and sends it to discriminator；Text judges
Unit 4 is used to make discriminator differentiate whether Object representation text is real data, and identification result is sent into maker, right
Maker is adjusted training.By generating the object detecting method of confrontation network, it can not only improve accuracy rate, Er Qiejian
Degree of testing the speed is also very fast, and object detection efficiency and discrimination are higher.
In the device of the present embodiment, the internal structure of maker is convolutional neural networks, and the structure of discriminator is BLSTM
Structural model.
In a word, in the present embodiment, a picture is given, by being entered into generation confrontation network, you can output inspection
Target object present in mapping piece, it can not only improve accuracy rate, and detection speed is also very fast, object detection efficiency and
Discrimination is higher.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all essences in the present invention
God any modification, equivalent substitution and improvements made etc., should be included in the scope of the protection with principle.
Claims (6)
1. a kind of object detecting method based on generation confrontation network, it is characterised in that the generation confrontation network includes generation
Device and discriminator, methods described comprise the following steps：
A the original image of input) is subjected to size change over and obtains the first image, and described first image is filtered at denoising
Reason obtains the second image；
B) second image is input in generation confrontation network；
C) maker is trained to second image, and the Object representation text of corresponding second image of generation simultaneously will
It is sent to the discriminator；
D) discriminator differentiates whether the Object representation text is real data, and identification result is sent into the life
Grow up to be a useful person, training is adjusted to the maker.
2. the object detecting method according to claim 1 based on generation confrontation network, it is characterised in that the maker
Internal structure be convolutional neural networks, the structure of the discriminator is BLSTM structural models.
3. the object detecting method according to claim 1 or 2 based on generation confrontation network, it is characterised in that described the
One image is 256*256 image.
4. a kind of device for realizing the object detecting method as claimed in claim 1 based on generation confrontation network, its feature exist
In the generation confrontation network includes maker and discriminator, and described device includes：
Image transforming unit：The first image is obtained for the original image of input to be carried out into size change over, and to first figure
The second image is obtained as being filtered denoising；
Image input units：For second image to be input in generation confrontation network；
Training unit：For making the maker be trained second image, the thing of corresponding second image of generation
Body describes text and sends it to the discriminator；
Text judging unit：For making the discriminator differentiate whether the Object representation text is real data, and will mirror
Other result is sent to the maker, and training is adjusted to the maker.
5. object detecting method of the realization according to claim 4 as claimed in claim 1 based on generation confrontation network
Device, it is characterised in that the internal structure of the maker is convolutional neural networks, and the structure of the discriminator is BLSTM
Structural model.
6. the object detection side realized as claimed in claim 1 based on generation confrontation network according to claim 4 or 5
The device of method, it is characterised in that described first image is 256*256 image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710614277.7A CN107451994A (en) | 2017-07-25 | 2017-07-25 | Object detecting method and device based on generation confrontation network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710614277.7A CN107451994A (en) | 2017-07-25 | 2017-07-25 | Object detecting method and device based on generation confrontation network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107451994A true CN107451994A (en) | 2017-12-08 |
Family
ID=60487550
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710614277.7A Pending CN107451994A (en) | 2017-07-25 | 2017-07-25 | Object detecting method and device based on generation confrontation network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107451994A (en) |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108334848A (en) * | 2018-02-06 | 2018-07-27 | 哈尔滨工业大学 | A kind of small face identification method based on generation confrontation network |
| CN108509952A (en) * | 2018-04-10 | 2018-09-07 | 深圳市唯特视科技有限公司 | A kind of instance-level image interpretation technology paying attention to generating confrontation network based on depth |
