CN109086742A - scene recognition method, scene recognition device and mobile terminal - Google Patents
scene recognition method, scene recognition device and mobile terminal Download PDFInfo
- Publication number
- CN109086742A CN109086742A CN201810981469.6A CN201810981469A CN109086742A CN 109086742 A CN109086742 A CN 109086742A CN 201810981469 A CN201810981469 A CN 201810981469A CN 109086742 A CN109086742 A CN 109086742A
- Authority
- CN
- China
- Prior art keywords
- scene
- image
- processed
- tag
- scene recognition
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The application belongs to scene Recognition technical field, provides a kind of scene recognition method, scene Recognition device, mobile terminal and computer readable storage medium, comprising: obtain image to be processed；Scene Recognition is carried out to the image to be processed, obtains the scene tag matrix comprising multiple scene tags, wherein the scene tag indicates scene type；According to the scene tag matrix, the scene type of the image to be processed is determined.When can solve the prior art to image progress scene Recognition by the application, the problem of being easily lost information in image, reduce scene Recognition rate.
Description
Technical field
The application belongs to scene Recognition technical field more particularly to a kind of scene recognition method, scene Recognition device, movement
Terminal and computer readable storage medium.
Background technique
Scene Recognition refers to that the scene gone out in image according to content recognition similar in scene image, scene Recognition are images
One basic preprocessing process of process field.Currently, usually using the scene detection model based on deep learning to image
Carry out scene Recognition, deep learning be machine learning research in a new field, motivation be establish, simulation human brain into
The neural network of row analytic learning, it imitates the mechanism of human brain to explain data, such as image, sound and text etc..It is using
When scene detection model carries out scene Recognition to image, need first to carry out image to be cut to fixed dimension, then will be after cutting
Image is input to scene detection model, and scene detection model exports a scene tag.However, no matter using which kind of mode to figure
As being cut, information in image can be lost, scene Recognition rate is reduced.
Summary of the invention
It can in view of this, this application provides a kind of scene recognition method, scene Recognition device, mobile terminal and computers
Storage medium is read, when carrying out scene Recognition to image to solve the prior art, information in image is easily lost, reduces scene Recognition
The problem of rate.
The first aspect of the application provides a kind of scene recognition method, comprising:
Obtain image to be processed；
Scene Recognition is carried out to the image to be processed, obtains the scene tag matrix comprising multiple scene tags, wherein
The scene tag indicates scene type；
According to the scene tag matrix, the scene type of the image to be processed is determined.
The second aspect of the application provides a kind of scene Recognition device, comprising:
Image collection module, for obtaining image to be processed；
Scene Recognition module is obtained for carrying out scene Recognition to the image to be processed comprising multiple scene tags
Scene tag matrix, wherein the scene tag indicates scene type；
Scene determining module, for determining the scene type of the image to be processed according to the scene tag matrix.
The third aspect of the application provides a kind of mobile terminal, including memory, processor and is stored in described deposit
In reservoir and the computer program that can run on the processor, the processor are realized such as when executing the computer program
The step of scene recognition method.
The fourth aspect of the application provides a kind of computer readable storage medium, and the computer readable storage medium is deposited
Computer program is contained, is realized when the computer program is executed by processor such as the step of the scene recognition method.
The 5th aspect of the application provides a kind of computer program product, and the computer program product includes computer
Program realizes the scene recognition method as described in above-mentioned first aspect when the computer program is executed by one or more processors
The step of.
Therefore application scheme first obtains image to be processed, carries out scene Recognition to image to be processed, is included
The scene tag matrix of multiple scene tags, the scene type of image to be processed is determined further according to the scene tag matrix of acquisition.
Application scheme carries out scene Recognition to the image to be processed not cut, determines figure to be processed according to multiple scene tags of acquisition
The scene type of picture can retain information in image to be processed as far as possible, scene Recognition rate be improved, to solve the prior art
When carrying out scene Recognition to image, the problem of being easily lost information in image, reduce scene Recognition rate.
Detailed description of the invention
It in order to more clearly explain the technical solutions in the embodiments of the present application, below will be to embodiment or description of the prior art
Needed in attached drawing be briefly described, it should be apparent that, the accompanying drawings in the following description is only some of the application
Embodiment for those of ordinary skill in the art without any creative labor, can also be according to these
Attached drawing obtains other attached drawings.
Fig. 1 is the implementation process schematic diagram for the scene recognition method that the embodiment of the present application one provides；
Fig. 2 is the implementation process schematic diagram for the scene recognition method that the embodiment of the present application two provides；
Fig. 3 is the implementation process schematic diagram for the scene recognition method that the embodiment of the present application three provides；
Fig. 4 is the schematic diagram for the scene Recognition device that the embodiment of the present application four provides；
Fig. 5 is the schematic diagram for the mobile terminal that the embodiment of the present application five provides；
Fig. 6 is the schematic diagram for the mobile terminal that the embodiment of the present application six provides.
Specific embodiment
In being described below, for illustration and not for limitation, the tool of such as particular system structure, technology etc is proposed
Body details, so as to provide a thorough understanding of the present application embodiment.However, it will be clear to one skilled in the art that there is no these specific
The application also may be implemented in the other embodiments of details.In other situations, it omits to well-known system, device, electricity
The detailed description of road and method, so as not to obscure the description of the present application with unnecessary details.
It should be appreciated that ought use in this specification and in the appended claims, term " includes " instruction is described special
Sign, entirety, step, operation, the presence of element and/or component, but be not precluded one or more of the other feature, entirety, step,
Operation, the presence or addition of element, component and/or its set.
It is also understood that mesh of the term used in this present specification merely for the sake of description specific embodiment
And be not intended to limit the application.As present specification and it is used in the attached claims, unless on
Other situations are hereafter clearly indicated, otherwise " one " of singular, "one" and "the" are intended to include plural form.
As used in this specification and in the appended claims, term " if " can be according to context quilt
Be construed to " when ... " or " once " or " in response to determination " or " in response to detecting ".
