CN109829503A - A kind of intensive frightened picture method of discrimination, system, equipment and its storage medium - Google Patents
A kind of intensive frightened picture method of discrimination, system, equipment and its storage medium Download PDFInfo
- Publication number
- CN109829503A CN109829503A CN201910110928.8A CN201910110928A CN109829503A CN 109829503 A CN109829503 A CN 109829503A CN 201910110928 A CN201910110928 A CN 201910110928A CN 109829503 A CN109829503 A CN 109829503A
- Authority
- CN
- China
- Prior art keywords
- picture
- discriminated
- characteristic value
- frightened
- intensive
- 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.)
- Granted
Links
Landscapes
- Image Analysis (AREA)
Abstract
The present invention relates to technical field of network information, more particularly to a kind of intensive frightened picture method of discrimination, system, equipment and its storage medium, the method comprise the steps that being selected respectively to positive and negative samples picture, the invariable rotary characteristic value of chosen picture is calculated;Its invariable rotary characteristic value is subjected to deep learning respectively, obtains identifying model;Picture to be discriminated is selected, the invariable rotary characteristic value of chosen picture is calculated;And import its invariable rotary characteristic value in the identification model, obtain the type of picture to be discriminated;The present invention can differentiate most pictures, versatile, and accuracy rate is high, meets the requirement of automatic discrimination, greatly improve the working efficiency of picture differentiation.
Description
Technical field
The present invention relates to technical field of network information, and in particular to a kind of intensive frightened picture method of discrimination and a kind of close
Collect the system of frightened picture method of discrimination.
Background technique
With the fast development of network technology, network has become the important channel that people obtain information;And existing
It there is various image documents on modern network, and since there is the phases of dense arrangement in part of image document
To lesser object, the globe neurosis of people is easily caused, the discomfort of people is caused.
For this purpose, website now all can to causing the intensive frightened picture of globe neurosis to be differentiated and screened,
But for existing discriminant approach mainly based on artificial cognition, working efficiency is more low, is not enough to deal with network Shanghai amount
Image document.
Summary of the invention
To overcome drawbacks described above, the purpose of the present invention is that providing one kind is feared based on machine depth learning technology to intensive
Fear picture to carry out sentencing method for distinguishing and its system.
The purpose of the present invention is achieved through the following technical solutions:
The present invention is a kind of intensive frightened picture method of discrimination, comprising:
Establish the sample graph valut containing positive sample picture and negative sample picture;
Positive and negative samples picture in the sample graph valut is selected respectively, not to the rotation of chosen picture
Become characteristic value to be calculated;
Invariable rotary characteristic value in the positive and negative samples picture is subjected to deep learning respectively, obtains identifying model, institute
It states to identify and contains positive template and negative norm plate in model;
Picture to be discriminated is selected, the invariable rotary characteristic value of chosen picture is calculated;
Invariable rotary characteristic value in the picture to be discriminated is imported in the identification model, figure to be discriminated is obtained
The type of piece.
In the present invention, include: before the invariable rotary characteristic value of described pair of chosen picture is calculated
Whether the chosen picture of judgement is scheduled format, if not scheduled format, then picture is adjusted to predetermined
Format.
In the present invention, described to judge whether picture is scheduled format, if not scheduled format, then adjust picture
Include: for scheduled format
Judge whether the positive and negative samples picture is grayscale image, if not grayscale image, is then adjusted to grayscale image.
In the present invention, described to judge whether picture is scheduled format, if not scheduled format, then adjust picture
For scheduled format further include:
Whether the size value for judging the positive and negative samples picture is predetermined size, if its size value is not predetermined size,
It is adjusted to predetermined size.
In the present invention, described to select picture to be discriminated and include:
Picture to be discriminated is imported in picture library to be discriminated, and to described to be discriminated from the picture library to be discriminated
Picture is selected.
