CN105915801A - Self-learning method and device capable of improving snap shot effect - Google Patents
Self-learning method and device capable of improving snap shot effect Download PDFInfo
- Publication number
- CN105915801A CN105915801A CN201610407762.2A CN201610407762A CN105915801A CN 105915801 A CN105915801 A CN 105915801A CN 201610407762 A CN201610407762 A CN 201610407762A CN 105915801 A CN105915801 A CN 105915801A
- Authority
- CN
- China
- Prior art keywords
- shooting condition
- condition parameter
- shooting
- parameter
- robot
- 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
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/60—Control of cameras or camera modules
- H04N23/64—Computer-aided capture of images, e.g. transfer from script file into camera, check of taken image quality, advice or proposal for image composition or decision on when to take image
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/60—Control of cameras or camera modules
- H04N23/617—Upgrading or updating of programs or applications for camera control
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/60—Control of cameras or camera modules
- H04N23/66—Remote control of cameras or camera parts, e.g. by remote control devices
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
Abstract
The invention provides a self-learning method capable of improving a snap shot effect. The self-learning method comprises the following steps: turning on a shooting function of an intelligent robot; executing a shooting operation based on a preset shooting condition parameter; acquiring preference data and a picture evaluation parameter specific to picture composition; performing machine learning on the picture evaluation parameter and the preference data specific to picture information, and modifying the preset shooting condition parameter based on a learning process in order to obtain an optimal shooting condition parameter for self-learning of the robot; and saving the optimal shooting condition parameter instead of the preset shooting condition parameter in order to perform a next shooting operation. Through adoption of the method, the storage space can be saved without manually deleting undesirable photos specially. Moreover, the intelligent robot can automatically delete undesirable pictures taken by a user or modifying the shooting parameter in a passive shooting scene.
Description
Technical field
The present invention relates to Information Technology Agreement field, specifically, relate to a kind of self study improving and capturing effect
Method and device.
Background technology
Either for amusement design, or for safety monitoring, if machine can be captured automatically, then
Photographer's hard work will be replaced, and the situation that some important moment has little time to shoot also can be kept away
Exempt from.At present, what the intelligent robot developed had have captures function automatically, can partly solve
The problems referred to above.There is the robot of candid photograph function automatically under screening-mode, the movement of meeting automatically track target thing
Maintain it in the indicia framing scope of shooting picture, and special according to the such as specific face of shooting trigger condition
Levy, facial emotions, action or special sound etc. trigger shooting operation.
But, in prior art, robot is based only upon above-mentioned shooting trigger condition and automatically starts shooting, and to picture
Quality or shooting effect do not screen, thus cause without selectively shooting.The a large amount of pictures so taken are used
Family may be dissatisfied for the effect of picture.These unsatisfied picture not only committed memories, but also can consume
Their time is deleted at expense family.
Therefore, under the scene automatically captured of object manipulator, need a kind of can improve capture effect from
Learning method and device thereof.
Summary of the invention
It is an object of the invention to solve the above-mentioned photo poor effect problem that oneself takes pictures of prior art, and oneself
Occur under scene of taking pictures that the bad picture of a large amount of effect takies memory space and time-consuming problem, it is provided that a kind of
The self-learning method improving candid photograph effect of object manipulator, said method comprising the steps of:
Open the shoot function of intelligent robot;
Shooting operation is performed based on default shooting condition parameter;
Obtain the preference data for composition and picture evaluating;
The picture evaluating made for described picture information and preference data are carried out machine learning, and base
Described default shooting condition parameter is revised, to obtain the optimum shooting condition of robot self study in learning process
Parameter;
Described optimum shooting condition parameter is preserved as the replacement of default shooting condition parameter, for
Shooting operation next time.
Picture evaluating described in the self-learning method improving candid photograph effect according to an embodiment of the invention
Including from the forward evaluation of user and negative sense evaluation, in the step that described default shooting condition parameter is modified
In Zhou, the picture information obtaining forward evaluation carried out analysis of image data and obtains the first correction value set, to obtaining
The picture information of negative sense evaluation carries out analysis of image data and obtains the second correction value set, wherein, by first
Correction value set and the second correction value set, as evaluating data, make shooting condition parameter convergence by machine learning
In the shooting condition parameter more meeting user's specification.
The self-learning method improving candid photograph effect according to an embodiment of the invention, described intelligent robot exists
Under shooting trigger command, opening the shoot function of intelligent robot, automatically perform to capture operation, shooting triggers life
Make and producing based on the shooting condition set.