In the specific implementation, mobile terminal described in the embodiment of the present application is including but not limited to such as with the sensitive table of touch
Mobile phone, laptop computer or the tablet computer in face (for example, touch-screen display and/or touch tablet) etc it is other
Portable device.It is to be further understood that in certain embodiments, the equipment is not portable communication device, but is had
The desktop computer of touch sensitive surface (for example, touch-screen display and/or touch tablet).
In following discussion, the mobile terminal including display and touch sensitive surface is described.However, should manage
Solution, mobile terminal may include that one or more of the other physical User of such as physical keyboard, mouse and/or control-rod connects
Jaws equipment.
Mobile terminal supports various application programs, such as one of the following or multiple: drawing application program, demonstration application
Program, word-processing application, website creation application program, disk imprinting application program, spreadsheet applications, game are answered
With program, telephony application, videoconference application, email application, instant messaging applications, forging
Refining supports application program, photo management application program, digital camera application program, digital camera application program, web-browsing to answer
With program, digital music player application and/or video frequency player application program.
At least one of such as touch sensitive surface can be used in the various application programs that can be executed on mobile terminals
Public physical user-interface device.It can be adjusted among applications and/or in corresponding application programs and/or change touch is quick
Feel the corresponding information shown in the one or more functions and terminal on surface.In this way, terminal public physical structure (for example,
Touch sensitive surface) it can support the various application programs with user interface intuitive and transparent for a user.
It should be understood that the size of the serial number of each step is not meant that the order of the execution order in the present embodiment, each process
Execution sequence should be determined by its function and internal logic, and the implementation process without coping with the embodiment of the present application constitutes any restriction.
In order to illustrate technical solution described herein, the following is a description of specific embodiments.
It is the implementation process schematic diagram for the scene recognition method that the embodiment of the present application one provides, the scene Recognition referring to Fig. 1
Method is applied to mobile terminal, and the scene recognition method as shown in the figure may comprise steps of:
Step S101 obtains image to be processed.
In the embodiment of the present application, the image to be processed can be user and pass through photo captured by this ground camera, example
Such as, user starts the camera application program in mobile terminal, utilizes photo captured by camera application program；Alternatively, can be
User is by the received picture of other applications, for example, transmitted by other wechats contact person that user receives in wechat
Picture；Alternatively, can be the picture that user downloads from internet；Alternatively, being also possible to a certain frame image in video, example
Such as, the wherein frame image in cartoon or TV play that user is watched；Alternatively, can also be that mobile terminal starts camera
Preview mode after camera current preview picture.The source of image to be processed is not construed as limiting herein.
It should be noted that before obtaining image to be processed, user be can choose whether to need directly to not cutting
Image to be processed carries out scene Recognition, when user selects "Yes" or do not select, after getting image to be processed, directly
Scene Recognition is carried out to the image to be processed not cut, first image to be processed can be cut when user selects "No",
Image to be processed after cutting is input to scene detection model, exports a scene tag.
Step S102 carries out scene Recognition to the image to be processed, obtains the scene tag comprising multiple scene tags
Matrix.
Wherein, the scene tag indicates that scene type, the scene type can refer to scene generic, the scene
It can refer to the background (such as night scene, indoor scene, backlight scene, snowfield scene, sandy beach scene etc.) in image, be also possible to
Refer to the main body (such as baby, animal, snow scenes etc.) in image, user can the specific scene class of sets itself according to actual needs
Not, it is not limited thereto.
In the embodiment of the present application, after getting image to be processed, in order to retain the figure to be processed as far as possible
Information as in, can not cut the image to be processed, but carry out scene knowledge to the image to be processed not cut
Not, acquisition includes the scene tag matrix (such as scene tag of 3x3) of multiple scene tags, rather than single scene tag
(i.e. the scene tag of 1x1).The multiple scene tag is the possible scene type of image to be processed, i.e., described to be processed
The scene type of image may be the scene type that some scene tag indicates in the multiple scene tag.
Step S103 determines the scene type of the image to be processed according to the scene tag matrix.
In the embodiment of the present application, it after obtaining the scene tag matrix comprising multiple scene tags, can use more
The method of number voting determines the scene type of the image to be processed, such as scene tag matrix is 3x3 matrix, i.e. nine scenes
Label, if there are five be above indoor scene in nine scene tags, it is determined that the scene type of the image to be processed
For indoor scene.
The embodiment of the present application carries out scene Recognition to the image to be processed not cut, true according to multiple scene tags of acquisition
The scene type of fixed image to be processed, can retain information in image to be processed as far as possible, scene Recognition rate be improved, to solve
When certainly the prior art carries out scene Recognition to image, the problem of being easily lost information in image, reduce scene Recognition rate.
It referring to fig. 2, is the implementation process schematic diagram for the scene recognition method that the embodiment of the present application two provides, the scene Recognition
Method is applied to mobile terminal, and the scene recognition method as shown in the figure may comprise steps of:
Step S201 obtains image to be processed.
The step is identical as step S101, and for details, reference can be made to the associated descriptions of step S101, and details are not described herein.
The image to be processed is input to trained convolutional neural networks, obtains the convolutional Neural by step S202
The scene tag matrix comprising multiple scene tags of the last one convolutional layer output of network.
Wherein, the scene tag indicates scene type.The scene type can refer to scene generic, the scene
It can refer to the background (such as night scene, indoor scene, backlight scene, snowfield scene, sandy beach scene etc.) in image, be also possible to
Refer to the main body (such as baby, animal, text etc.) in image, user can the specific scene class of sets itself according to actual needs
Not, it is not limited thereto.
In the embodiment of the present application, after getting image to be processed, in order to retain the figure to be processed as far as possible
Information as in, can not cut the image to be processed, but the image to be processed not cut is directly defeated
Enter to control the convolutional neural networks to trained convolutional neural networks and no longer exports single scene tag (the i.e. scene of 1x1
Label), but a label matrix is exported, which includes multiple scene tags, such as the scene tag of 3x3.
In the embodiment of the present application, the convolutional neural networks are used to carry out scene Recognition to the image to be processed, defeated
It out include the scene tag matrix of multiple scene tags, the multiple scene tag is the possible scene class of the image to be processed
Not, i.e., the scene type of the described image to be processed may be the scene class that some scene tag indicates in the multiple scene tag
Not.