In the present invention, the invariable rotary characteristic value of described pair of chosen picture, which calculate, includes:
It is center pixel that a pixel is chosen from the chosen picture, and obtains pixel adjacent thereto
Point, all pixels adjacent with center pixel form the neighborhood of the center pixel;
The gray value of the center pixel is compared with the gray value of all adjacent pixels in the neighborhood
And count, obtain the primitive character value of the center pixel;
Centered on the center pixel, moving in rotation is carried out to the adjacent pixel in the neighborhood, is carried out again
It calculates, obtains the new primitive character value of the center pixel;
All primitive character values are subjected to size comparison, and special using the smallest primitive character value of numerical value as invariable rotary
Value indicative.
In the present invention, the invariable rotary characteristic value by the picture to be discriminated imports the identification model
In, the type for obtaining picture to be discriminated includes:
Invariable rotary characteristic value in the picture to be discriminated is imported in the identification model, with the positive template and
Negative norm plate is matched respectively;
If described match with the positive template, which is defined as intensive frightened picture;If with institute
It states negative norm plate to match, then picture to be discriminated is defined as non-dense set fear picture by this.
The present invention is a kind of intensive frightened picture judgement system comprising:
Sample graph valut, the samples pictures inventory contain positive sample picture and negative sample picture;
Picture library to be discriminated is stored with picture to be discriminated in the picture library to be discriminated;
Picture chosen module, the picture chosen module respectively with the sample graph valut and the picture library phase to be discriminated
Connection, for being selected to positive and negative samples picture or picture to be discriminated;
Characteristic value calculating module, the characteristic value calculating module are connected with the picture chosen module, for selected
The invariable rotary characteristic value of fixed positive and negative samples picture or picture to be discriminated is calculated;
Model building module, the model building module are connected with the characteristic value calculating module, for by it is described just,
Invariable rotary characteristic value in negative sample picture carries out deep learning respectively, obtains identifying model, contain in the identification model
Positive template and negative norm plate;
Picture match module, the picture match module respectively with the characteristic value calculating module and the model foundation mould
Block is connected, and for importing the invariable rotary characteristic value in the picture to be discriminated in the identification model, obtains wait sentence
The type of other picture.
The present invention is a kind of electronic equipment, and the electronic equipment includes:
Processor;
Memory is stored with computer-readable instruction on the memory, and the computer-readable instruction is by the processing
When device executes, intensive frightened picture method of discrimination as described above is realized.
The present invention is a kind of computer readable storage medium, is stored thereon with computer program, the computer program quilt
When processor executes, intensive frightened picture method of discrimination as described above is realized.
The present invention combines depth learning technology to generate and identifies model by obtaining the feature of positive and negative sample pictures, then
Picture to be discriminated is imported and is identified in model, show whether picture to be discriminated is intensive frightened picture by comparing;This hair
It is bright most pictures to be differentiated, it is versatile, and accuracy rate is high, meets the requirement of automatic discrimination, greatly mentions
The working efficiency that high picture differentiates.
Detailed description of the invention
The present invention is described in detail by following preferred embodiments and attached drawing for ease of explanation,.
Fig. 1 is the workflow schematic diagram of the intensive frightened picture method of discrimination one embodiment of the present invention;
Fig. 2 is the workflow schematic diagram of intensive another embodiment of frightened picture method of discrimination of the present invention;
Fig. 3 is the workflow schematic diagram that the invariable rotary characteristic value of samples pictures is calculated in the present invention;
Fig. 4 is the logical construction schematic diagram of one embodiment of the intensive frightened picture judgement system of the present invention.
Specific embodiment
In order to make the objectives, technical solutions, and advantages of the present invention clearer, with reference to the accompanying drawings and embodiments, right
The present invention is further described.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, and do not have to
It is of the invention in limiting.
