CN106373121A - Fuzzy image identification method and apparatus - Google Patents
Fuzzy image identification method and apparatus Download PDFInfo
- Publication number
- CN106373121A CN106373121A CN201610828484.8A CN201610828484A CN106373121A CN 106373121 A CN106373121 A CN 106373121A CN 201610828484 A CN201610828484 A CN 201610828484A CN 106373121 A CN106373121 A CN 106373121A
- Authority
- CN
- China
- Prior art keywords
- image
- difference
- broad
- fuzzy
- gray
- 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
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30168—Image quality inspection
Abstract
The invention discloses a fuzzy image identification method and apparatus. The fuzzy image identification method comprises the following steps: obtaining an image; calculating differences between adjacent pixel points in the image; and if the proportion of a difference smaller than a preset difference threshold to a total difference is greater than a preset proportion threshold, determining that the image is a fuzzy image. According to the technical scheme provided by the invention, the fuzzy image can be automatically identified.
Description
Technical field
The present invention relates to picture recognition technical field is and in particular to a kind of fuzzy image recognition method and apparatus.
Background technology
Searching topic class application program (i.e. searching class app, wherein, app is the english abbreviation of application) is to refer to obtain
After taking exercise question and automatically searching for solution approach and the answer of this exercise question from exam pool, show the solution approach of this exercise question and answer
app.
Topic class app of searching on the market mainly obtains, by photographic head, the picture comprising examination question, afterwards from this picture now
Middle extraction text message carries out examination question search.If because when shooting, photographic head shake or other reason lead to the picture mould getting
During paste, then will be difficult to extract accurate text message from fuzzy picture, and this also will affect examination question Search Results
Accuracy.
Not can recognize that the related mechanism of broad image at present.
Content of the invention
The present invention provides a kind of fuzzy image recognition method and apparatus, for realizing the automatic identification to broad image.
First aspect present invention provides a kind of fuzzy image recognition method, comprising:
Obtain image；
Calculate the difference between each neighbor pixel in above-mentioned image；
If less than default difference threshold difference account for whole differences ratio be more than default proportion threshold value it is determined that on
Stating image is broad image.
Second aspect present invention provides a kind of fuzzy image recognition device, comprising:
Acquiring unit, for obtaining image；
Computing unit, for calculating the difference between each neighbor pixel in the image that above-mentioned acquiring unit obtains；
Determining unit, the ratio for accounting for whole differences when the difference less than default difference threshold is more than default ratio
During threshold value, determine that above-mentioned image is broad image.
Therefore, the present invention program passes through to obtain image and calculate the difference between each neighbor pixel in image,
Difference if less than default difference threshold accounts for the ratio of whole differences more than default proportion threshold value, then automatically by this image
It is identified as broad image.Due to smoother between broad image neighbor pixel, most of adjacent therefore in broad image
The difference of pixel also can be less, therefore passes through the present invention program, can more efficiently identify whether image is broad image.
Brief description
In order to be illustrated more clearly that the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
Have technology description in required use accompanying drawing be briefly described it should be apparent that, drawings in the following description be only this
Some embodiments of invention, for those of ordinary skill in the art, without having to pay creative labor, also may be used
So that other accompanying drawings are obtained according to these accompanying drawings.
One embodiment schematic flow sheet of fuzzy image recognition method that Fig. 1 provides for the present invention；
Another embodiment schematic flow sheet of fuzzy image recognition method that Fig. 2 provides for the present invention
One example structure schematic diagram of fuzzy image recognition device that Fig. 3 provides for the present invention.
Specific embodiment
For enabling the goal of the invention of the present invention, feature, advantage more obvious and understandable, below in conjunction with the present invention
Accompanying drawing in embodiment, is clearly and completely described the reality it is clear that described to the technical scheme in the embodiment of the present invention
Applying example is only a part of embodiment of the present invention, and not all embodiments.Based on the embodiment in the present invention, the common skill in this area
The every other embodiment that art personnel are obtained under the premise of not making creative work, broadly falls into the model of present invention protection
Enclose.
The embodiment of the present invention provides a kind of fuzzy image recognition method, comprising: obtain image；Calculate in above-mentioned image each
Difference between neighbor pixel；If the ratio accounting for whole differences less than the difference of default difference threshold is more than default ratio
Threshold value is it is determined that above-mentioned image is broad image.The embodiment of the present invention also provides corresponding fuzzy image recognition device, below divides
It is not described in detail.