Wherein, the convolutional neural networks can train in advance before mobile terminal factory, be solidificated in mobile terminal
In；Also mobile terminal can be used, and captured photo is as training set within a preset period of time, to convolutional neural networks progress
Personalization training, the scene type that the convolutional neural networks of different mobile terminal are accurately identified are different.For example, user
A often takes pictures indoors, then the convolutional neural networks in mobile terminal used in user A are accurate to the identification of indoor scene
Rate is relatively high；User B is often in shooting night scene, then the convolutional neural networks in mobile terminal used in user B are to night scene
Recognition accuracy it is relatively high.
Convolutional neural networks usually consist of two parts, respectively convolutional layer and full articulamentum, and since full articulamentum needs
The dimension of input vector is fixed, it is therefore desirable to the image for inputting convolutional neural networks is fixed size, such as
224x224, it usually needs the image of input convolutional neural networks is cut, fixed size is cut to, is easily lost image
In information.Since convolutional layer does not have size to limit image to be processed, then in order to avoid the information in image is lost, this
Apply embodiment can by output of the output as the convolutional neural networks of the last one convolutional layer of convolutional neural networks,
Without using the full articulamentum in convolutional neural networks, due to not using full articulamentum, then convolutional neural networks can input
The image of arbitrary dimension improves scene Recognition rate to retain the information in image.
Step S203 determines the scene type of the image to be processed according to the scene tag matrix.
The step is identical as step S103, and for details, reference can be made to the associated descriptions of step S103, and details are not described herein.
The embodiment of the present application is by the output of the last one convolutional layer of convolutional neural networks as the convolutional neural networks
Output can not limit the size of the image of input convolutional neural networks, so as to which the image to be processed not cut is directly defeated
Enter to trained convolutional neural networks, remain information in image to be processed as far as possible, improves scene Recognition rate.
It is the implementation process schematic diagram for the scene recognition method that the embodiment of the present application three provides, the scene Recognition referring to Fig. 3
Method is applied to mobile terminal, and the scene recognition method as shown in the figure may comprise steps of:
Step S301 starts the preview mode of camera.
In the embodiment of the present application, the preview mode, which can be, refers to the mould that preview camera currently wants shooting picture
Formula.Mobile terminal, into preview mode, can show the current preview picture of camera, so that user is true after starting camera
Determine whether current picture is the picture to be shot, when receiving photographing instruction, closes the preview mode of camera, camera is worked as
Preceding preview screen is taken pictures.
Step S302 obtains current preview picture under the preview mode, using the current preview picture as to
Handle image.
In the embodiment of the present application, it after the preview mode for starting camera, can obtain in real time under the preview mode
Current preview picture, using the current preview screen as image to be processed；It can also first obtain what mobile terminal was remain stationary
Duration when the duration is more than preset duration (such as 3 seconds), then obtains the current preview picture under the preview mode, will
The current preview screen is as image to be processed, because user is in just starting camera, the current preview picture of camera may be simultaneously
It is not meant to the picture of shooting, user may need mobile camera to be adjusted, then can keep by obtaining mobile terminal
Whether static duration judges the camera in movement.
The image to be processed is input to trained convolutional neural networks, obtains the convolutional Neural by step S303
The scene tag matrix comprising multiple scene tags of the last one convolutional layer output of network.
The step is identical as step S202, and for details, reference can be made to the associated descriptions of step S202, and details are not described herein.
Step S304 counts the scene tag quantity that Same Scene classification is indicated in the scene tag matrix.
Step S305, using the most scene type of scene tag quantity as the scene type of the image to be processed.
In the embodiment of the present application, since scene tag indicates scene type, then getting convolutional neural networks
After the scene tag matrix comprising multiple scene tags of the last one convolutional layer output, the scene tag can be first obtained
The scene type of each scene tag instruction, counts the scene that Same Scene classification is indicated in the scene tag matrix in matrix
Number of labels, using the most scene type of scene tag quantity as the scene type of image to be processed.It can also be by scene mark
Signing quantity is more than scene type of the scene type of preset threshold as image to be processed, is not deposited when in the scene tag matrix
When scene number of labels is more than the scene type of preset threshold, multiple scene tags in the scene tag matrix can be shown
The different scenes classification of instruction, using the scene type selected from the different scenes classification of display as the scene of image to be processed
Classification.Wherein, the preset threshold can refer to the total quantity of the multiple scene tagFor example, convolutional neural networks
The last one convolutional layer output include nine scene tags scene tag matrix, nine scene tags instruction scene class
It is not divided into snowfield scene, indoor scene, indoor scene, indoor scene, night scene, night scene, indoor scene, indoor scene, indoor field
Scape indicates that the scene tag quantity of snowfield scene is 1 then can count, indicates indoor scene in above-mentioned nine scene tags
Scene tag quantity be 6, indicate that the scene tag quantity of night scene is 2, since the scene tag quantity 6 of indoor scene is more than pre-
If threshold value 4.5, at this time it is determined that the scene type of image to be processed is indoor scene；What if nine scene tags indicated
Scene type is respectively snowfield scene, snowfield scene, snowfield scene, indoor scene, night scene, night scene, indoor scene, indoor field
Scape, indoor scene indicate that the scene tag quantity of snowfield scene is 3 then can count, refer in above-mentioned nine scene tags
The scene tag quantity for showing indoor scene is 4, indicates that the scene tag quantity of night scene is 2, due to the scene of above three scene
Number of labels is less than preset threshold 4.5, at this time can be in the current interface of mobile phone display snowfield, interior and three kinds of night scene
Scene type, for selection by the user.
Step S306 adjusts the parameter value of the camera according to the scene type of the image to be processed.
In the embodiment of the present application, due to shooting different scene types, the parameter value of required camera may also be different, that
Mobile terminal, can be according to the scene type adjust automatically of image to be processed after the scene type for determining image to be processed
The parameter value of camera in mobile terminal improves the regulated efficiency of camera parameter without manually adjusting the parameter value of camera.