The self-learning method improving candid photograph effect according to an embodiment of the invention, is opening intelligent robot
Shoot function step after, described intelligent robot according to user instruction passively perform capture function.
The self-learning method improving candid photograph effect according to an embodiment of the invention, passively opens in robot
In the case of capturing function, robot carries out Automatic sieve for the photo taken based on optimum shooting condition parameter
Choosing, deletes the picture information not meeting described optimum shooting condition parameter, preserves and meets described optimum shooting condition
The picture information of parameter./ this is a kind of implementation therein, in the range of being included in above technical scheme, not
Limitation and unique/
According to another aspect of the present invention, additionally provide a kind of self study device improving and capturing effect, described
Device includes:
Shoot function opens unit, and it is in order to open the shoot function of intelligent robot;
Shooting operation execution unit, it performs shooting operation based on default shooting condition parameter;
Parameter acquiring unit, it is for obtaining the preference data for composition and picture evaluating;
Shooting condition parameter modifying unit, it is in order to the picture evaluating made for described picture information
Carry out machine learning with preference data, and revise described default shooting condition parameter based on learning process, with
Optimum shooting condition parameter to robot self study;
Parameter storage unit, preserves described optimum shooting condition parameter, grasps for shooting next time
Make.
According to the present invention improve capture effect self study device in, described picture evaluating include from
The forward evaluation of user and negative sense evaluation, in described shooting condition parameter modifying unit, to obtaining forward evaluation
Picture information carry out analysis of image data and obtain the first correction value set, to obtaining the picture information that negative sense is evaluated
Carry out analysis of image data and obtain the second correction value set, wherein, by the first correction value set and the second correction value
Gather as evaluating data, make shooting condition parameter level off to more by machine learning to meet the shooting of user's specification
Conditional parameter.
According to the present invention improve capture effect self study device also include automatically capturing unit, its with so that
Described intelligent robot, under shooting trigger command, is opened the shoot function of intelligent robot, is automatically performed candid photograph
Operation, shooting trigger command produces based on the shooting condition set.
The self study device improving candid photograph effect according to the present invention also includes passively capturing unit, described intelligent machine
Device people passively opens candid photograph function according to user instruction.
According to the present invention improve capture effect self study device also include automatic screening unit, its with so that
Robot carries out automatic screening for the photo taken based on optimum shooting condition parameter, deletes described in not meeting
The picture information of optimum shooting condition parameter, preserves the picture information meeting described optimum shooting condition parameter.
The self study side improving candid photograph effect of the had the benefit that the application of the invention of the present invention
Method, picture can be evaluated data parameter according to conventional user and enter the preference of composition by intelligent robot
Row is analyzed and is drawn optimum shooting condition parameter required during automatic shooting, automatic under optimum shooting condition parameter
The effect of picture information captured is gratifying, thus having saved memory space, to be not required to labor intensive special
Door deletes undesirable photo.Additionally, the method according to the invention, intelligent robot can also passively shoot field
Automatically the picture having shot undesirable to user under scape is deleted or revises acquisition parameters.
Other features and advantages of the present invention will illustrate in the following description, and, partly from description
In become apparent, or by implement the present invention and understand.The purpose of the present invention and other advantages can be passed through
Structure specifically noted in description, claims and accompanying drawing realizes and obtains.
Accompanying drawing explanation
Accompanying drawing is for providing a further understanding of the present invention, and constitutes a part for description, with the present invention
Embodiment be provided commonly for explain the present invention, be not intended that limitation of the present invention.In the accompanying drawings:
Fig. 1 is the method flow diagram that the intelligent robot of prior art is captured automatically;
Fig. 2 is the self-learning method flow process automatically capturing effect for improving intelligent robot according to the present invention
Figure;
Fig. 3 is the self study automatically capturing effect according to one embodiment of the present of invention for improving intelligent robot
Method flow diagram;And
Fig. 4 is the structured flowchart improving the self study device capturing effect according to one embodiment of the present of invention.
Detailed description of the invention
For making the object, technical solutions and advantages of the present invention clearer, below in conjunction with accompanying drawing, the present invention is implemented
Example is described in further detail.
As it is shown in figure 1, which show intelligent robot in prior art the most automatically capture the side of operation
Method flow chart.In the method flow process, the audition detection function going back detail display intelligent robot is examined with vision
The automatic conversion of brake.
