CN108989684A - The method and apparatus for controlling shooting distance - Google Patents
The method and apparatus for controlling shooting distance Download PDFInfo
- Publication number
- CN108989684A CN108989684A CN201810969478.3A CN201810969478A CN108989684A CN 108989684 A CN108989684 A CN 108989684A CN 201810969478 A CN201810969478 A CN 201810969478A CN 108989684 A CN108989684 A CN 108989684A
- Authority
- CN
- China
- Prior art keywords
- apart
- section
- shooting
- affiliated
- information
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Studio Devices (AREA)
- Image Analysis (AREA)
- Traffic Control Systems (AREA)
Abstract
This specification embodiment provides a kind of method and apparatus for controlling shooting distance, method includes: that terminal obtains preview image, and determines the current shooting of the terminal apart from affiliated apart from section using neural network classification model trained in advance according to the preview image；When the current shooting apart from it is affiliated be consistent apart from section pre-determined distance section corresponding with current car damage identification scene when, the terminal exports the first prompt information, and first prompt information is for reminding user that can shoot；When the current shooting apart from it is affiliated inconsistent apart from section pre-determined distance section corresponding with current car damage identification scene when, the terminal exports the second prompt information, second prompt information is used to remind user that shooting distance is improper, and user is reminded to need close or walk remote.This specification embodiment can control shooting distance, so that the photo of shooting be made to meet the needs of to car damage identification.
Description
Technical field
This specification one or more embodiment is related to computer field, more particularly to the method and dress of control shooting distance
It sets.
Background technique
Traditional vehicle insurance settle a claim scene, insurance company need send profession survey setting loss personnel to the scene of the accident to vehicle
Dam site investigation setting loss is carried out, the maintenance program and indemnity of vehicle, and floor photo and setting loss photo are provided, by photo
Shelves are stayed to verify prices for backstage inspector's core damage.In addition in the self-service Claims Resolution scene of vehicle insurance, car owner needs floor photo and determines
Photo is damaged so that the car damage identification of insurance company uses.
In traditional vehicle insurance Claims Resolution scene, application (application, App) is surveyed for insurance by each insurance company
Company setting loss person uses.This kind of App has the function of camera shooting, photo classification, and the personnel of surveying, which can be used for shooting surveying, to be determined
Photo is damaged, and sort operation is carried out to photo.In the self-service Claims Resolution scene of vehicle insurance, car owner is usually using the original camera App of mobile phone
It is shot.Existing insurance company surveys App and only provides the shooting function of general camera, in the unsuitable situation of shooting distance
Still allow to shoot photo, the quality of photo places one's entire reliance upon operator to the understanding standardized is shot, systematic can not protect
Demonstrate,prove shooting quality.The original camera of system also has same problem, is completely dependent on the operation of car owner.
According to often there is the unsuitable situation of shooting distance, the photo of shooting is not able to satisfy to vehicle for the setting loss of live shooting
The demand of setting loss, this causes, and the core of inspector damages and the work for operation of verifying prices is affected.
Accordingly, it would be desirable to there is improved plan, shooting distance can be controlled, so that it is fixed to vehicle to meet the photo of shooting
The demand of damage.
Summary of the invention
This specification one or more embodiment describes a kind of method and apparatus for controlling shooting distance, can control bat
Photographic range, so that the photo of shooting be made to meet the needs of to car damage identification.
In a first aspect, providing a kind of method for controlling shooting distance, method includes:
Terminal obtains preview image, and is determined according to the preview image using neural network classification model trained in advance
The current shooting of the terminal is apart from affiliated apart from section；
When the current shooting apart from affiliated apart from section pre-determined distance area corresponding with current car damage identification scene
Between when being consistent, the terminal exports the first prompt information, and first prompt information is for reminding user that can shoot；
When the current shooting apart from affiliated apart from section pre-determined distance area corresponding with current car damage identification scene
Between it is inconsistent when, the terminal exports the second prompt information, and second prompt information is for reminding user's shooting distance not
Properly, and user is reminded to need close or walk remote.