It is taken pictures using parameter value adjusted to image to be processed, the quality of photo can be improved, because for image to be processed
Scene type for, camera parameter adjusted is usually optimal parameter value, can shoot the photo of high quality.It needs
It is noted that the parameter value of camera can be restored to adjustment after being taken pictures using parameter value adjusted
State is restored to initial parameter value, which can be the corresponding ginseng of scene type that mobile terminal is most often shot
Numerical value, can be to avoid the parameter value of frequent adjustment camera.Wherein, the parameter of the camera include but is not limited to exposure, sensitivity,
Aperture, white balance, focal length, Exposure Metering, flash lamp etc., user can modify the parameter of selected camera according to actual needs,
It is not limited thereto.Parameter value refers to the value of the parameter of above-mentioned camera.
Optionally, it before the scene type according to the image to be processed, the parameter value for adjusting the camera, also wraps
It includes:
Establish the corresponding relationship of M different scenes classification Yu N group different parameters value, wherein N is the integer greater than zero, and M is
Integer more than or equal to N；
Correspondingly, in the scene type according to the image to be processed, the parameter value for adjusting the camera includes:
From in the corresponding relationship of the M different scenes classification and N group different parameters value, search and the image to be processed
The corresponding one group of parameter value of scene type, using this group of parameter value as the parameter value of the camera.
In the embodiment of the present application, M different scenes classification pass corresponding with N group different parameters value can be pre-established
System, can be from the M different scenes classification ginseng different from N group pre-established after the scene type for determining image to be processed
In the corresponding relationship of numerical value, corresponding with the scene type of image to be processed parameter is searched, by the parameter found and camera
Initial parameter is compared, if the parameter found is identical as the initial parameter of camera, without being adjusted；If finding
Parameter and the initial parameter of camera be not identical, then the parameter of camera is adjusted to the parameter found by initial parameter.Wherein, often
Group parameter includes but is not limited to exposure, sensitivity, aperture, white balance, focal length, Exposure Metering, flash lamp etc..
The embodiment of the present application is using the current preview picture of camera as image to be processed, in the field for identifying image to be processed
After scape classification, it can be manually adjusted according to the parameter of the scene type adjust automatically camera of image to be processed without user,
The regulated efficiency of camera parameter is improved, and the photo of high quality can be shot.
It referring to fig. 4, is that the schematic diagram of scene Recognition device that the embodiment of the present application four provides only shows for ease of description
Part relevant to the embodiment of the present application is gone out.
The scene Recognition device includes:
Image collection module 41, for obtaining image to be processed；
Scene Recognition module 42, for carrying out scene Recognition to the image to be processed, obtaining includes multiple scene tags
Scene tag matrix, wherein the scene tag indicates scene type；
Scene determining module 43, for determining the scene type of the image to be processed according to the scene tag matrix.
Optionally, the scene Recognition module 42 is specifically used for:
The image to be processed is input to trained convolutional neural networks, obtains the convolutional neural networks output
Scene tag matrix comprising multiple scene tags.
Optionally, the scene Recognition module 42 is specifically used for:
Obtain the scene tag comprising multiple scene tags of the last one convolutional layer output of the convolutional neural networks
Matrix.
Optionally, the scene Recognition device further include:
Mode starting module 44, for starting the preview mode of camera；
Described image obtains module 41 and is specifically used for:
The current preview picture under the preview mode is obtained, using the current preview picture as the figure to be processed
Picture.
Optionally, the scene Recognition device further include:
Parameter adjustment module 45 adjusts the parameter value of the camera for the scene type according to the image to be processed.
Optionally, the scene Recognition device further include:
Relationship establishes module 46, for establishing the corresponding relationship of M different scenes classification Yu N group different parameters value, wherein
N is the integer greater than zero, and M is the integer more than or equal to N；
The parameter adjustment module 45 is specifically used for:
From in the corresponding relationship of the M different scenes classification and N group different parameters value, search and the image to be processed
The corresponding one group of parameter value of scene type, using this group of parameter value as the parameter value of the camera.
Optionally, the scene determining module 43 includes:
Statistic unit, for counting the scene tag quantity for indicating Same Scene classification in the scene tag matrix；
Determination unit, for using the most scene type of scene tag quantity as the scene class of the image to be processed
Not.
Scene Recognition device provided by the embodiments of the present application can be applied in preceding method embodiment one, embodiment two and reality
It applies in example three, details are referring to the description of above method embodiment one, embodiment two and embodiment three, and details are not described herein.
Fig. 5 is the schematic diagram for the mobile terminal that the embodiment of the present application five provides.The mobile terminal as shown in the figure can wrap
Include: one or more processors 501 (only show one) in figure；One or more input equipments 502 (one is only shown in figure),
One or more output equipments 503 (one is only shown in figure) and memory 504.It is above-mentioned processor 501, input equipment 502, defeated
Equipment 503 and memory 504 are connected by bus 505 out.Memory 504 for storing instruction, deposit for executing by processor 501
The instruction that reservoir 504 stores.Wherein:
The processor 501, for obtaining image to be processed；Scene Recognition is carried out to the image to be processed, is wrapped
Scene tag matrix containing multiple scene tags, wherein the scene tag indicates scene type；According to the scene tag square
Battle array, determines the scene type of the image to be processed.
Optionally, the processor 501 is specifically used for:
The image to be processed is input to trained convolutional neural networks, obtains the convolutional neural networks output
Scene tag matrix comprising multiple scene tags.
Optionally, the processor 501 is specifically used for:
Obtain the scene tag comprising multiple scene tags of the last one convolutional layer output of the convolutional neural networks
Matrix.
Optionally, before obtaining image to be processed, the processor 501 is also used to:
Start the preview mode of camera；
Optionally, the processor 501 is specifically used for:
The current preview picture under the preview mode is obtained, using the current preview picture as the figure to be processed
Picture.
Optionally, after the scene for determining the image to be processed, the processor 501 is also used to:
According to the scene type of the image to be processed, the parameter value of the camera is adjusted.