In order to automatically capture, it be first turned on the vision-based detection function of intelligent robot, and object is carried out
Detection and location.In prior art, the step of detection and location object generally includes: determined by vision-based detection
Object is in coverage;If it is determined that object is in outside coverage, then the sound that detection object sends
Sound, utilizes sound detection to position object;If it is determined that outside object is in coverage and object detected
Do not send sound, then utilize vision-based detection to carry out Scan orientation object.
Specifically, vision-based detection can first be started in step S1011.In judging shooting picture afterwards
Whether there is the existence of object, such as step S1012.If having the existence of object, then auto-steering such as Fig. 1
Shown tracking target step.
But, without the existence of object, then robot starts audition detection function, such as step S1013.
According to audition detection function (step S1014), if robot detects the sound of object, then according to being somebody's turn to do
The source of sound positions the position (step S1016) of object, and robot rotates health automatically subsequently, makes
Photographic head alignment object orientation (step S1017).After alignment, robot is again turned on vision-based detection function
And carry out follow-up visual tracking (S102).
If in step S1013, robot is also not detected by the source of sound, then be again started up vision-based detection
Function is scanned finding target, step S1015.In this step, the necessary portion of revolute's health
Part so that robot can scan object in the range of planar 360 degree and vertical 360 degree.In scanning
During, if target can be found, then start visual tracking function, carry out visual tracking S102.If swept
Target, then repeated execution of steps S1013 can not be found during retouching, thus turn again to audition detection process,
Start audition detection function and carry out sound detection location or Application on Voiceprint Recognition.
But, as long as according to this object of prior art in coverage, and trigger shooting condition,
Such as specific action or sound are just taken pictures automatically, have substantial amounts of shooting effect and do not meet user and want
The photo asked exists, because in this technology, is not configured acquisition parameters required during shooting,
Or it is single according to default acquisition parameters effectiveness comparison.Such as, for stationary object with for moving object
Parameter of taking pictures is the most different.And, different people is the most different to the requirement of composition.Additionally, robot
Owner the preference data of picture information is had each opposite sex.Therefore, carry out capturing (either in robot
Automatically capture or passive capture) time, in addition it is also necessary to it can process according to user in operation to picture information in the past
It is continuously generated learning data, thus optimizes the shooting condition parameter automatically captured or under passive candid photograph.
As in figure 2 it is shown, which show according to an embodiment of the invention for improving above-mentioned intelligent machine
The self-learning method flow chart of the picture effect that people captures out automatically.
In fig. 2, method starts from step S201, in this step, is first turned on the shooting of intelligent robot
Function.The shoot function of intelligent robot can automatically open up under in response to shooting trigger condition, it is possible to
To be the instruction passive open according to user.Under robot is in screening-mode, it will be constantly by regarding
Feel or sound detection system follow the tracks of object, and rotate self, make object be in all the time in coverage frame,
Method as shown in Figure 1.
It follows that in step S202, be carried out shooting operation after default shooting condition parameter meets.
The moment that such as will appear from triggering the object of the specific action taken pictures captures, or carries out video record behaviour
Make.The shooting condition parameter preset is only most basic acquisition parameters, under the instruction of shooting trigger condition,
Such acquisition parameters can only meet basic shooting demand.
Therefore, in order to make picture information that candid photograph goes out or video data satisfactory, robot operating system is also
Need acquisition user to learn for preference data and the picture evaluating of composition, see step S203.?
When automatically capturing function of initial use robot, user can come by deleting some photo or video clip
Abandon those undesirable candid photographs, or by preserve some photo or video clip to retain those preferable
Capture.For the candid photograph video clip remained or picture, user can carry out using photoshop further
Etc U.S. figure instrument carry out editor and cutting evidence.Additionally, capture pictures can also be uploaded with good by user
Friend shares, thus shows user's forward evaluation to this candid photograph picture.Other user evaluation to being shared for
The composition of picture evaluating is also that comparison is crucial.Such as, praise operation by the point of acquisition good friends or praise
Beautiful evaluation or further picture forward operation to know the picture evaluating that forward is evaluated.These operations
Details, robot operating system is required in this step obtain, thus knows user's preference data to composition
With picture evaluating.
It follows that in step S204, to the picture evaluating made for described picture information with inclined
Good data carry out machine learning, and revise above-mentioned default shooting condition parameter based on learning process, to obtain machine
The optimum shooting condition parameter of device people's self study.
These data of machine learning include that mass data study can be processed, draw optimal value of the parameter by machine, and
Constantly circulate this process.Particularly, it is simply that composition preference data and the picture evaluation of user are joined by robot
Number is analyzed and learns.