In a kind of possible embodiment, the car damage identification scene includes vista shot scene and close shot shooting field
Scape；
The method also includes:
When the car damage identification scene is vista shot scene, the pre-determined distance section is set as first distance area
Between；
When the car damage identification scene is close shot photographed scene, the pre-determined distance section is set as second distance area
Between；
Wherein, the average distance in the first distance section is greater than the average distance in the second distance section.
In a kind of possible embodiment, the neural network classification model is based on training sample and trains in advance, institute
Stating training sample includes the multiple images under car damage identification scene, and each image has the mark in proven shooting distance section
Label.
It is described that neural network classification trained in advance is utilized according to the preview image in a kind of possible embodiment
Model determines the current shooting of the terminal apart from affiliated apart from section, comprising:
Using the preview image as the input of the neural network classification model, the neural network classification model is utilized
Determine that the corresponding tag along sort of the preview image, the tag along sort correspond to the current shooting of the terminal apart from affiliated
Apart from section.
In a kind of possible embodiment, the terminal also obtains at least one sensor information of the terminal；
It is described that the current of the terminal is determined using neural network classification model trained in advance according to the preview image
Apart from section belonging to shooting distance, comprising:
Using at least one sensor information of the preview image and the terminal as the neural network classification model
Input, determine the corresponding tag along sort of the preview image, the tag along sort pair using the neural network classification model
The current shooting of terminal described in Ying Yu is apart from affiliated apart from section.
Further, at least one sensor information, comprising: gyroscope information, depth transducer information or infrared
Sensor information.
In a kind of possible embodiment, the method also includes:
When the current shooting apart from affiliated apart from section pre-determined distance area corresponding with current car damage identification scene
Between it is inconsistent when, the terminal changes shooting button form or blocks shooting function.
In a kind of possible embodiment, first prompt information be voice messaging, text information, image information and
At least one of vibration information；And/or
Second prompt information is at least one of voice messaging, text information, image information and vibration information.
In a kind of possible embodiment, when the current shooting is apart from affiliated fixed apart from section and current vehicle
When the corresponding pre-determined distance section of damage scene is inconsistent, the terminal exports the second prompt information, comprising:
When it is described it is affiliated apart from section apart from mean value be greater than pre-determined distance section apart from mean value when, the terminal is defeated
Second prompt information out, second prompt information is to remind user close；
When it is described it is affiliated apart from section apart from mean value be less than pre-determined distance section apart from mean value when, the terminal is defeated
Second prompt information out, second prompt information is to remind user to walk far.
Second aspect, provides a kind of device for controlling shooting distance, and device includes:
Determination unit, for utilizing neural network trained in advance when acquisition preview image, and according to the preview image
Disaggregated model determines current shooting apart from affiliated apart from section；
First output unit, the current shooting for determining when the determination unit is apart from affiliated apart from section and current
Car damage identification scene corresponding pre-determined distance section when being consistent, export the first prompt information, first prompt information is used
It can be shot in prompting user；
Second output unit, the current shooting for determining when the determination unit is apart from affiliated apart from section and current
The corresponding pre-determined distance section of car damage identification scene it is inconsistent when, export the second prompt information, second prompt information
For reminding user that shooting distance is improper, and user is reminded to need close or walk remote.
The third aspect provides a kind of computer readable storage medium, is stored thereon with computer program, when the calculating
When machine program executes in a computer, enable computer execute first aspect method.
Fourth aspect provides a kind of calculating equipment, including memory and processor, and being stored in the memory can hold
Line code, when the processor executes the executable code, the method for realizing first aspect.
The method and apparatus provided by this specification embodiment, terminal obtain preview image, and according to the preview graph
As determined first with neural network classification model trained in advance the current shooting of the terminal apart from affiliated apart from section, so
It is apart from section pre-determined distance section corresponding with current car damage identification scene apart from affiliated according to the current shooting afterwards
It is no to be consistent, so that corresponding prompt information is exported, specifically, when the current shooting is apart from affiliated apart from section and current
Car damage identification scene corresponding pre-determined distance section when being consistent, the terminal exports the first prompt information, and described first mentions
Show information for reminding user that can shoot；When the current shooting apart from affiliated apart from section and current car damage identification field
When the corresponding pre-determined distance section of scape is inconsistent, the terminal exports the second prompt information, and second prompt information is used for
It reminds user's shooting distance improper, and user is reminded to need close or walk remote.Therefore this specification embodiment mentions
The method and apparatus of confession, terminal obtain preview image, find shooting distance problem by intelligent algorithm, and remind use in time
How family adjusts shooting distance, can control shooting distance by the interaction of terminal and user, so that the photo of shooting be made to meet
Demand to car damage identification.