Embodiment one
Refer to Fig. 1, the fuzzy image recognition method in the embodiment of the present invention includes:
Step 101, acquisition image；
In the embodiment of the present invention, obtaining needs to carry out the image of fuzzy image recognition, for example, can obtain gained of taking pictures in real time
Image, or, obtain take pictures gained and prepare for carry out examination question search or for other application image.
Difference between each neighbor pixel in step 102, the above-mentioned image of calculating；
In the embodiment of the present invention, the difference between each neighbor pixel in the image that calculation procedure 101 obtains.Illustrate
Bright, for the image of a 100*100, by calculate each neighbor pixel in this image (for example left and right neighbor pixel or
The neighbouring pixel of person) between difference, 99*99 difference can be obtained.
If the difference that step 103 is less than default difference threshold accounts for the ratio of whole differences more than default proportion threshold value,
Then determine that above-mentioned image is broad image；
In the embodiment of the present invention, preset above-mentioned difference threshold and proportion threshold value, when the difference less than this difference threshold
When the ratio accounting for whole differences is more than this proportion threshold value, determine that above-mentioned image is broad image.Illustrate, if proportion threshold value is
10%, for the image of a 100*100, by the calculating of step 102,99*99 difference can be obtained.When less than this difference
When the difference of threshold value accounts for the ratio of whole differences more than 10% (namely it is more than 99*99* less than the quantity of the difference of this difference threshold
When 10%), determine that above-mentioned image is broad image.Specifically, above-mentioned difference threshold and aforementioned proportion threshold value can be based on a large amount of
Research and experiment determine, are not construed as limiting herein.
Further, can also be before step 102, the image first step 101 being obtained carries out gray proces, obtains this figure
The gray-scale maps of picture, then step 102 be embodied in: calculate the difference between each neighbor pixel in the gray-scale maps of this image.
Further, after step 102, above-mentioned difference threshold whole differences calculated to step 102 can be based on
Carry out binary conversion treatment, above-mentioned binary conversion treatment also all differences more than or equal to this difference threshold will be set to 255, and will
The above-mentioned difference less than this difference threshold is set to 0, then, after binary conversion treatment, the calculated difference of step 102 will be turned
Turn to 0 or 255 numerical value.Step 103 is embodied in: if numerical value 0 is in the whole numerical value obtaining after above-mentioned binary conversion treatment
In shared ratio be more than aforementioned proportion threshold value it is determined that above-mentioned image is broad image.Illustrate, if aforementioned proportion threshold value
For 10%, the numerical value 0 that in image a, the difference of each neighbor pixel obtains after above-mentioned binary conversion treatment has 2000, and warp
The numerical value 255 obtaining after above-mentioned binary conversion treatment has 10000, then numerical value 0 proportion is 2000/10000 (namely 20%),
Because the shared ratio 20% in the whole numerical value obtaining after above-mentioned binary conversion treatment of numerical value 0 is more than 10%, therefore determine figure
As a is broad image.
Further, after step 103 determines that above-mentioned image is broad image, can be with the above-mentioned image of output indication as mould
The information of paste image.
It should be noted that the fuzzy image recognition method in the embodiment of the present invention can be real by fuzzy image recognition device
Existing, it is (for example intelligent that this fuzzy image recognition device specifically can be integrated in above-mentioned electric terminal with software (for example in the form of app)
The terminals such as mobile phone, panel computer, learning machine) in.
Therefore, the present invention program passes through to obtain image and calculate the difference between each neighbor pixel in image,
Difference if less than default difference threshold accounts for the ratio of whole differences more than default proportion threshold value, then automatically by this image
It is identified as broad image.Due to smoother between broad image neighbor pixel, most of adjacent therefore in broad image
The difference of pixel also can be less, therefore passes through the present invention program, can more efficiently identify whether image is broad image.