Composition preference data includes the people relevant to composition or the number of thing, people or thing position in the picture, people
Or the size of thing, people or thing and the distance of camera lens, the angle of face, the size of face, bright and dark light degree, thing
Part, face emotion, type of emotion, emotion degree, human action, human face action, kind of object, object and
The position of people, age, sex, face value etc..Above-mentioned data can use existing Visual identification technology to obtain
?.Visual identification technology can analyze the bright and dark light degree in picture, people/object position in the picture.
Additionally, also by using speech recognition technology, semantic analysis technology to analyze user and to the comment of photo be
Forward or negative sense, thus obtain the picture evaluating of user.Such as, by user to capture pictures
Process carries out speech recognition and semantic analysis.If user to she good friend or at this photo of Web realease or regard
Frequently short-movie, then can draw this picture or forward evaluating corresponding to video clip.And if the user while browse
After photo or video clip, carry out the operation deleted, then can obtain joining corresponding to the negative sense evaluation of this data
Number.Other users can also be as picture evaluating for robot learning to the evaluation of photo.Such as, use
The good friend at the family picture to being issued has carried out a little praising operation or forward comment, or forwards further, then may be used
To obtain forward evaluating.And if this good friend steps on, negative sense has commented on this picture information, then obtain is negative
To evaluating.
In summary, include commenting from forward evaluation and the negative sense of user according to the picture evaluating of the present invention
Valency, in the step modifying default shooting condition parameter, is carried out the picture information obtaining forward evaluation
Analysis of image data obtains the first correction value set, the picture information obtaining negative sense evaluation is carried out view data and divides
Analysis obtains the second correction value set, wherein, superposition the first correction value collection in original default shooting condition parameter
Close, and in described shooting condition parameter, deduct described second correction value set, thus by default shooting condition
Parameter modification is optimum shooting condition parameter.
Finally, in step S205, using optimum shooting condition parameter replacing as default shooting condition parameter
Swap-in row preserves, and operates for shooting next time.In other words, the optimum preserved in robot system
Shooting condition parameter is dynamically change, and once user carries out operation process to picture information, and machine will be carried out
Study, extracts the data being wherein correlated with composition or picture evaluating.By such machine learning, intelligence
In energy robot system, the optimum shooting condition parameter of storage ensure that always optimal conditions parameter.
Learn by oneself as it is shown on figure 3, wherein show in more detail robot according to an embodiment of the invention
Practise and improve the method flow diagram automatically capturing effect.
In figure 3, first given a set of parameter preset for triggering candid photograph automatically and initial parameter value.Machine
These ginsengs in the relevant device of device people such as vision detection system or sound detection system tracing and monitoring reality
Whether number changes, thus judges whether to trigger the condition of candid photograph automatically.The most automatically capture
Operation.If do not triggered, then continuing to keep following the tracks of object, line parameter of going forward side by side monitors.
After candid photograph operation completes automatically, picture or the video of shooting are sent to user, the most remotely lead to
Letter is sent on the terminal units such as the mobile phone of user, panel computer.Certainly, local can also preserve here, use
Family is browsed by display.When user browses captured photo, the process being taked them is carried out
Pay close attention to.Judge that user is forward or negative sense to the result of picture.Either forward or negative sense evaluation,
Machine carries out graphical analysis per capita to corresponding picture, thus the parameter that the photo of user's process is corresponding is carried out machine
Device learns.Specifically, machine by excavate, analyze composition preference data, light intensity, time of exposure,
Personage's size etc., and they are respectively formed as the first correction value set (corresponding to forward evaluation) and
Second correction value set (corresponding to negative sense evaluation), in order to default shooting condition parameter is modified.
Amendment parameter is included in superposition the first correction value set in default shooting condition parameter, and at default shooting bar
Deducting the second correction value set in part parameter, thus obtain new parameter and parameter value, machine is as
Excellent shooting condition parameter is saved in order to shoot next time.
It should be strongly noted that the present invention method describe realize in computer systems.This calculating
Machine system such as can be arranged in the control core processor of robot.Such as, method described herein is permissible
Being embodied as can be to control the software that logic performs, and it is performed by the CPU in robot control system.Herein
Described function can be implemented as the programmed instruction set being stored in non-transitory tangible computer computer-readable recording medium.