Detailed description of the invention
In order to illustrate the technical solution of the embodiments of the present invention more clearly, required use in being described below to embodiment
Attached drawing be briefly described, it should be apparent that, drawings in the following description are only some embodiments of the invention, for this
For the those of ordinary skill of field, without creative efforts, it can also be obtained according to these attached drawings others
Attached drawing.
Fig. 1 is the implement scene schematic diagram of one embodiment that this specification discloses；
Fig. 2 shows the method flow diagrams for controlling shooting distance according to one embodiment；
Fig. 3 shows the user according to one embodiment and the interaction schematic diagram of terminal；
Fig. 4 shows the schematic block diagram of the device of the control shooting distance according to one embodiment.
Specific embodiment
With reference to the accompanying drawing, the scheme provided this specification is described.
Fig. 1 is the implement scene schematic diagram of one embodiment that this specification discloses.As shown in Figure 1, the implement scene is
Car damage identification scene in vehicle insurance Claims Resolution.11 using terminal 12 of user shoots vehicle 13, and the image of shooting is used for vehicle
13 carry out setting loss.Due in car damage identification scene to image include in have specific requirement, correspondingly, to shooting distance
There is specific requirement.It is understood that above-mentioned shooting distance is plane and vehicle where the camera of terminal 12 when shooting image
13 the distance between vehicle damage surface.In a typical implement scene, by shooting process be divided into vista shot and
Close shot shoots two steps.It needs to shoot in vista shot to understand damage location overall picture, needs to shoot clear damage in close shot shooting
Hurt details.
It should be noted that above-mentioned terminal 12 can be any with camera and with the terminal of processing function, for example,
Mobile phone or plate, this specification embodiment do not limit this.
In one or more embodiments that this specification discloses, a kind of improved terminal 12 is provided and is shot.It shot
Shooting distance is estimated according to the information of picture using intelligent algorithm in journey, alternatively, according to picture, gyroscope, depth
Degree sensor and the information of infrared sensor estimates shooting distance, helps to improve the accurate of vista shot and close shot shooting
Degree.It is reminded and is stopped when finding that shooting distance is improper.Reminded contents include: voice or text, image or shake
Dynamic prompting informs that there are problems for photographer's distance, and informs and need close or walk remote.Obstruction includes: change shooting button
Form blocks shooting function.Remind user that can shoot when shooting distance is suitble to.
It should be noted that above-mentioned implement scene is only the method for the control shooting distance that this specification embodiment provides
A kind of possible implement scene, is not intended as the restriction to this specification embodiment implement scene.For example, the control shooting distance
Method can be used for the car damage identification scene of car to car impact, also can be applied to the car damage identification scene to shunt into one another, shoot
Image in can only include a vehicle, also may include more vehicles, this specification embodiment does not limit this.
Fig. 2 shows the method flow diagram according to the control shooting distance of one embodiment, the executing subject of this method can be with
It is terminal 12 shown in FIG. 1.As shown in Fig. 2, the method for shooting distance is controlled in the embodiment the following steps are included: step 21,
Terminal obtains preview image, and determines the terminal using neural network classification model trained in advance according to the preview image
Current shooting apart from affiliated apart from section；Step 22, when the current shooting is apart from affiliated apart from section and current
When the corresponding pre-determined distance section of car damage identification scene is consistent, the terminal exports the first prompt information, first prompt
Information is for reminding user that can shoot；Step 23, when the current shooting apart from affiliated apart from section and current vehicle
When the corresponding pre-determined distance section of setting loss scene is inconsistent, the terminal exports the second prompt information, the second prompt letter
Breath is for reminding user that shooting distance is improper, and user is reminded to need close or walk remote.Above each step is described below
Rapid specific executive mode.
First in step 21, terminal obtains preview image, and utilizes nerve net trained in advance according to the preview image
Network disaggregated model determines the current shooting of the terminal apart from affiliated apart from section.