Embodiment two
The embodiment of the present invention is with the difference of embodiment one, the embodiment of the present invention each neighbor in calculating image
Before difference between point, first gray proces are carried out to the image obtaining, as shown in Fig. 2 the fuzzy graph in the embodiment of the present invention
As recognition methodss include:
Step 201, acquisition image；
In the embodiment of the present invention, obtaining needs to carry out the image of fuzzy image recognition, for example, can obtain gained of taking pictures in real time
Image, or, obtain take pictures gained and prepare for carry out examination question search or for other application image.
Step 202, gray proces are carried out to above-mentioned image, obtain the gray-scale maps of above-mentioned image；
In the embodiment of the present invention, the image that step 201 is obtained carries out gray proces, obtains the gray value of this image, tool
Body ground, carries out gray proces and is referred to prior art realization, be not construed as limiting herein to image.
Difference between each neighbor pixel in step 203, the gray-scale maps of the above-mentioned image of calculating；
In the embodiment of the present invention, the difference between each neighbor pixel in the gray-scale maps that calculation procedure 202 obtains after processing
Value.Illustrate, for the gray-scale maps of a 100*100, by calculating (the such as left and right of each neighbor pixel in this gray-scale map
Neighbor pixel or neighbouring pixel) between difference, 99*99 difference can be obtained.
If the difference that step 204 is less than default difference threshold accounts for the ratio of whole differences more than default proportion threshold value,
Then determine that above-mentioned image is broad image；
In the embodiment of the present invention, preset above-mentioned difference threshold and proportion threshold value, when the difference less than this difference threshold
When the ratio accounting for whole differences is more than this proportion threshold value, determine that above-mentioned image is broad image.
Further, after step 203, above-mentioned difference threshold whole differences calculated to step 203 can be based on
Carry out binary conversion treatment, above-mentioned binary conversion treatment also all differences more than or equal to this difference threshold will be set to 255, and will
The above-mentioned difference less than this difference threshold is set to 0, then, after binary conversion treatment, the calculated difference of step 102 will be turned
Turn to 0 or 255 numerical value.Step 204 is embodied in: if numerical value 0 is in the whole numerical value obtaining after above-mentioned binary conversion treatment
In shared ratio be more than aforementioned proportion threshold value it is determined that above-mentioned image is broad image.Illustrate, if aforementioned proportion threshold value
For 10%, the numerical value 0 that in image a, the difference of each neighbor pixel obtains after above-mentioned binary conversion treatment has 2000, and warp
The numerical value 255 obtaining after above-mentioned binary conversion treatment has 10000, then numerical value 0 proportion is 2000/10000 (namely 20%),
Because the shared ratio 20% in the whole numerical value obtaining after above-mentioned binary conversion treatment of numerical value 0 is more than 10%, therefore determine figure
As a is broad image.
Further, after step 204 determines that above-mentioned image is broad image, can be with the above-mentioned image of output indication as mould
The information of paste image.
It should be noted that the fuzzy image recognition method in the embodiment of the present invention can be real by fuzzy image recognition device
Existing, it is (for example intelligent that this fuzzy image recognition device specifically can be integrated in above-mentioned electric terminal with software (for example in the form of app)
The terminals such as mobile phone, panel computer, learning machine) in.
Therefore, the present invention program passes through to obtain image and calculate the difference between each neighbor pixel in image,
Difference if less than default difference threshold accounts for the ratio of whole differences more than default proportion threshold value, then automatically by this image
It is identified as broad image.Due to smoother between broad image neighbor pixel, most of adjacent therefore in broad image
The difference of pixel also can be less, therefore passes through the present invention program, can more efficiently identify whether image is broad image.
Embodiment three
In the embodiment of the present invention, a kind of fuzzy image recognition device is described, refers to Fig. 3, in the embodiment of the present invention
Fuzzy image recognition device 300 include:
Acquiring unit 301, for obtaining image；
Computing unit 302, for calculating the difference between each neighbor pixel in the image that acquiring unit 301 obtains；
Determining unit 303, the ratio for accounting for whole differences when the difference less than default difference threshold is more than default
During proportion threshold value, determine that above-mentioned image is broad image.
Optionally, the fuzzy image recognition device in the embodiment of the present invention also includes: gray proces unit, for above-mentioned
The image that acquiring unit obtains carries out gray proces, obtains the gray-scale maps of above-mentioned image；Computing unit 302 specifically for: calculate
Difference between each neighbor pixel in the gray-scale maps obtaining after above-mentioned gray proces cell processing.