When implemented in this fashion, this computer program includes one group of instruction, when the instruction of this group is run by computer
It promotes the method that computer performs to implement above-mentioned functions.FPGA can temporarily or permanently be arranged on
In non-transitory tangible computer computer-readable recording medium, such as ROM chip, computer storage, disk or
Other storage mediums.In addition to realizing with software, logic as herein described may utilize discrete parts, integrated
Circuit and programmable logic device (such as, field programmable gate array (FPGA) or microprocessor) knot
Close the FPGA used, or include that any other equipment of they combination in any embodies.All this type of
Within embodiment is intended to fall under the scope of the present invention.
As shown in Figure 4, which show and improve the self study device capturing effect according to an embodiment of the invention
400。
According to another aspect of the present invention, additionally provide a kind of self study device 400 improving and capturing effect,
Described device 400 includes:
Shoot function opens unit 401, and it is in order to open the shoot function of intelligent robot;
Shooting operation execution unit 402, it performs shooting operation based on default shooting condition parameter;
Parameter acquiring unit 403, it is for obtaining the preference data for composition and picture evaluating;
Shooting condition parameter modifying unit 404, it is in order to the picture evaluation made for described picture information
Parameter and preference data carry out machine learning, and revise described default shooting condition parameter based on learning process,
To obtain the optimum shooting condition parameter of robot self study;
Parameter storage unit 405, preserves described optimum shooting condition parameter, for bat next time
Take the photograph operation.
Improving in the self study device 400 capturing effect according to the present invention, described picture evaluating includes
From forward evaluation and the negative sense evaluation of user, in described shooting condition parameter modifying unit, to obtaining forward
The picture information evaluated carries out analysis of image data and obtains the first correction value set, to obtaining the picture that negative sense is evaluated
Data carries out analysis of image data and obtains the second correction value set, wherein, the first correction value set and second is repaiied
On the occasion of set as evaluating data, make shooting condition parameter level off to more by machine learning and meet user's specification
Shooting condition parameter.
According to the present invention improve capture effect self study device also include automatically capturing unit 406, its in order to
Make described intelligent robot under shooting trigger command, open the shoot function of intelligent robot, automatically perform
Capturing operation, shooting trigger command produces based on the shooting condition set.
The self study device 400 improving candid photograph effect according to the present invention also includes passively capturing unit 407, institute
State intelligent robot and passively open candid photograph function according to user instruction.
According to the present invention improve capture effect self study device also include automatic screening unit 408, its in order to
Making robot carry out automatic screening for the photo taken based on optimum shooting condition parameter, deletion does not meets
The picture information of described optimum shooting condition parameter, preserves the picture money meeting described optimum shooting condition parameter
Material.
Due to, it is judged that Wonderful time to be captured, almost all people judges with perception, such as
Feel the most beautiful, be a lot of fun ....Even if being the photographer of specialty, it is also difficult to all of main points are quantified out.
Machine is captured exactly needs n quantifiable parameter just can get rid of the photo that people are satisfied.By according to this
Bright machine learning, robot self with the quantized data of these parameters of sustained improvement, thus can allow machine capture
Photo out is become better and better.
It should be understood that disclosed embodiment of this invention is not limited to ad hoc structure disclosed herein, process
Step or material, and the equivalent that should extend to these features that those of ordinary skill in the related art are understood is replaced
Generation.It is to be further understood that term as used herein is only used for describing the purpose of specific embodiment, and and unexpectedly
Taste restriction.
" embodiment " mentioned in description or " embodiment " mean the specific spy in conjunction with the embodiments described
Levy, structure or characteristic are included at least one embodiment of the present invention.Therefore, description various places throughout
The phrase " embodiment " or " embodiment " that occur might not refer both to same embodiment.
While it is disclosed that embodiment as above, but described content is only to facilitate understand the present invention
And the embodiment used, it is not limited to the present invention.Technology people in any the technical field of the invention
Member, on the premise of without departing from spirit and scope disclosed in this invention, can be in the formal and details implemented
On make any amendment and change, but the scope of patent protection of the present invention, still must be with appending claims institute
Define in the range of standard.
Claims (10)
1. one kind is improved the self-learning method capturing effect, it is characterised in that said method comprising the steps of:
Open the shoot function of intelligent robot;
Shooting operation is performed based on default shooting condition parameter;
Obtain the preference data for composition and picture evaluating;
The picture evaluating made for described picture information and preference data are carried out machine learning, and base
Described default shooting condition parameter is revised, to obtain the optimum shooting condition of robot self study in learning process
Parameter;
Described optimum shooting condition parameter is preserved as the replacement of default shooting condition parameter, for
Shooting operation next time.