Optionally, before the scene type according to the image to be processed, the parameter value for adjusting the camera, the place
Reason device 501 is also used to:
Establish the corresponding relationship of M different scenes classification Yu N group different parameters value, wherein N is the integer greater than zero, and M is
Integer more than or equal to N；
Optionally, the processor 501 is specifically used for:
From in the corresponding relationship of the M different scenes classification and N group different parameters value, search and the image to be processed
The corresponding one group of parameter value of scene type, using this group of parameter value as the parameter value of the camera.
Optionally, the processor 501 is specifically used for:
Count the scene tag quantity that Same Scene classification is indicated in the scene tag matrix；
Using the most scene type of scene tag quantity as the scene type of the image to be processed.
It should be appreciated that in the embodiment of the present application, the processor 501 can be central processing unit (Central
Processing Unit, CPU), which can also be other general processors, digital signal processor (Digital
Signal Processor, DSP), specific integrated circuit (Application Specific Integrated Circuit,
ASIC), ready-made programmable gate array (Field-Programmable Gate Array, FPGA) or other programmable logic
Device, discrete gate or transistor logic, discrete hardware components etc..General processor can be microprocessor or this at
Reason device is also possible to any conventional processor etc..
Input equipment 502 may include that Trackpad, fingerprint adopt sensor (for acquiring the finger print information and fingerprint of user
Directional information), microphone, data receiver interface etc..Output equipment 503 may include display (LCD etc.), loudspeaker, data
Transmission interface etc..
The memory 504 may include read-only memory and random access memory, and to processor 501 provide instruction and
Data.The a part of of memory 504 can also include nonvolatile RAM.For example, memory 504 can also be deposited
Store up the information of device type.
In the specific implementation, processor 501, input equipment 502 described in the embodiment of the present application, 503 and of output equipment
Implementation described in the embodiment of scene recognition method provided by the embodiments of the present application can be performed in memory 504, can also
Implementation described in scene Recognition device described in example IV is executed, details are not described herein.
Fig. 6 is the schematic diagram for the mobile terminal that the embodiment of the present application six provides.As shown in fig. 6, the mobile end of the embodiment
End 6 includes: processor 60, memory 61 and is stored in the meter that can be run in the memory 61 and on the processor 60
Calculation machine program 62.The processor 60 is realized when executing the computer program 62 in above-mentioned each scene recognition method embodiment
The step of, such as step S101 to S103 shown in FIG. 1.Alternatively, reality when the processor 60 executes the computer program 62
The function of each module/unit in existing above-mentioned each Installation practice, such as the function of module 41 to 46 shown in Fig. 4.
Illustratively, the computer program 62 can be divided into one or more module/units, it is one or
Multiple module/units are stored in the memory 61, and are executed by the processor 60, to complete the application.Described one
A or multiple module/units can be the series of computation machine program instruction section that can complete specific function, which is used for
Implementation procedure of the computer program 62 in the mobile terminal 6 is described.For example, the computer program 62 can be divided
It is cut into image collection module, scene Recognition module, scene determining module, mode starting module, parameter adjustment module and relationship
Module is established, each module concrete function is as follows:
Image collection module, for obtaining image to be processed；
Label output module is obtained for carrying out scene Recognition to the image to be processed comprising multiple scene tags
Scene tag matrix, wherein the scene tag indicates scene type；
Scene determining module, for determining the scene type of the image to be processed according to the scene tag matrix.
Optionally, the scene Recognition module is specifically used for:
The image to be processed is input to trained convolutional neural networks, obtains the convolutional neural networks output
Scene tag matrix comprising multiple scene tags.
Optionally, the scene Recognition module is specifically used for:
Obtain the scene tag comprising multiple scene tags of the last one convolutional layer output of the convolutional neural networks
Matrix.
Mode starting module, for starting the preview mode of camera；
Described image obtains module and is specifically used for:
The current preview picture under the preview mode is obtained, using the current preview picture as the figure to be processed
Picture.
Parameter adjustment module adjusts the parameter value of the camera for the scene type according to the image to be processed.
Relationship establishes module, for establishing the corresponding relationship of M different scenes classification Yu N group different parameters value, wherein N
For the integer greater than zero, M is the integer more than or equal to N；
The parameter adjustment module is specifically used for:
From in the corresponding relationship of the M different scenes classification and N group different parameters value, search and the image to be processed
The corresponding one group of parameter value of scene type, using this group of parameter value as the parameter value of the camera.
Optionally, the scene determining module includes:
Statistic unit, for counting the scene tag quantity for indicating Same Scene classification in the scene tag matrix；
Determination unit, for using the most scene type of scene tag quantity as the scene class of the image to be processed
Not.
The mobile terminal 6 can be the calculating such as desktop PC, notebook, palm PC and cloud server and set
It is standby.The mobile terminal may include, but be not limited only to, processor 60, memory 61.It will be understood by those skilled in the art that Fig. 6
The only example of mobile terminal 6 does not constitute the restriction to mobile terminal 6, may include than illustrating more or fewer portions
Part perhaps combines certain components or different components, such as the mobile terminal can also include input-output equipment, net
Network access device, bus etc..
Alleged processor 60 can be central processing unit CPU, can also be other general processors, Digital Signal Processing
Device DSP, application-specific integrated circuit ASIC, ready-made programmable gate array FPGA or other programmable logic device, discrete gate or
Transistor logic, discrete hardware components etc..General processor can be microprocessor or the processor is also possible to appoint
What conventional processor etc..
The memory 61 can be the internal storage unit of the mobile terminal 6, such as the hard disk or interior of mobile terminal 6
It deposits.The memory 61 is also possible to the External memory equipment of the mobile terminal 6, such as be equipped on the mobile terminal 6
Plug-in type hard disk, intelligent memory card (Smart Media Card, SMC), secure digital (Secure Digital, SD) card dodge
Deposit card (Flash Card) etc..Further, the memory 61 can also both include the storage inside list of the mobile terminal 6
Member also includes External memory equipment.The memory 61 is for storing needed for the computer program and the mobile terminal
Other programs and data.The memory 61 can be also used for temporarily storing the data that has exported or will export.