In the description of the present invention, it is to be understood that, term " center ", " longitudinal direction ", " transverse direction ", " length ", " width ",
" thickness ", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outside", " up time
The orientation or positional relationship of the instructions such as needle ", " counterclockwise " is to be based on the orientation or positional relationship shown in the drawings, and is merely for convenience of
The description present invention and simplified description, rather than the device or element of indication or suggestion meaning must have a particular orientation, with spy
Fixed orientation construction and operation, therefore be not considered as limiting the invention.In addition, term " first ", " second " are only used for
Purpose is described, relative importance is not understood to indicate or imply or implicitly indicates the quantity of indicated technical characteristic.
" first " is defined as a result, the feature of " second " can explicitly or implicitly include one or more feature.?
In description of the invention, the meaning of " plurality " is two or more, unless otherwise specifically defined.
In the description of the present invention, it should be noted that unless otherwise clearly defined and limited, term " installation ", " phase
Even ", " connection " shall be understood in a broad sense, for example, it may be being fixedly connected, may be a detachable connection, or be integrally connected.It can
To be mechanical connection, it is also possible to be electrically connected.It can be directly connected, can also can be indirectly connected through an intermediary
The interaction relationship of connection or two elements inside two elements.It for the ordinary skill in the art, can be with
The concrete meaning of above-mentioned term in the present invention is understood as the case may be.
Below by taking one embodiment as an example, intensive frightened picture method of discrimination of the invention is specifically described, please be join
Read Fig. 1 comprising:
S101. sample graph valut is established
Establish the sample graph valut containing positive sample picture and negative sample picture;Wherein, positive sample picture is to be judged to
It is set to the picture of intensive frightened picture, negative sample picture is the picture for being judged as non-dense set fear picture.
S102. the invariable rotary characteristic value of samples pictures is calculated
Positive and negative samples picture in the sample graph valut is selected respectively, to chosen positive and negative samples picture
Invariable rotary characteristic value calculated, respectively obtain the invariable rotary characteristic value of positive sample picture and the rotation of negative sample picture
Invariant features value.
S103. deep learning is carried out to obtain identifying model
Invariable rotary characteristic value in multiple positive sample pictures is subjected to deep learning, obtains positive template;It will be multiple
Invariable rotary characteristic value in the negative sample picture carries out deep learning, obtains negative norm plate;The positive template and negative norm plate are constituted
Identification model.Wherein, deep learning is a kind of based on the method for carrying out representative learning to data in machine learning.Observation
Various ways can be used to indicate in (such as piece image), such as the vector of each pixel intensity value, or more abstractively indicate
At a series of sides, region of specific shape etc..And certain specific representation methods is used to be easier the learning tasks (example from example
Such as, recognition of face or human facial expression recognition).The benefit of deep learning is with the feature learning of non-supervisory formula or Semi-supervised and to divide
Layer feature extraction highly effective algorithm obtains feature to substitute by hand.
S104. the invariable rotary characteristic value of chosen picture is calculated
Picture to be discriminated is selected, the invariable rotary characteristic value of chosen picture is calculated;
S105. the type for identifying and obtaining picture in model is imported
Invariable rotary characteristic value in the picture to be discriminated is imported in the identification model, figure to be discriminated is obtained
The type of piece.
For a better understanding of the present invention, below by taking another embodiment as an example, intensive frightened picture of the invention is sentenced
Other method is specifically described, referring to Fig. 2, comprising:
S201. sample graph valut is established
Establish the sample graph valut containing positive sample picture and negative sample picture;Wherein, positive sample picture is to be judged to
It is set to the picture of intensive frightened picture, negative sample picture is the picture for being judged as non-dense set fear picture.
S202. the picture format of positive and negative samples picture is adjusted
Positive and negative samples picture in the sample graph valut is selected respectively, judge chosen picture whether be
Picture is then adjusted to scheduled format if not scheduled format by scheduled format;
Wherein, described picture is adjusted to scheduled format to include:
Judge whether the positive and negative samples picture is grayscale image, if not grayscale image, is then adjusted to grayscale image;Its
Chosen image must be grayscale image, if it is cromogram, need to be first converted into grayscale image.