2. the self-learning method improving candid photograph effect as claimed in claim 1, it is characterised in that described figure
Sheet evaluating includes:
From forward evaluation and the negative sense evaluation of user, in the step that described default shooting condition parameter is modified
In Zhou, the picture information obtaining forward evaluation carried out analysis of image data and obtains the first correction value set, to obtaining
The picture information of negative sense evaluation carries out analysis of image data and obtains the second correction value set, wherein, repair first
On the occasion of set with the second correction value set as evaluating data, shooting condition parameter is made to level off to by machine learning
More meet the shooting condition parameter of user's specification.
3. the self-learning method improving candid photograph effect as claimed in claim 2, it is characterised in that described intelligence
Energy robot, under shooting trigger command, opens the shoot function of intelligent robot, automatically performs candid photograph and operates,
Shooting trigger command produces based on the shooting condition set.
4. the self-learning method improving candid photograph effect as claimed in claim 3, it is characterised in that opening
After the step of the shoot function of intelligent robot, described intelligent robot passively performs according to user instruction to capture behaviour
Make.
The self-learning method improving candid photograph effect the most as described in any of claims 4, it is characterised in that
In the case of candid photograph function is passively opened by robot, robot shoots bar for the photo taken based on optimum
Part parameter carries out automatic screening, deletes the picture information not meeting described optimum shooting condition parameter, and preservation meets
The picture information of described optimum shooting condition parameter.
6. one kind is improved the self study device capturing effect, it is characterised in that described device includes:
Shoot function opens unit, and it is in order to open the shoot function of intelligent robot;
Shooting operation execution unit, it performs shooting operation based on default shooting condition parameter;
Parameter acquiring unit, it is for obtaining the preference data for composition and picture evaluating;
Shooting condition parameter modifying unit, it is in order to the picture evaluating made for described picture information
Carry out machine learning with preference data, and revise described default shooting condition parameter based on learning process, with
Optimum shooting condition parameter to robot self study;
Parameter storage unit, preserves described optimum shooting condition parameter, grasps for shooting next time
Make.
7. the self study device improving candid photograph effect as claimed in claim 6, it is characterised in that described figure
Sheet evaluating includes the forward evaluation from user and negative sense evaluation, in described shooting condition parameter modifying unit
In, the picture information obtaining forward evaluation is carried out analysis of image data and obtains the first correction value set, to acquisition
The picture information of negative sense evaluation carries out analysis of image data and obtains the second correction value set, wherein, revises first
Value set and the second correction value set, as evaluating data, make shooting condition parameter level off to more by machine learning
Meet the shooting condition parameter of user's specification.
8. the self study device improving candid photograph effect as claimed in claim 7, it is characterised in that described dress
Putting and also include automatically capturing unit, it, with so that described intelligent robot is under shooting trigger command, opens intelligence
The shoot function of energy robot, automatically performs to capture operation, shoots trigger command and produce based on the shooting condition set
Raw.
9. the self study device improving candid photograph effect as claimed in claim 8, it is characterised in that described dress
Putting and also include passively capturing unit, candid photograph function passively opened by described intelligent robot according to user instruction.
10. the self study device improving candid photograph effect as claimed in claim 9, it is characterised in that described dress
Put and also include automatic screening unit, its with so that robot for the photo taken based on optimum shooting condition
Parameter carries out automatic screening, deletes the picture information not meeting described optimum shooting condition parameter, preserves and meet institute