Optionally, the fuzzy image recognition device in the embodiment of the present invention also includes: binary conversion treatment unit, for being based on
Above-mentioned difference threshold whole differences calculated to computing unit 302 carry out binary conversion treatment；Determining unit 303 is specifically used
In: when the shared ratio in the whole numerical value obtaining after above-mentioned binary conversion treatment cell processing of numerical value 0 is more than aforementioned proportion
During threshold value, determine that above-mentioned image is broad image.
Optionally, acquiring unit 301 specifically for: obtain the gained and prepare the image for carrying out examination question search of taking pictures.
Optionally, the fuzzy image recognition device in the embodiment of the present invention also includes: Tip element, for output indication
State the information that image is broad image.
It should be noted that fuzzy image recognition device in the embodiment of the present invention specifically can in the way of software (example
Form as app) it is integrated in above-mentioned electric terminal (the such as terminal such as smart mobile phone, panel computer, learning machine).
It should be understood that the fuzzy image recognition device in the embodiment of the present invention can be used for realizing in said method embodiment
All technical schemes, the function of its each functional module can implement according to the method in said method embodiment, its tool
Body realizes the associated description that process can refer in above-described embodiment, and here is omitted.
Therefore, the present invention program passes through to obtain image and calculate the difference between each neighbor pixel in image,
Difference if less than default difference threshold accounts for the ratio of whole differences more than default proportion threshold value, then automatically by this image
It is identified as broad image.Due to smoother between broad image neighbor pixel, most of adjacent therefore in broad image
The difference of pixel also can be less, therefore passes through the present invention program, can more efficiently identify whether image is broad image.
It should be understood that disclosed apparatus and method in several embodiments provided herein, can be passed through it
Its mode is realized.For example, device embodiment described above is only schematically, for example, the division of said units, and only
It is only a kind of division of logic function, actual can have other dividing mode when realizing, and for example multiple units or assembly can be tied
Close or be desirably integrated into another system, or some features can be ignored, or do not execute.Another, shown or discussed
Coupling each other or direct-coupling or communication connection can be by some interfaces, the INDIRECT COUPLING of device or unit or logical
Letter connects, and can be electrical, mechanical or other forms.
The above-mentioned unit illustrating as separating component can be or may not be physically separate, show as unit
The part showing can be or may not be physical location, you can with positioned at a place, or can also be distributed to multiple
On NE.The mesh to realize this embodiment scheme for some or all of unit therein can be selected according to the actual needs
's.
In addition, can be integrated in a processing unit in each functional unit in each embodiment of the present invention it is also possible to
It is that unit is individually physically present it is also possible to two or more units are integrated in a unit.Above-mentioned integrated list
Unit both can be to be realized in the form of hardware, it would however also be possible to employ the form of SFU software functional unit is realized.
If above-mentioned integrated unit is realized and as independent production marketing or use using in the form of SFU software functional unit
When, can be stored in a computer read/write memory medium.Based on such understanding, technical scheme is substantially
The part in other words prior art being contributed or all or part of this technical scheme can be in the form of software products
Embody, this computer software product is stored in a storage medium, including some instructions with so that a computer
Equipment (can be personal computer, server, or network equipment etc.) executes the complete of each embodiment said method of the present invention
Portion or part steps.And aforesaid storage medium includes: u disk, portable hard drive, read only memory (rom, read-only
Memory), random access memory (ram, random access memory), magnetic disc or CD etc. are various can store journey
The medium of sequence code.
It should be noted that for aforesaid each method embodiment, for easy description, therefore it is all expressed as a series of
Combination of actions, but those skilled in the art should know, the present invention is not limited by described sequence of movement because
According to the present invention, some steps can be carried out using other orders or simultaneously.Secondly, those skilled in the art also should know
Know, embodiment described in this description belongs to preferred embodiment, and involved action and module might not be all these
Bright necessary.
In the above-described embodiments, the description to each embodiment all emphasizes particularly on different fields, and does not have the portion described in detail in certain embodiment
Point, may refer to the associated description of other embodiments.
It is more than to a kind of description of fuzzy image recognition method and apparatus provided by the present invention, for the one of this area
As technical staff, according to the embodiment of the present invention thought, all will change in specific embodiments and applications, comprehensive
On, this specification content should not be construed as limitation of the present invention.