Optionally, the neural network classification model is based on training sample and trains in advance, and the training sample includes vehicle
Multiple images under setting loss scene, each image have the label in proven shooting distance section.
Above-mentioned label can be divided at least two classes.
In one example, above-mentioned label includes close shot and component.Wherein, label is the corresponding shooting distance section of close shot
It is denoted as the first shooting distance section, when terminal is in the first shooting distance section, the image of shooting can see damage details clearly,
Meets the needs of close shot shooting.Label is that the corresponding shooting distance section of component is denoted as the second shooting distance section, at terminal
When the second shooting distance section, the image of shooting can see defective component clearly, meet the needs of vista shot.First shooting away from
The average distance of average distance from section less than the second shooting distance section.
In another example, above-mentioned label not only includes close shot and component, further includes half vehicle, full vehicle and scene.Wherein,
Label is that the corresponding shooting distance section of half vehicle is denoted as third shooting distance section, when terminal is in third shooting distance section
When, the image of shooting is it can be seen that about half vehicle, the position all around of the terminal that can distinguish one from the other relative vehicle.Label is
The complete corresponding shooting distance section of vehicle is denoted as the 4th shooting distance section, when terminal is in the 4th shooting distance section, shooting
Image it can be seen that whole vehicle.Label is that the corresponding shooting distance section in scene is denoted as the 5th shooting distance section, works as terminal
When in the 5th shooting distance section, the image of shooting can see the vehicle for leveling a vehicle and accident another party, be able to reflect thing
Therefore field condition.The average distance in the first shooting distance section is less than third shot less than the average distance in the second shooting distance section
Average distance being averaged less than fiveth shooting distance section of the average distance in photographic range section less than the 4th shooting distance section
Distance.
It is understood that the number and meaning of label can be set according to the demand of car damage identification, it is not limited on
State example.
In one example, the terminal obtains preview image, using the preview image as the neural network classification
The input of model determines the corresponding tag along sort of the preview image, the contingency table using the neural network classification model
The current shooting for corresponding to the terminal is signed apart from affiliated apart from section.
Optionally, the neural network classification model is based on training sample and trains in advance, and the training sample includes vehicle
Multiple images under setting loss scene and at least one sensor information with the associated terminal of each image, each image tool
There is the label in proven shooting distance section.
Correspondingly, in another example, the terminal obtains at least one sensor of preview image and the terminal
Information；Using at least one sensor information of the preview image and the terminal as the defeated of the neural network classification model
Enter, determines that the corresponding tag along sort of the preview image, the tag along sort correspond to using the neural network classification model
The current shooting of the terminal is apart from affiliated apart from section.
Wherein, at least one sensor information, comprising: gyroscope information, depth transducer information or infrared sensing
Device information.
Then in step 22, when the current shooting is apart from affiliated corresponding with current car damage identification scene apart from section
Pre-determined distance section when being consistent, the terminal exports the first prompt information, and first prompt information is for reminding user
It can shoot.It is understood that the number in above-mentioned pre-determined distance section can be equal to the mark of aforementioned neurological network class model
The number of label, alternatively, the number in above-mentioned pre-determined distance section might be less that the number of the label of aforementioned neurological network class model
Mesh.
In one example, the car damage identification scene includes vista shot scene and close shot photographed scene；When the vehicle
When setting loss scene is vista shot scene, the pre-determined distance section is set as first distance section；When the car damage identification
When scene is close shot photographed scene, the pre-determined distance section is set as second distance section；Wherein, the first distance section
Average distance be greater than the second distance section average distance.
It is understood that it is the corresponding shooting distance section of component that first distance section, which can be set to label, also
It is to say, setting first distance section is identical as the second shooting distance section；It is close shot that second distance section, which can be set to label,
Corresponding shooting distance section, that is to say, that setting second distance section is identical as the first shooting distance section.
Wherein, first prompt information can in voice messaging, text information, image information and vibration information extremely
Few one kind.
Finally in step 23, when the current shooting is apart from affiliated corresponding with current car damage identification scene apart from section
Pre-determined distance section it is inconsistent when, the terminal exports the second prompt information, and second prompt information is used for reminding
Family shooting distance is improper, and user is reminded to need close or walk remote.Wherein, second prompt information can be voice
At least one of information, text information, image information and vibration information.