State the picture information of optimum shooting condition parameter.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610407762.2A CN105915801A (en) | 2016-06-12 | 2016-06-12 | Self-learning method and device capable of improving snap shot effect |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610407762.2A CN105915801A (en) | 2016-06-12 | 2016-06-12 | Self-learning method and device capable of improving snap shot effect |
Publications (1)
Publication Number | Publication Date |
---|---|
CN105915801A true CN105915801A (en) | 2016-08-31 |
Family
ID=56749928
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610407762.2A Pending CN105915801A (en) | 2016-06-12 | 2016-06-12 | Self-learning method and device capable of improving snap shot effect |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105915801A (en) |
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107749952A (en) * | 2017-11-09 | 2018-03-02 | 睿魔智能科技(东莞)有限公司 | A kind of Intelligent unattended method for imaging and system based on deep learning |
CN107995428A (en) * | 2017-12-21 | 2018-05-04 | 广东欧珀移动通信有限公司 | Image processing method, device and storage medium and mobile terminal |
CN108284434A (en) * | 2017-01-10 | 2018-07-17 | 发那科株式会社 | Machine learning device, the impact suppression system of teaching apparatus and machine learning method |
CN108289169A (en) * | 2018-01-09 | 2018-07-17 | 北京小米移动软件有限公司 | Image pickup method, device, electronic equipment and storage medium |
CN108471499A (en) * | 2018-05-06 | 2018-08-31 | 深圳市保千里电子有限公司 | A method of taking pictures and select photo automatically based on artificial intelligence |
CN108833766A (en) * | 2018-03-21 | 2018-11-16 | 北京猎户星空科技有限公司 | Control method, device, smart machine and the storage medium of smart machine |
CN110574040A (en) * | 2018-02-14 | 2019-12-13 | 深圳市大疆创新科技有限公司 | Automatic snapshot method and device, unmanned aerial vehicle and storage medium |
CN110679114A (en) * | 2017-05-24 | 2020-01-10 | 国际商业机器公司 | Method for estimating deletability of data object |
CN110913120A (en) * | 2018-09-17 | 2020-03-24 | 北京小米移动软件有限公司 | Image shooting method and device, electronic equipment and storage medium |
WO2020195642A1 (en) * | 2019-03-28 | 2020-10-01 | 富士フイルム株式会社 | Imaging device, control method, program, and non-transitory storage medium |
CN112843678A (en) * | 2020-12-31 | 2021-05-28 | 上海米哈游天命科技有限公司 | Method and device for shooting image, electronic equipment and storage medium |
CN116369362A (en) * | 2023-05-30 | 2023-07-04 | 乳山新达食品有限公司 | Control method and control system for seafood product classification extraction device |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20080192129A1 (en) * | 2003-12-24 | 2008-08-14 | Walker Jay S | Method and Apparatus for Automatically Capturing and Managing Images |
US20110292248A1 (en) * | 2010-05-26 | 2011-12-01 | David De Leon | Camera system and method for taking photographs that correspond to user preferences |
CN103971547A (en) * | 2014-05-30 | 2014-08-06 | 深圳市中兴移动通信有限公司 | Mobile-terminal-based photography simulation teaching method and system |
CN104683697A (en) * | 2015-03-17 | 2015-06-03 | 努比亚技术有限公司 | Shooting parameter adjusting method and device |
CN105430267A (en) * | 2015-12-01 | 2016-03-23 | 厦门瑞为信息技术有限公司 | Method for adaptively adjusting camera parameters based on face image illumination parameters |
CN105578058A (en) * | 2016-02-03 | 2016-05-11 | 北京光年无限科技有限公司 | Shooting control method and device for intelligent robot and robot |
-
2016
- 2016-06-12 CN CN201610407762.2A patent/CN105915801A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20080192129A1 (en) * | 2003-12-24 | 2008-08-14 | Walker Jay S | Method and Apparatus for Automatically Capturing and Managing Images |
US20110292248A1 (en) * | 2010-05-26 | 2011-12-01 | David De Leon | Camera system and method for taking photographs that correspond to user preferences |
CN103971547A (en) * | 2014-05-30 | 2014-08-06 | 深圳市中兴移动通信有限公司 | Mobile-terminal-based photography simulation teaching method and system |
CN104683697A (en) * | 2015-03-17 | 2015-06-03 | 努比亚技术有限公司 | Shooting parameter adjusting method and device |
CN105430267A (en) * | 2015-12-01 | 2016-03-23 | 厦门瑞为信息技术有限公司 | Method for adaptively adjusting camera parameters based on face image illumination parameters |
CN105578058A (en) * | 2016-02-03 | 2016-05-11 | 北京光年无限科技有限公司 | Shooting control method and device for intelligent robot and robot |
Cited By (19)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108284434A (en) * | 2017-01-10 | 2018-07-17 | 发那科株式会社 | Machine learning device, the impact suppression system of teaching apparatus and machine learning method |
CN108284434B (en) * | 2017-01-10 | 2019-08-20 | 发那科株式会社 | The impact of machine learning device, teaching apparatus inhibits system and machine learning method |