It is apparent to those skilled in the art that for convenience of description and succinctly, only with above-mentioned each function
Can unit, module division progress for example, in practical application, can according to need and by above-mentioned function distribution by different
Functional unit, module are completed, i.e., the internal structure of described device is divided into different functional unit or module, more than completing
The all or part of function of description.Each functional unit in embodiment, module can integrate in one processing unit, can also
To be that each unit physically exists alone, can also be integrated in one unit with two or more units, it is above-mentioned integrated
Unit both can take the form of hardware realization, can also realize in the form of software functional units.In addition, each function list
Member, the specific name of module are also only for convenience of distinguishing each other, the protection scope being not intended to limit this application.Above system
The specific work process of middle unit, module, can refer to corresponding processes in the foregoing method embodiment, and details are not described herein.
In the above-described embodiments, it all emphasizes particularly on different fields to the description of each embodiment, is not described in detail or remembers in some embodiment
The part of load may refer to the associated description of other embodiments.
Those of ordinary skill in the art may be aware that list described in conjunction with the examples disclosed in the embodiments of the present disclosure
Member and algorithm steps can be realized with the combination of electronic hardware or computer software and electronic hardware.These functions are actually
It is implemented in hardware or software, the specific application and design constraint depending on technical solution.Professional technician
Each specific application can be used different methods to achieve the described function, but this realization is it is not considered that exceed
Scope of the present application.
In embodiment provided herein, it should be understood that disclosed device/mobile terminal and method, it can be with
It realizes by another way.For example, device described above/mobile terminal embodiment is only schematical, for example, institute
The division of module or unit is stated, only a kind of logical function partition, there may be another division manner in actual implementation, such as
Multiple units or components can be combined or can be integrated into another system, or some features can be ignored or not executed.Separately
A bit, shown or discussed mutual coupling or direct-coupling or communication connection can be through some interfaces, device
Or the INDIRECT COUPLING or communication connection of unit, it can be electrical property, mechanical or other forms.
The unit as illustrated by the separation member may or may not be physically separated, aobvious as unit
The component shown may or may not be physical unit, it can and it is in one place, or may be distributed over multiple
In network unit.It can select some or all of unit therein according to the actual needs to realize the mesh of this embodiment scheme
's.
It, can also be in addition, each functional unit in each embodiment of the application can integrate in one processing unit
It is that each unit physically exists alone, can also be integrated in one unit with two or more units.Above-mentioned integrated list
Member both can take the form of hardware realization, can also realize in the form of software functional units.
If the integrated module/unit be realized in the form of SFU software functional unit and as independent product sale or
In use, can store in a computer readable storage medium.Based on this understanding, the application realizes above-mentioned implementation
All or part of the process in example method, can also instruct relevant hardware to complete, the meter by computer program
Calculation machine program can be stored in a computer readable storage medium, the computer program when being executed by processor, it can be achieved that on
The step of stating each embodiment of the method.Wherein, the computer program includes computer program code, the computer program generation
Code can be source code form, object identification code form, executable file or certain intermediate forms etc..The computer-readable medium
It may include: any entity or device, recording medium, USB flash disk, mobile hard disk, magnetic that can carry the computer program code
Dish, CD, computer storage, read-only memory (ROM, Read-Only Memory), random access memory (RAM,
Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium etc..It should be noted that described
The content that computer-readable medium includes can carry out increasing appropriate according to the requirement made laws in jurisdiction with patent practice
Subtract, such as does not include electric carrier signal and electricity according to legislation and patent practice, computer-readable medium in certain jurisdictions
Believe signal.
Embodiment described above is only to illustrate the technical solution of the application, rather than its limitations；Although referring to aforementioned reality
Example is applied the application is described in detail, those skilled in the art should understand that: it still can be to aforementioned each
Technical solution documented by embodiment is modified or equivalent replacement of some of the technical features；And these are modified
Or replacement, the spirit and scope of each embodiment technical solution of the application that it does not separate the essence of the corresponding technical solution should all
Comprising within the scope of protection of this application.
Claims (10)
1. a kind of scene recognition method characterized by comprising
Obtain image to be processed；
Scene Recognition is carried out to the image to be processed, obtains the scene tag matrix comprising multiple scene tags, wherein described
Scene tag indicates scene type；
According to the scene tag matrix, the scene type of the image to be processed is determined.
2. scene recognition method as described in claim 1, which is characterized in that described to carry out scene knowledge to the image to be processed
Not, obtaining the scene tag matrix comprising multiple scene tags includes:
The image to be processed is input to trained convolutional neural networks, obtain the convolutional neural networks output includes
The scene tag matrix of multiple scene tags.
3. scene recognition method as claimed in claim 2, which is characterized in that described to obtain what the convolutional neural networks exported
Scene tag matrix comprising multiple scene tags includes:
Obtain the scene tag matrix comprising multiple scene tags of the last one convolutional layer output of the convolutional neural networks.
4. scene recognition method as described in claim 1, which is characterized in that before obtaining image to be processed, further includes:
Start the preview mode of camera；
Correspondingly, obtaining image to be processed and including:
The current preview picture under the preview mode is obtained, using the current preview picture as the image to be processed.
5. scene recognition method as claimed in claim 4, which is characterized in that the scene for determining the image to be processed it
Afterwards, further includes:
According to the scene type of the image to be processed, the parameter value of the camera is adjusted.
6. scene recognition method as claimed in claim 5, which is characterized in that in the scene class according to the image to be processed
Not, before the parameter value for adjusting the camera, further includes:
Establish the corresponding relationship of M different scenes classification Yu N group different parameters value, wherein N is integer greater than zero, M be greater than
Or the integer equal to N；
Correspondingly, in the scene type according to the image to be processed, the parameter value for adjusting the camera includes:
From in the corresponding relationship of the M different scenes classification and N group different parameters value, the field with the image to be processed is searched
The corresponding one group of parameter value of scape classification, using this group of parameter value as the parameter value of the camera.
7. such as scene recognition method as claimed in any one of claims 1 to 6, which is characterized in that described according to the scene tag
Matrix determines that the scene type of the image to be processed includes:
Count the scene tag quantity that Same Scene classification is indicated in the scene tag matrix；
Using the most scene type of scene tag quantity as the scene type of the image to be processed.