It whether is predetermined size with the size value for judging the positive and negative samples picture, if its size value is not predetermined size,
Then it is adjusted to predetermined size;The size of its preferred grayscale image is 224*224, is handled convenient for system image.
S203. the invariable rotary characteristic value of samples pictures is calculated
The invariable rotary characteristic value of chosen positive and negative samples picture is calculated, positive sample picture is respectively obtained
The invariable rotary characteristic value of invariable rotary characteristic value and negative sample picture.
Wherein, carry out calculate picture invariable rotary characteristic value include:
S2031. center pixel and its neighborhood are selected
It is center pixel that a pixel is chosen from the chosen picture, and obtains pixel adjacent thereto
Point, all pixels adjacent with center pixel form the neighborhood of the center pixel;A center pixel is chosen, and is obtained in this
Neighborhood composed by pixel in the adjacent region 3*3 of imago element.
S2032. the primitive character value of center pixel is obtained
The gray value of the center pixel is compared with the gray value of all adjacent pixels in the neighborhood
And count, obtain the primitive character value of the center pixel;In the neighborhood of center pixel 3*3, using centre of neighbourhood pixel as threshold value,
The adjacent gray value of 8 pixels is compared with the pixel value of the centre of neighbourhood, should if surrounding pixel is greater than center pixel value
The position of pixel is marked as 1, is otherwise 0.In this way, 8 points in 3*3 neighborhood can produce 8 bits by comparing,
This 8 bit is arranged successively to form a binary digit, this binary value is the primitive character of center pixel
Value.
S2033. neighborhood moving in rotation obtains new primitive character value
Centered on the center pixel, moving in rotation is carried out to the adjacent pixel in the neighborhood, is carried out again
It calculates, obtains the new primitive character value of the center pixel;
Its specifically: neighborhood is rotated clockwise first, and according to selection, different starting points obtains a series of LBP feature
Value, from the original LBP feature of these LBP characteristic values selection LBP characteristic value pixel centered on the smallest.
Above procedure is formulated are as follows:
Wherein (xc, yc) be center pixel coordinate, b be field in specify starting pixels, PbTo be calculated by starting point of b
P when single LBP value in neighborhoodbA pixel,For PbThe gray value of a neighborhood territory pixel, icFor the gray value of center pixel,
And s (x) is sign function:
S2034. invariable rotary characteristic value is determined
All primitive character values are subjected to size comparison, and special using the smallest primitive character value of numerical value as invariable rotary
Value indicative.
S204. deep learning is carried out to obtain identifying model
Invariable rotary characteristic value in multiple positive sample pictures is subjected to deep learning, obtains positive template;It will be multiple
Invariable rotary characteristic value in the negative sample picture carries out deep learning, obtains negative norm plate;The positive template and negative norm plate are constituted
Identification model.Wherein, deep learning is a kind of based on the method for carrying out representative learning to data in machine learning.Observation
Various ways can be used to indicate in (such as piece image), such as the vector of each pixel intensity value, or more abstractively indicate
At a series of sides, region of specific shape etc..And certain specific representation methods is used to be easier the learning tasks (example from example
Such as, recognition of face or human facial expression recognition).The benefit of deep learning is with the feature learning of non-supervisory formula or Semi-supervised and to divide
Layer feature extraction highly effective algorithm obtains feature to substitute by hand.
S205. the picture format of picture to be discriminated is adjusted
Picture to be discriminated is selected, judges whether chosen picture to be discriminated is scheduled format, if not predetermined
Format, then picture is adjusted to scheduled format;
Wherein, described picture is adjusted to scheduled format to include:
Judge whether the positive and negative samples picture is grayscale image, if not grayscale image, is then adjusted to grayscale image;Its
Chosen image must be grayscale image, if it is cromogram, need to be first converted into grayscale image.
It whether is predetermined size with the size value for judging the positive and negative samples picture, if its size value is not predetermined size,
Then it is adjusted to predetermined size;The size of its preferred grayscale image is 224*224, is handled convenient for system image.