CN110679114A (en) * | 2017-05-24 | 2020-01-10 | 国际商业机器公司 | Method for estimating deletability of data object |
CN110679114B (en) * | 2017-05-24 | 2021-08-06 | 国际商业机器公司 | Method for estimating deletability of data object |
CN107749952B (en) * | 2017-11-09 | 2020-04-10 | 睿魔智能科技(东莞)有限公司 | Intelligent unmanned photographing method and system based on deep learning |
CN107749952A (en) * | 2017-11-09 | 2018-03-02 | 睿魔智能科技(东莞)有限公司 | A kind of Intelligent unattended method for imaging and system based on deep learning |
CN107995428A (en) * | 2017-12-21 | 2018-05-04 | 广东欧珀移动通信有限公司 | Image processing method, device and storage medium and mobile terminal |
CN108289169A (en) * | 2018-01-09 | 2018-07-17 | 北京小米移动软件有限公司 | Image pickup method, device, electronic equipment and storage medium |
CN110574040A (en) * | 2018-02-14 | 2019-12-13 | 深圳市大疆创新科技有限公司 | Automatic snapshot method and device, unmanned aerial vehicle and storage medium |
CN108833766A (en) * | 2018-03-21 | 2018-11-16 | 北京猎户星空科技有限公司 | Control method, device, smart machine and the storage medium of smart machine |
CN108471499A (en) * | 2018-05-06 | 2018-08-31 | 深圳市保千里电子有限公司 | A method of taking pictures and select photo automatically based on artificial intelligence |
CN110913120A (en) * | 2018-09-17 | 2020-03-24 | 北京小米移动软件有限公司 | Image shooting method and device, electronic equipment and storage medium |
CN110913120B (en) * | 2018-09-17 | 2021-11-30 | 北京小米移动软件有限公司 | Image shooting method and device, electronic equipment and storage medium |
WO2020195642A1 (en) * | 2019-03-28 | 2020-10-01 | 富士フイルム株式会社 | Imaging device, control method, program, and non-transitory storage medium |
JPWO2020195642A1 (en) * | 2019-03-28 | 2021-12-23 | 富士フイルム株式会社 | Imaging equipment, control methods, programs, and non-temporary recording media |
JP7090802B2 (en) | 2019-03-28 | 2022-06-24 | 富士フイルム株式会社 | Imaging equipment, control methods, programs, and non-temporary recording media |
JP7344348B2 (en) | 2019-03-28 | 2023-09-13 | 富士フイルム株式会社 | Imaging device, control method, program, and non-temporary recording medium |
CN112843678A (en) * | 2020-12-31 | 2021-05-28 | 上海米哈游天命科技有限公司 | Method and device for shooting image, electronic equipment and storage medium |
CN116369362A (en) * | 2023-05-30 | 2023-07-04 | 乳山新达食品有限公司 | Control method and control system for seafood product classification extraction device |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105915801A (en) | Self-learning method and device capable of improving snap shot effect | |
US20220366219A1 (en) | Systems and Methods for Providing Feedback for Artificial Intelligence-Based Image Capture Devices | |
CN101427266B (en) | Method and apparatus for selective rejection of digital images | |
US20200371535A1 (en) | Automatic image capturing method and device, unmanned aerial vehicle and storage medium | |
EP2753066A2 (en) | Apparatus having camera and method for image photographing | |
CN108737717A (en) | Image pickup method, device, smart machine and storage medium | |
CN105981368A (en) | Photo composition and position guidance in an imaging device | |
JP4605039B2 (en) | Automatic photographing method, automatic photographing apparatus and automatic photographing program | |
JP2008198062A (en) | Image processor and image processing method | |
CN109948525A (en) | It takes pictures processing method, device, mobile terminal and storage medium | |
CN107395957B (en) | Photographing method and device, storage medium and electronic equipment | |
JP2007067560A (en) | Imaging apparatus and its control method, computer program and recording medium | |
CN112019739A (en) | Shooting control method and device, electronic equipment and storage medium | |
CN104820567B (en) | It is a kind of to control method, equipment and the system taken pictures using writing pencil | |
US20200380690A1 (en) | Image processing method, apparatus, and storage medium | |
CN108419012A (en) | Photographic method, device, storage medium and electronic equipment | |
CN111597938A (en) | Living body detection and model training method and device | |
CN107959795A (en) | A kind of information collecting method, equipment and computer-readable recording medium | |
CN108513066A (en) | It takes pictures composition guidance method, mobile terminal and storage medium | |
CN107979727A (en) | A kind of document image processing method, mobile terminal and computer-readable storage medium | |
JP2005045600A (en) | Image photographing apparatus and program | |
US20190320116A1 (en) | Image pickup device and method for controlling same | |
CN106131439A (en) | A kind of image pickup method and filming apparatus | |
US20130308829A1 (en) | Still image extraction apparatus | |
CN107563975A (en) | Brightness of image adjusting method, terminal and computer-readable recording medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160831 |