| CN108710892A (en) * | 2018-04-04 | 2018-10-26 | 浙江工业大学 | Synergetic immunity defence method towards a variety of confrontation picture attacks |
| CN109242841A (en) * | 2018-08-30 | 2019-01-18 | 广东工业大学 | A kind of transmission tower defect inspection method based on generation confrontation network |
| CN109685097A (en) * | 2018-11-08 | 2019-04-26 | 银河水滴科技（北京）有限公司 | A kind of image detecting method and device based on GAN |
| CN110147830A (en) * | 2019-05-07 | 2019-08-20 | 东软集团股份有限公司 | Training image data generates method, image data classification method and the device of network |
| WO2020019738A1 (en) * | 2018-07-24 | 2020-01-30 | 深圳先进技术研究院 | Plaque processing method and device capable of performing magnetic resonance vessel wall imaging, and computing device |
| WO2020108165A1 (en) * | 2018-11-30 | 2020-06-04 | 腾讯科技（深圳）有限公司 | Image description information generation method and device, and electronic device |
| CN111242112A (en) * | 2018-11-29 | 2020-06-05 | 马上消费金融股份有限公司 | Image processing method, identity information processing method and device |
| US10810460B2 (en) | 2018-06-13 | 2020-10-20 | Cosmo Artificial Intelligence—AI Limited | Systems and methods for training generative adversarial networks and use of trained generative adversarial networks |
| US10909421B2 (en) | 2018-12-13 | 2021-02-02 | Industrial Technology Research Institute | Training method for phase image generator and training method of phase image classifier |
| US11100633B2 (en) | 2018-06-13 | 2021-08-24 | Cosmo Artificial Intelligence—Al Limited | Systems and methods for processing real-time video from a medical image device and detecting objects in the video |
| US11250329B2 (en) | 2017-10-26 | 2022-02-15 | Nvidia Corporation | Progressive modification of generative adversarial neural networks |
| US11263525B2 (en) | 2017-10-26 | 2022-03-01 | Nvidia Corporation | Progressive modification of neural networks |
| US11315231B2 (en) | 2018-06-08 | 2022-04-26 | Industrial Technology Research Institute | Industrial image inspection method and system and computer readable recording medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015195609A1 (en) * | 2014-06-16 | 2015-12-23 | Siemens Healthcare Diagnostics Inc. | Analyzing digital holographic microscopy data for hematology applications |
| CN106530220A (en) * | 2016-11-09 | 2017-03-22 | 深圳市唯特视科技有限公司 | Method for intelligent visual operation on product pictures according to client preferences |
| CN106803082A (en) * | 2017-01-23 | 2017-06-06 | 重庆邮电大学 | A kind of online handwriting recognition methods based on conditional generation confrontation network |
| CN106845471A (en) * | 2017-02-20 | 2017-06-13 | 深圳市唯特视科技有限公司 | A kind of vision significance Forecasting Methodology based on generation confrontation network |
-
2017
- 2017-07-25 CN CN201710614277.7A patent/CN107451994A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015195609A1 (en) * | 2014-06-16 | 2015-12-23 | Siemens Healthcare Diagnostics Inc. | Analyzing digital holographic microscopy data for hematology applications |
| CN106530220A (en) * | 2016-11-09 | 2017-03-22 | 深圳市唯特视科技有限公司 | Method for intelligent visual operation on product pictures according to client preferences |
| CN106803082A (en) * | 2017-01-23 | 2017-06-06 | 重庆邮电大学 | A kind of online handwriting recognition methods based on conditional generation confrontation network |
| CN106845471A (en) * | 2017-02-20 | 2017-06-13 | 深圳市唯特视科技有限公司 | A kind of vision significance Forecasting Methodology based on generation confrontation network |
Non-Patent Citations (2)
| Title |
|---|
| MEHDI MIRZA: "Conditional Generative Adversarial Nets", 《ARXIV:1411.1784V1 [CS.LG]》 * |
| 王坤峰 等: "生成式对抗网络GAN的研究进展与展望", 《自动化学报》 * |
Cited By (22)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11763168B2 (en) | 2017-10-26 | 2023-09-19 | Nvidia Corporation | Progressive modification of generative adversarial neural networks |
| US11263525B2 (en) | 2017-10-26 | 2022-03-01 | Nvidia Corporation | Progressive modification of neural networks |
| US11250329B2 (en) | 2017-10-26 | 2022-02-15 | Nvidia Corporation | Progressive modification of generative adversarial neural networks |
| CN108334848A (en) * | 2018-02-06 | 2018-07-27 | 哈尔滨工业大学 | A kind of small face identification method based on generation confrontation network |
| CN108710892B (en) * | 2018-04-04 | 2020-09-01 | 浙江工业大学 | Cooperative immune defense method for multiple anti-picture attacks |
| CN108710892A (en) * | 2018-04-04 | 2018-10-26 | 浙江工业大学 | Synergetic immunity defence method towards a variety of confrontation picture attacks |
| CN108509952A (en) * | 2018-04-10 | 2018-09-07 | 深圳市唯特视科技有限公司 | A kind of instance-level image interpretation technology paying attention to generating confrontation network based on depth |
| US11315231B2 (en) | 2018-06-08 | 2022-04-26 | Industrial Technology Research Institute | Industrial image inspection method and system and computer readable recording medium |
| US10810460B2 (en) | 2018-06-13 | 2020-10-20 | Cosmo Artificial Intelligence—AI Limited | Systems and methods for training generative adversarial networks and use of trained generative adversarial networks |