S206. the invariable rotary characteristic value of chosen picture is calculated
The invariable rotary characteristic value of chosen picture is calculated;
Wherein, carry out calculating the method and step S2031-S2034 phase of the invariable rotary characteristic value of picture in this step
Unanimously.
S207. the identification model is imported to be matched
Invariable rotary characteristic value in the picture to be discriminated is imported in the identification model, with the positive template and
Negative norm plate is matched respectively;
S208. the type of picture to be discriminated is obtained
If matching with the positive template, which is defined as intensive frightened picture;If with described negative
Template matches, then picture to be discriminated is defined as non-dense set fear picture by this;According to matching result, figure to be discriminated is obtained
The type of piece.
Fig. 3 is please referred to, the present invention is a kind of intensive frightened picture judgement system comprising:
Sample graph valut 301, the sample graph valut 301 are stored with positive sample picture and negative sample picture;
Picture library 302 to be discriminated is stored with picture to be discriminated in the picture library 302 to be discriminated;
Picture chosen module 303, the picture chosen module 303 respectively with the sample graph valut 301 and described wait sentence
Other picture library 302 is connected, for selecting to positive and negative samples picture or picture to be discriminated;
Characteristic value calculating module 304, the characteristic value calculating module 304 are connected with the picture chosen module 303, use
It is calculated in the invariable rotary characteristic value to chosen positive and negative samples picture or picture to be discriminated;
Model building module 305, the model building module 305 are connected with the characteristic value calculating module 304, are used for
Invariable rotary characteristic value in the positive and negative samples picture is subjected to deep learning respectively, obtains identifying model, the identification mould
Contain positive template and negative norm plate in type;
Picture match module 306, the picture match module 306 respectively with the characteristic value calculating module 304 and described
Model building module 305 is connected, for the invariable rotary characteristic value in the picture to be discriminated to be imported the identification mould
In type, the type of picture to be discriminated is obtained.
Involved module can be realized by way of software in the present embodiment, can also by way of hardware come
It realizes, described module also can be set in the processor.Wherein, the title of these modules is not constituted under certain conditions
Restriction to the unit itself.
The present invention can be a kind of electronic equipment, and the electronic equipment includes:
Processor;
Memory is stored with computer-readable instruction on the memory, and the computer-readable instruction is by the processing
When device executes, intensive frightened picture method of discrimination as described above is realized.
As on the other hand, present invention also provides a kind of computer-readable medium, which be can be
Included in electronic equipment described in above-described embodiment;It is also possible to individualism, and without in the supplying electronic equipment.
Above-mentioned computer-readable medium carries one or more program, when the electronics is set by one for said one or multiple programs
When standby execution, so that the electronic equipment is realized such as above-mentioned intensive frightened picture method of discrimination as described in the examples.
The present invention can also be a kind of computer readable storage medium, be stored thereon with computer program, the computer
When program is executed by processor, intensive frightened picture method of discrimination as described above is realized.For example, the present embodiment includes a kind of meter
Calculation machine program product comprising carry computer program on a computer-readable medium, which includes for holding
The program code of method shown in row above method process.
It should be noted that computer-readable medium shown in the present invention can be computer-readable signal media or meter
Calculation machine readable storage medium storing program for executing either the two any combination.Computer readable storage medium for example can be --- but not
Be limited to --- electricity, magnetic, optical, electromagnetic, infrared ray or semiconductor system, device or device, or any above combination.Meter
The more specific example of calculation machine readable storage medium storing program for executing can include but is not limited to: have the electrical connection, just of one or more conducting wires
Taking formula computer disk, hard disk, random access storage device (RAM), read-only memory (ROM), erasable type may be programmed read-only storage
Device (EPROM or flash memory), optical fiber, portable compact disc read-only memory (CD-ROM), light storage device, magnetic memory device,
Or above-mentioned any appropriate combination.In the present invention, computer readable storage medium can be it is any include or storage journey
The tangible medium of sequence, the program can be commanded execution system, device or device use or in connection.And at this
In invention, computer-readable signal media may include in a base band or as carrier wave a part propagate data-signal,
Wherein carry computer-readable program code.The data-signal of this propagation can take various forms, including but unlimited
In electromagnetic signal, optical signal or above-mentioned any appropriate combination.Computer-readable signal media can also be that computer can
Any computer-readable medium other than storage medium is read, which can send, propagates or transmit and be used for
By the use of instruction execution system, device or device or program in connection.Include on computer-readable medium
Program code can transmit with any suitable medium, including but not limited to: wireless, electric wire, optical cable, RF etc. are above-mentioned
Any appropriate combination.