Claims (10)
1. a kind of fuzzy image recognition method is it is characterised in that include:
Obtain image；
Calculate the difference between each neighbor pixel in described image；
If the ratio accounting for whole differences less than the difference of default difference threshold is more than default proportion threshold value it is determined that described figure
Picture is broad image.
2. method according to claim 1 is it is characterised in that in described calculating described image between each neighbor pixel
Difference before include:
Gray proces are carried out to described image, obtains the gray-scale maps of described image；
Difference between each neighbor pixel in described calculating described image, particularly as follows:
Calculate the difference between each neighbor pixel in the gray-scale maps of described image.
3. method according to claim 2 is it is characterised in that each adjacent picture in the gray-scale maps of described calculating described image
Also include after difference between vegetarian refreshments:
Binary conversion treatment is carried out to calculated whole differences based on described difference threshold；
If the ratio that the described difference less than default difference threshold accounts for whole differences is more than default proportion threshold value it is determined that institute
Stating image is broad image, particularly as follows:
If the shared ratio in the whole numerical value obtaining after described binary conversion treatment of numerical value 0 is more than described proportion threshold value,
Determine that described image is broad image.
4. the method according to any one of claims 1 to 3 is it is characterised in that described acquisition image is taken pictures particularly as follows: obtaining
Gained and prepare the image for carrying out examination question search.
5. the method according to any one of claims 1 to 3 it is characterised in that described determination described image be broad image,
Also include afterwards:
Output indication described image is the information of broad image.
6. a kind of fuzzy image recognition device is it is characterised in that include:
Acquiring unit, for obtaining image；
Computing unit, for calculating the difference between each neighbor pixel in the image that described acquiring unit obtains；
Determining unit, the ratio for accounting for whole differences when the difference less than default difference threshold is more than default proportion threshold value
When, determine that described image is broad image.
7. fuzzy image recognition device according to claim 6 is it is characterised in that described fuzzy image recognition device also wraps
Include:
Gray proces unit, the image for obtaining to described acquiring unit carries out gray proces, obtains the gray scale of described image
Figure；
Described computing unit specifically for: calculate each adjacent picture in the gray-scale maps that obtain after described gray proces cell processing
Difference between vegetarian refreshments.
8. fuzzy image recognition device according to claim 7 is it is characterised in that described fuzzy image recognition device also wraps
Include:
Binary conversion treatment unit, for carrying out two based on described difference threshold whole differences calculated to described computing unit
Value is processed；
Described determining unit specifically for: when numerical value 0 is in the whole numerical value obtaining after described binary conversion treatment cell processing
When shared ratio is more than described proportion threshold value, determine that described image is broad image.
9. the fuzzy image recognition device according to any one of claim 6 to 8 is it is characterised in that described acquiring unit has
Body is used for: obtains and takes pictures gained and prepare the image for carrying out examination question search.
10. the fuzzy image recognition device according to any one of claim 6 to 8 is it is characterised in that described broad image is known
Other device also includes:
Tip element, is the information of broad image for output indication described image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610828484.8A CN106373121A (en) | 2016-09-18 | 2016-09-18 | Fuzzy image identification method and apparatus |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610828484.8A CN106373121A (en) | 2016-09-18 | 2016-09-18 | Fuzzy image identification method and apparatus |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106373121A true CN106373121A (en) | 2017-02-01 |
Family
ID=57897595
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610828484.8A Pending CN106373121A (en) | 2016-09-18 | 2016-09-18 | Fuzzy image identification method and apparatus |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106373121A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107545251A (en) * | 2017-08-31 | 2018-01-05 | 北京图铭视界科技有限公司 | Face quality discrimination and the method and device of picture enhancing |