8. a kind of scene Recognition device characterized by comprising
Image collection module, for obtaining image to be processed；
Scene Recognition module obtains the scene comprising multiple scene tags for carrying out scene Recognition to the image to be processed
Label matrix, wherein the scene tag indicates scene type；
Scene determining module, for determining the scene type of the image to be processed according to the scene tag matrix.
9. a kind of mobile terminal, including memory, processor and storage are in the memory and can be on the processor
The computer program of operation, which is characterized in that the processor realizes such as claim 1 to 7 when executing the computer program
The step of any one scene recognition method.
10. a kind of computer readable storage medium, the computer-readable recording medium storage has computer program, and feature exists
In the step of realization scene recognition method as described in any one of claim 1 to 7 when the computer program is executed by processor
Suddenly.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810981469.6A CN109086742A (en) | 2018-08-27 | 2018-08-27 | scene recognition method, scene recognition device and mobile terminal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810981469.6A CN109086742A (en) | 2018-08-27 | 2018-08-27 | scene recognition method, scene recognition device and mobile terminal |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109086742A true CN109086742A (en) | 2018-12-25 |
Family
ID=64794647
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810981469.6A Pending CN109086742A (en) | 2018-08-27 | 2018-08-27 | scene recognition method, scene recognition device and mobile terminal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109086742A (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109919116A (en) * | 2019-03-14 | 2019-06-21 | Oppo广东移动通信有限公司 | Scene recognition method, device, electronic equipment and storage medium |
| CN109995999A (en) * | 2019-03-14 | 2019-07-09 | Oppo广东移动通信有限公司 | Scene recognition method, device, electronic equipment and storage medium |
| CN110503099A (en) * | 2019-07-23 | 2019-11-26 | 平安科技（深圳）有限公司 | Information identifying method and relevant device based on deep learning |
| CN110516590A (en) * | 2019-08-26 | 2019-11-29 | 国网河北省电力有限公司保定供电分公司 | Operation or work standard prompt system based on scene Recognition |
| CN111131889A (en) * | 2019-12-31 | 2020-05-08 | 深圳创维－Rgb电子有限公司 | Method and system for adaptively adjusting images and sounds in scene and readable storage medium |
| CN111723606A (en) * | 2019-03-19 | 2020-09-29 | 北京搜狗科技发展有限公司 | Data processing method and device and data processing device |
| CN111814812A (en) * | 2019-04-09 | 2020-10-23 | Oppo广东移动通信有限公司 | Modeling method, modeling device, storage medium, electronic device and scene recognition method |
| CN111832491A (en) * | 2020-07-16 | 2020-10-27 | Oppo广东移动通信有限公司 | Text detection method and device and processing equipment |
| CN113435499A (en) * | 2021-06-25 | 2021-09-24 | 平安科技（深圳）有限公司 | Label classification method and device, electronic equipment and storage medium |
| CN113673275A (en) * | 2020-05-13 | 2021-11-19 | 北京达佳互联信息技术有限公司 | Indoor scene layout estimation method and device, electronic equipment and storage medium |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105184312A (en) * | 2015-08-24 | 2015-12-23 | 中国科学院自动化研究所 | Character detection method and device based on deep learning |
| CN105631426A (en) * | 2015-12-29 | 2016-06-01 | 中国科学院深圳先进技术研究院 | Image text detection method and device |
| CN105933607A (en) * | 2016-05-26 | 2016-09-07 | 维沃移动通信有限公司 | Photographing effect adjusting method of mobile terminal and mobile terminal |
| CN106228158A (en) * | 2016-07-25 | 2016-12-14 | 北京小米移动软件有限公司 | The method and apparatus of picture detection |
| CN106357983A (en) * | 2016-11-15 | 2017-01-25 | 上海传英信息技术有限公司 | Photographing parameter adjustment method and user terminal |
| CN107145904A (en) * | 2017-04-28 | 2017-09-08 | 北京小米移动软件有限公司 | Determination method, device and the storage medium of image category |
| CN107194318A (en) * | 2017-04-24 | 2017-09-22 | 北京航空航天大学 | The scene recognition method of target detection auxiliary |
| WO2017192194A2 (en) * | 2016-02-09 | 2017-11-09 | Hrl Laboratories, Llc | System and method for the fusion of bottom-up whole-image features and top-down entity classification for accurate image/video scene classification |
| CN107577983A (en) * | 2017-07-11 | 2018-01-12 | 中山大学 | It is a kind of to circulate the method for finding region-of-interest identification multi-tag image |
| CN107679580A (en) * | 2017-10-21 | 2018-02-09 | 桂林电子科技大学 | A kind of isomery shift image feeling polarities analysis method based on the potential association of multi-modal depth |
| CN107886064A (en) * | 2017-11-06 | 2018-04-06 | 安徽大学 | A kind of method that recognition of face scene based on convolutional neural networks adapts to |
| CN108133233A (en) * | 2017-12-18 | 2018-06-08 | 中山大学 | A kind of multi-tag image-recognizing method and device |
-
2018
- 2018-08-27 CN CN201810981469.6A patent/CN109086742A/en active Pending
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105184312A (en) * | 2015-08-24 | 2015-12-23 | 中国科学院自动化研究所 | Character detection method and device based on deep learning |
| CN105631426A (en) * | 2015-12-29 | 2016-06-01 | 中国科学院深圳先进技术研究院 | Image text detection method and device |
| WO2017192194A2 (en) * | 2016-02-09 | 2017-11-09 | Hrl Laboratories, Llc | System and method for the fusion of bottom-up whole-image features and top-down entity classification for accurate image/video scene classification |
| CN105933607A (en) * | 2016-05-26 | 2016-09-07 | 维沃移动通信有限公司 | Photographing effect adjusting method of mobile terminal and mobile terminal |
| CN106228158A (en) * | 2016-07-25 | 2016-12-14 | 北京小米移动软件有限公司 | The method and apparatus of picture detection |
| CN106357983A (en) * | 2016-11-15 | 2017-01-25 | 上海传英信息技术有限公司 | Photographing parameter adjustment method and user terminal |