In the description of this specification, reference term " embodiment ", " some embodiments ", " schematically implementation
The description of mode ", " example ", " specific example " or " some examples " etc. means embodiment or example is combined to describe specific
Feature, structure, material or feature are contained at least one embodiment or example of the invention.In the present specification, right
The schematic representation of above-mentioned term is not necessarily referring to identical embodiment or example.Moreover, the specific features of description, knot
Structure, material or feature can be combined in any suitable manner in any one or more embodiments or example.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all in essence of the invention
Made any modifications, equivalent replacements, and improvements etc., should all be included in the protection scope of the present invention within mind and principle.
Claims (10)
1. a kind of intensive frightened picture method of discrimination characterized by comprising
Establish the sample graph valut containing positive sample picture and negative sample picture;
Positive and negative samples picture in the sample graph valut is selected respectively, it is special to the invariable rotary of chosen picture
Value indicative is calculated;
Invariable rotary characteristic value in the positive and negative samples picture is subjected to deep learning respectively, obtains identifying model, the mirror
Contain positive template and negative norm plate in other model;
Picture to be discriminated is selected, the invariable rotary characteristic value of chosen picture is calculated;
Invariable rotary characteristic value in the picture to be discriminated is imported in the identification model, picture to be discriminated is obtained
Type.
2. intensive frightened picture method of discrimination according to claim 1, which is characterized in that described pair of chosen picture
Invariable rotary characteristic value includes: before being calculated
Whether the chosen picture of judgement is scheduled format, if not scheduled format, then picture is adjusted to scheduled lattice
Formula.
3. intensive frightened picture method of discrimination according to claim 2, which is characterized in that described to judge whether picture is pre-
Fixed format, if not scheduled format, then picture is adjusted to scheduled format includes:
Judge whether the positive and negative samples picture is grayscale image, if not grayscale image, is then adjusted to grayscale image.
4. intensive frightened picture method of discrimination according to claim 3, which is characterized in that described to judge whether picture is pre-
Picture is then adjusted to scheduled format if not scheduled format by fixed format further include:
Whether the size value for judging the positive and negative samples picture is predetermined size, if its size value is not predetermined size, by it
It is adjusted to predetermined size.
5. intensive frightened picture method of discrimination according to claim 4, which is characterized in that described to select picture to be discriminated
Include:
Picture to be discriminated is imported in picture library to be discriminated, and to the picture to be discriminated from the picture library to be discriminated
It is selected.
6. intensive frightened picture method of discrimination according to claim 5, which is characterized in that described pair of chosen picture
Invariable rotary characteristic value calculate
It is center pixel that a pixel is chosen from the chosen picture, and obtains pixel adjacent thereto, institute
The neighborhood of the center pixel is made of the pixel adjacent with center pixel;
The gray value of the center pixel is compared and is counted with the gray value of all adjacent pixels in the neighborhood
Number, obtains the primitive character value of the center pixel;
Centered on the center pixel, moving in rotation is carried out to the adjacent pixel in the neighborhood, is calculated again,
Obtain the new primitive character value of the center pixel;
All primitive character values are subjected to size comparison, and using the smallest primitive character value of numerical value as invariable rotary feature
Value.