| CN107992872A (en) * | 2017-12-25 | 2018-05-04 | 广东小天才科技有限公司 | A kind of method and mobile terminal that text identification is carried out to picture |
| CN109558878A (en) * | 2017-09-27 | 2019-04-02 | 北京国双科技有限公司 | Image-recognizing method and device |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101051351A (en) * | 2007-05-23 | 2007-10-10 | 重庆医科大学 | Image band parameter two-valued method and device using said method |
| CN104239909A (en) * | 2014-08-11 | 2014-12-24 | 北京捷通华声语音技术有限公司 | Method and device for recognizing images |
| CN105069783A (en) * | 2015-07-23 | 2015-11-18 | 北京金山安全软件有限公司 | Fuzzy picture identification method and device |
-
2016
- 2016-09-18 CN CN201610828484.8A patent/CN106373121A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101051351A (en) * | 2007-05-23 | 2007-10-10 | 重庆医科大学 | Image band parameter two-valued method and device using said method |
| CN104239909A (en) * | 2014-08-11 | 2014-12-24 | 北京捷通华声语音技术有限公司 | Method and device for recognizing images |
| CN105069783A (en) * | 2015-07-23 | 2015-11-18 | 北京金山安全软件有限公司 | Fuzzy picture identification method and device |
Non-Patent Citations (4)
| Title |
|---|
| SANG KU KIM 等: "Simultaneous Out-of-focus Blur Estimation And Restoration For A Digital Auto-focusing System", 《INTERNATIONAL CONFERENCE ON CONSUMER ELECTRONICS》 * |
| 孙农亮 等: "运行模糊图像清晰度评价函数的研究", 《仪器仪表学报》 * |
| 洪裕珍 等: "离焦模糊图像清晰度评价函数的分析与改进", 《光学 精密工程》 * |
| 龚声蓉 等: "《数字图像处理与分析》", 31 May 2014, 清华大学出版社 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107545251A (en) * | 2017-08-31 | 2018-01-05 | 北京图铭视界科技有限公司 | Face quality discrimination and the method and device of picture enhancing |
| CN109558878A (en) * | 2017-09-27 | 2019-04-02 | 北京国双科技有限公司 | Image-recognizing method and device |
| CN109558878B (en) * | 2017-09-27 | 2022-11-22 | 北京国双科技有限公司 | Image recognition method and device |
| CN107992872A (en) * | 2017-12-25 | 2018-05-04 | 广东小天才科技有限公司 | A kind of method and mobile terminal that text identification is carried out to picture |
| CN107992872B (en) * | 2017-12-25 | 2020-04-28 | 广东小天才科技有限公司 | Method for carrying out text recognition on picture and mobile terminal |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20210224598A1 (en) | Method for training deep learning model, electronic equipment, and storage medium | |
| CN106295591A (en) | Gender identification method based on facial image and device | |
| CN107194158A (en) | A kind of disease aided diagnosis method based on image recognition | |
| CN110827236B (en) | Brain tissue layering method, device and computer equipment based on neural network | |
| CN111598860B (en) | Lithium battery defect detection method based on yolov3 network embedded into self-attention door module | |
| CN106407911A (en) | Image-based eyeglass recognition method and device | |
| CN109858476B (en) | Tag expansion method and electronic equipment | |
| CN106326857A (en) | Gender identification method and gender identification device based on face image | |
| CN113033537A (en) | Method, apparatus, device, medium and program product for training a model | |
| CN111291825A (en) | Focus classification model training method and device, computer equipment and storage medium | |
| CN112257665A (en) | Image content recognition method, image recognition model training method, and medium | |
| CN110046941A (en) | A kind of face identification method, system and electronic equipment and storage medium | |
| CN106373121A (en) | Fuzzy image identification method and apparatus | |
| CN112862005A (en) | Video classification method and device, electronic equipment and storage medium | |
| CN112580666A (en) | Image feature extraction method, training method, device, electronic equipment and medium | |
| CN114022748B (en) | Target identification method, device, equipment and storage medium | |
| CN114639152A (en) | Multi-modal voice interaction method, device, equipment and medium based on face recognition | |
| CN111738080A (en) | Face detection and alignment method and device | |
| CN109543744B (en) | Multi-category deep learning image identification method based on Loongson group and application thereof | |
| CN112364828B (en) | Face recognition method and financial system | |
| CN114419087A (en) | Focus image generation method and device, electronic equipment and storage medium | |
| CN114783597B (en) | Method and device for diagnosing multi-class diseases, electronic equipment and storage medium | |
| EP4220552A1 (en) | Image processing method and system | |
| CN113591580B (en) | Image annotation method and device, electronic equipment and storage medium | |
| CN112329735B (en) | Training method of face recognition model and online education system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | 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: 20170201 |