| CN107194318A (en) * | 2017-04-24 | 2017-09-22 | 北京航空航天大学 | The scene recognition method of target detection auxiliary |
| CN107145904A (en) * | 2017-04-28 | 2017-09-08 | 北京小米移动软件有限公司 | Determination method, device and the storage medium of image category |
| CN107577983A (en) * | 2017-07-11 | 2018-01-12 | 中山大学 | It is a kind of to circulate the method for finding region-of-interest identification multi-tag image |
| CN107679580A (en) * | 2017-10-21 | 2018-02-09 | 桂林电子科技大学 | A kind of isomery shift image feeling polarities analysis method based on the potential association of multi-modal depth |
| CN107886064A (en) * | 2017-11-06 | 2018-04-06 | 安徽大学 | A kind of method that recognition of face scene based on convolutional neural networks adapts to |
| CN108133233A (en) * | 2017-12-18 | 2018-06-08 | 中山大学 | A kind of multi-tag image-recognizing method and device |
Non-Patent Citations (2)
| Title |
|---|
| 司马海峰 等: "《遥感图像分类中的智能计算方法》", 31 January 2018, 吉林大学出版社 * |
| 陈慧岩: "CNN的改进", 《智能车辆理论与应用》 * |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109919116B (en) * | 2019-03-14 | 2022-05-17 | Oppo广东移动通信有限公司 | Scene recognition method and device, electronic equipment and storage medium |
| CN109995999A (en) * | 2019-03-14 | 2019-07-09 | Oppo广东移动通信有限公司 | Scene recognition method, device, electronic equipment and storage medium |
| CN109919116A (en) * | 2019-03-14 | 2019-06-21 | Oppo广东移动通信有限公司 | Scene recognition method, device, electronic equipment and storage medium |
| CN111723606A (en) * | 2019-03-19 | 2020-09-29 | 北京搜狗科技发展有限公司 | Data processing method and device and data processing device |
| CN111814812A (en) * | 2019-04-09 | 2020-10-23 | Oppo广东移动通信有限公司 | Modeling method, modeling device, storage medium, electronic device and scene recognition method |
| CN110503099A (en) * | 2019-07-23 | 2019-11-26 | 平安科技（深圳）有限公司 | Information identifying method and relevant device based on deep learning |
| CN110503099B (en) * | 2019-07-23 | 2023-06-20 | 平安科技（深圳）有限公司 | Information identification method based on deep learning and related equipment |
| CN110516590A (en) * | 2019-08-26 | 2019-11-29 | 国网河北省电力有限公司保定供电分公司 | Operation or work standard prompt system based on scene Recognition |
| CN111131889A (en) * | 2019-12-31 | 2020-05-08 | 深圳创维－Rgb电子有限公司 | Method and system for adaptively adjusting images and sounds in scene and readable storage medium |
| CN113673275A (en) * | 2020-05-13 | 2021-11-19 | 北京达佳互联信息技术有限公司 | Indoor scene layout estimation method and device, electronic equipment and storage medium |
| CN113673275B (en) * | 2020-05-13 | 2024-02-20 | 北京达佳互联信息技术有限公司 | Indoor scene layout estimation method and device, electronic equipment and storage medium |
| CN111832491A (en) * | 2020-07-16 | 2020-10-27 | Oppo广东移动通信有限公司 | Text detection method and device and processing equipment |
| CN111832491B (en) * | 2020-07-16 | 2024-08-09 | Oppo广东移动通信有限公司 | Text detection method, device and processing equipment |
| CN113435499A (en) * | 2021-06-25 | 2021-09-24 | 平安科技（深圳）有限公司 | Label classification method and device, electronic equipment and storage medium |
| CN113435499B (en) * | 2021-06-25 | 2023-06-20 | 平安科技（深圳）有限公司 | Label classification method, device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109086742A (en) | scene recognition method, scene recognition device and mobile terminal | |
| CN110585726B (en) | User recall method, device, server and computer readable storage medium | |
| CN108961157B (en) | Picture processing method, picture processing device and terminal equipment | |
| CN110458360B (en) | Method, device, equipment and storage medium for predicting hot resources | |
| US11158057B2 (en) | Device, method, and graphical user interface for processing document | |
| CN108921806A (en) | A kind of image processing method, image processing apparatus and terminal device | |
| CN109284445B (en) | Network resource recommendation method and device, server and storage medium | |
| CN110246110B (en) | Image evaluation method, device and storage medium | |
| CN112069414A (en) | Recommendation model training method and device, computer equipment and storage medium | |
| CN109101931A (en) | A kind of scene recognition method, scene Recognition device and terminal device | |
| CN109784351B (en) | Behavior data classification method and device and classification model training method and device | |
| CN108898082B (en) | Picture processing method, picture processing device and terminal equipment | |
| CN108961183B (en) | Image processing method, terminal device and computer-readable storage medium | |
| CN108961267B (en) | Picture processing method, picture processing device and terminal equipment | |
| CN108737739A (en) | A kind of preview screen acquisition method, preview screen harvester and electronic equipment | |
| CN111506758A (en) | Method and device for determining article name, computer equipment and storage medium | |
| CN111984803B (en) | Multimedia resource processing method and device, computer equipment and storage medium | |
| CN108874134A (en) | Eyeshield mode treatment method, mobile terminal and computer readable storage medium | |
| CN110555102A (en) | media title recognition method, device and storage medium | |
| CN112464052A (en) | Feedback information processing method, feedback information display device and electronic equipment | |
| CN114154068A (en) | Media content recommendation method and device, electronic equipment and storage medium | |
| CN110929159A (en) | Resource delivery method, device, equipment and medium | |
| CN108985215B (en) | Picture processing method, picture processing device and terminal equipment | |
| CN108932703B (en) | Picture processing method, picture processing device and terminal equipment | |
| CN111931075B (en) | Content recommendation method and device, computer equipment and storage medium |
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: 20181225 |
|
| RJ01 | Rejection of invention patent application after publication |