7. intensive frightened picture method of discrimination according to claim 6, which is characterized in that described by the figure to be discriminated
Invariable rotary characteristic value in piece imports in the identification model, and the type for obtaining picture to be discriminated includes:
Invariable rotary characteristic value in the picture to be discriminated is imported in the identification model, with the positive template and negative norm
Plate is matched respectively;
If described match with the positive template, which is defined as intensive frightened picture;If with described negative
Template matches, then picture to be discriminated is defined as non-dense set fear picture by this.
8. a kind of intensive frightened picture judgement system characterized by comprising
Sample graph valut, the samples pictures inventory contain positive sample picture and negative sample picture;
Picture library to be discriminated is stored with picture to be discriminated in the picture library to be discriminated;
Picture chosen module, the picture chosen module are connected with the sample graph valut and the picture library to be discriminated respectively
It connects, for being selected to positive and negative samples picture or picture to be discriminated;
Characteristic value calculating module, the characteristic value calculating module is connected with the picture chosen module, for chosen
The invariable rotary characteristic value of positive and negative samples picture or picture to be discriminated is calculated;
Model building module, the model building module are connected with the characteristic value calculating module, are used for the positive and negative sample
Invariable rotary characteristic value in this picture carries out deep learning respectively, obtains identifying model, contains holotype in the identification model
Plate and negative norm plate;
Picture match module, the picture match module respectively with the characteristic value calculating module and the model building module phase
Connection obtains to be discriminated for importing the invariable rotary characteristic value in the picture to be discriminated in the identification model
The type of picture.
9. a kind of electronic equipment, which is characterized in that the electronic equipment includes:
Processor;
Memory is stored with computer-readable instruction on the memory, and the computer-readable instruction is held by the processor
When row, intensive frightened picture method of discrimination as described in any one of claim 1 to 7 is realized.
10. a kind of computer readable storage medium, which is characterized in that be stored thereon with computer program, the computer program
When being executed by processor, intensive frightened picture method of discrimination as described in any one of claim 1 to 7 is realized.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910110928.8A CN109829503B (en) | 2019-02-12 | 2019-02-12 | Dense fear picture distinguishing method, system, equipment and storage medium thereof |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910110928.8A CN109829503B (en) | 2019-02-12 | 2019-02-12 | Dense fear picture distinguishing method, system, equipment and storage medium thereof |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109829503A true CN109829503A (en) | 2019-05-31 |
CN109829503B CN109829503B (en) | 2021-12-17 |
Family
ID=66863574
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910110928.8A Active CN109829503B (en) | 2019-02-12 | 2019-02-12 | Dense fear picture distinguishing method, system, equipment and storage medium thereof |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109829503B (en) |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103605811A (en) * | 2013-12-10 | 2014-02-26 | 三峡大学 | Texture image retrieval method and device |
CN104077613A (en) * | 2014-07-16 | 2014-10-01 | 电子科技大学 | Crowd density estimation method based on cascaded multilevel convolution neural network |
CN104239896A (en) * | 2014-09-04 | 2014-12-24 | 四川省绵阳西南自动化研究所 | Method for classifying crowd density degrees in video image |
CN104268528A (en) * | 2014-09-28 | 2015-01-07 | 深圳市科松电子有限公司 | Method and device for detecting crowd gathered region |
CN107506692A (en) * | 2017-07-21 | 2017-12-22 | 天津大学 | A kind of dense population based on deep learning counts and personnel's distribution estimation method |