| US11574403B2 (en) | 2018-06-13 | 2023-02-07 | Cosmo Artificial Intelligence—AI Limited | Systems and methods for processing real-time video from a medical image device and detecting objects in the video |
| US12089820B2 (en) | 2018-06-13 | 2024-09-17 | Cosmo Artificial Intelligence—AI Limited | Systems and methods for processing real-time video from a medical image device and detecting objects in the video |
| US11844499B2 (en) | 2018-06-13 | 2023-12-19 | Cosmo Artificial Intelligence—AI Limited | Systems and methods for processing real-time video from a medical image device and detecting objects in the video |
| US11100633B2 (en) | 2018-06-13 | 2021-08-24 | Cosmo Artificial Intelligence—Al Limited | Systems and methods for processing real-time video from a medical image device and detecting objects in the video |
| US11195052B2 (en) | 2018-06-13 | 2021-12-07 | Cosmo Artificial Intelligence—AI Limited | Systems and methods for training generative adversarial networks and use of trained generative adversarial networks |
| WO2020019738A1 (en) * | 2018-07-24 | 2020-01-30 | 深圳先进技术研究院 | Plaque processing method and device capable of performing magnetic resonance vessel wall imaging, and computing device |
| CN109242841A (en) * | 2018-08-30 | 2019-01-18 | 广东工业大学 | A kind of transmission tower defect inspection method based on generation confrontation network |
| CN109685097A (en) * | 2018-11-08 | 2019-04-26 | 银河水滴科技（北京）有限公司 | A kind of image detecting method and device based on GAN |
| CN111242112A (en) * | 2018-11-29 | 2020-06-05 | 马上消费金融股份有限公司 | Image processing method, identity information processing method and device |
| US11783199B2 (en) | 2018-11-30 | 2023-10-10 | Tencent Technology (Shenzhen) Company Limited | Image description information generation method and apparatus, and electronic device |
| WO2020108165A1 (en) * | 2018-11-30 | 2020-06-04 | 腾讯科技（深圳）有限公司 | Image description information generation method and device, and electronic device |
| US10909421B2 (en) | 2018-12-13 | 2021-02-02 | Industrial Technology Research Institute | Training method for phase image generator and training method of phase image classifier |
| CN110147830A (en) * | 2019-05-07 | 2019-08-20 | 东软集团股份有限公司 | Training image data generates method, image data classification method and the device of network |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107451994A (en) | Object detecting method and device based on generation confrontation network | |
| CN109543606B (en) | Human face recognition method with attention mechanism | |
| CN109815826B (en) | Method and device for generating face attribute model | |
| WO2020098158A1 (en) | Pedestrian re-recognition method and apparatus, and computer readable storage medium | |
| CN107273872A (en) | The depth discrimination net model methodology recognized again for pedestrian in image or video | |
| CN108537743A (en) | A kind of face-image Enhancement Method based on generation confrontation network | |
| CN109614921B (en) | Cell segmentation method based on semi-supervised learning of confrontation generation network | |
| CN109190446A (en) | Pedestrian's recognition methods again based on triple focused lost function | |
| CN109101865A (en) | A kind of recognition methods again of the pedestrian based on deep learning | |
| CN110738161A (en) | face image correction method based on improved generation type confrontation network | |
| CN107194418B (en) | Rice aphid detection method based on antagonistic characteristic learning | |
| CN108629753A (en) | A kind of face image restoration method and device based on Recognition with Recurrent Neural Network | |
| CN107464217B (en) | Image processing method and device | |
| CN104091169A (en) | Behavior identification method based on multi feature fusion | |
| CN106372656B (en) | Obtain method, image-recognizing method and the device of the disposable learning model of depth | |
| CN107066963B (en) | A kind of adaptive people counting method | |
| CN109325513B (en) | Image classification network training method based on massive single-class images | |
| CN108564166A (en) | Based on the semi-supervised feature learning method of the convolutional neural networks with symmetrical parallel link | |
| CN106886763A (en) | The system and method for real-time detection face | |
| CN106874825A (en) | The training method of Face datection, detection method and device | |
| CN109359527A (en) | Hair zones extracting method and system neural network based | |
| CN111597978B (en) | Method for automatically generating pedestrian re-identification picture based on StarGAN network model | |
| CN109344709A (en) | A kind of face generates the detection method of forgery image | |
| CN109886170A (en) | A kind of identification of oncomelania intelligent measurement and statistical system | |
| CN111160356A (en) | Image segmentation and classification 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 |
Application publication date: 20171208 |
|
| RJ01 | Rejection of invention patent application after publication |