CN108986824A (en) * | 2018-07-09 | 2018-12-11 | 宁波大学 | A kind of voice playback detection method |
-
2019
- 2019-02-12 CN CN201910110928.8A patent/CN109829503B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103605811A (en) * | 2013-12-10 | 2014-02-26 | 三峡大学 | Texture image retrieval method and device |
CN104077613A (en) * | 2014-07-16 | 2014-10-01 | 电子科技大学 | Crowd density estimation method based on cascaded multilevel convolution neural network |
CN104239896A (en) * | 2014-09-04 | 2014-12-24 | 四川省绵阳西南自动化研究所 | Method for classifying crowd density degrees in video image |
CN104268528A (en) * | 2014-09-28 | 2015-01-07 | 深圳市科松电子有限公司 | Method and device for detecting crowd gathered region |
CN107506692A (en) * | 2017-07-21 | 2017-12-22 | 天津大学 | A kind of dense population based on deep learning counts and personnel's distribution estimation method |
CN108986824A (en) * | 2018-07-09 | 2018-12-11 | 宁波大学 | A kind of voice playback detection method |
Non-Patent Citations (3)
Title |
---|
YUE LI 等: "Estimation of Crowd Density based on Adaptive LBP", 《ADVANCED MATERIALS RESEARCH》 * |
何其佳 等: "基于LBP和极限学习机的脑部MR图像分类", 《山东大学学报(工学版)》 * |
成金庚: "基于深度学习的人群密度估计及行为分析方法研究", 《中国优秀硕士学位论文全文数据库社会科学II辑》 * |
Also Published As
Publication number | Publication date |
---|---|
CN109829503B (en) | 2021-12-17 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109657533A (en) | Pedestrian recognition methods and Related product again | |
CN111667066B (en) | Training method and device of network model, character recognition method and device and electronic equipment | |
CN104809452A (en) | Fingerprint identification method | |
CN104809464A (en) | Fingerprint information processing method | |
CN108229419A (en) | For clustering the method and apparatus of image | |
CN104809453A (en) | Authentication method based on fingerprints | |
CN106164932B (en) | The method and apparatus of music symbol for identification | |
CN112560753B (en) | Face recognition method, device, equipment and storage medium based on feature fusion | |
US11275970B2 (en) | Systems and methods for distributed data analytics | |
CN109154975A (en) | For generating the device and method of local binary patterns LBP | |
CN110362832B (en) | Paragraph merging method and device, storage medium and electronic equipment | |
US20240054802A1 (en) | System and method for spatial encoding and feature generators for enhancing information extraction | |
CN116681957B (en) | Image recognition method based on artificial intelligence and computer equipment | |
JP2022088602A (en) | Table generation method, device, electronic apparatus, storage medium and program | |
JP2022540101A (en) | POSITIONING METHOD AND APPARATUS, ELECTRONIC DEVICE, COMPUTER-READABLE STORAGE MEDIUM | |
CN111382791B (en) | Deep learning task processing method, image recognition task processing method and device | |
CN111639667B (en) | Image recognition method, device, electronic equipment and computer readable storage medium | |
CN109299692A (en) | A kind of personal identification method, computer readable storage medium and terminal device | |
CN110728172B (en) | Point cloud-based face key point detection method, device and system and storage medium | |
CN109829503A (en) | A kind of intensive frightened picture method of discrimination, system, equipment and its storage medium | |
CN113836297B (en) | Training method and device for text emotion analysis model | |
CN112507912B (en) | Method and device for identifying illegal pictures | |
CN107909091A (en) | A kind of iconic memory degree Forecasting Methodology based on sparse low-rank regression model | |
CN115131240A (en) | Target identification method and system for three-dimensional point cloud data | |
CN111931148A (en) | Image processing method and device and electronic equipment |
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 | ||
CB02 | Change of applicant information | ||
CB02 | Change of applicant information |
Address after: 361000 Area 1F-D1, Huaxun Building A, Software Park, Xiamen Torch High-tech Zone, Xiamen City, Fujian Province Applicant after: Xiamen Meishao Co., Ltd. Address before: Unit G03, Room 102, 22 Guanri Road, Phase II, Xiamen Software Park, Fujian Province Applicant before: XIAMEN MEIYOU INFORMATION SCIENCE & TECHNOLOGY Co.,Ltd. |
|
GR01 | Patent grant | ||
GR01 | Patent grant |