In one example, when the current shooting is apart from affiliated corresponding with current car damage identification scene apart from section
Pre-determined distance section it is inconsistent when, the terminal can also change shooting button form or block shooting function.
In one example, when it is described it is affiliated apart from section apart from mean value be greater than pre-determined distance section apart from mean value
When, the terminal exports the second prompt information, and second prompt information is to remind user close；When the affiliated distance
Section apart from mean value be less than pre-determined distance section apart from mean value when, the terminal exports the second prompt information, described second
Prompt information is to remind user to walk far.For example, if step 21 neural network classification model determines the affiliated mark apart from section
Label are component, and are at this time close shot photographed scene, then the terminal exports the second prompt information, second prompt information to
Remind user close.
It should be noted that meeting car damage identification field if the initial shooting distance of terminal in this specification embodiment
The demand of scape (such as close shot shooting) then adjusts the shooting distance of terminal without user.In addition, user is primary after step 23
The shooting distance of adjustment terminal may still not meet the demand of car damage identification scene (such as close shot shooting), then will be again
Step 23 is executed, until the shooting distance of user's adjustment terminal meets the demand of car damage identification scene (such as close shot shooting).
Fig. 3 shows the user according to one embodiment and the interaction schematic diagram of terminal.Referring to Fig. 3, terminal is tested using distance
Algorithm is demonstrate,proved according to the demand of car damage identification scene (such as close shot shooting), exports prompt information, which uses for prompting
Family operation, for example, reminding user close to vehicle or reminding user far from vehicle, user abides by guidance shooting according to prompt operation,
After the one or many adjustment shooting distances of user, terminal determine shooting distance meet car damage identification scene (such as close shot clap
Take the photograph) demand, export prompt information, the prompt information is for prompting user's operation, for example, reminding user that can shoot.Wherein,
Being averaged for the pre-determined distance section different with the preset pre-determined distance section of vista shot, close shot is shot is shot for close shot
Distance is less than the average distance in the pre-determined distance section of vista shot.
The method provided by this specification embodiment, terminal obtains preview image, and is utilized according to the preview image
Neural network classification model trained in advance determine the current shooting of the terminal apart from affiliated apart from section, then according to institute
State whether current shooting is consistent apart from affiliated apart from section pre-determined distance section corresponding with current car damage identification scene,
To export corresponding prompt information, specifically, when the current shooting is apart from affiliated fixed apart from section and current vehicle
When the corresponding pre-determined distance section of damage scene is consistent, the terminal exports the first prompt information, and first prompt information is used
It can be shot in prompting user；When the current shooting is apart from affiliated corresponding with current car damage identification scene apart from section
When pre-determined distance section is inconsistent, the terminal exports the second prompt information, and second prompt information is for reminding user
Shooting distance is improper, and user is reminded to need close or walk remote.Therefore the side that this specification embodiment provides
Method, terminal obtain preview image, find shooting distance problem by intelligent algorithm, and remind how user adjusts bat in time
Photographic range can control shooting distance by the interaction of terminal and user, so that the photo of shooting be made to meet to car damage identification
Demand.
According to the embodiment of another aspect, a kind of device for controlling shooting distance is also provided.Fig. 4 is shown to be implemented according to one
The schematic block diagram of the device of the control shooting distance of example.As shown in figure 4, the device 400 includes:
Determination unit 41 utilizes neural network trained in advance for obtaining preview image, and according to the preview image
Disaggregated model determines current shooting apart from affiliated apart from section；
First output unit 42, for when the determination unit 41 determine current shooting apart from it is affiliated apart from section with
When the corresponding pre-determined distance section of current car damage identification scene is consistent, the first prompt information, the first prompt letter are exported
Breath is for reminding user that can shoot；
Second output unit 43, for when the determination unit 41 determine current shooting apart from it is affiliated apart from section with
When the corresponding pre-determined distance section of current car damage identification scene is inconsistent, the second prompt information, second prompt are exported
Information is used to remind user that shooting distance is improper, and user is reminded to need close or walk remote.
In one example, the car damage identification scene includes vista shot scene and close shot photographed scene；
Described device further include:
Setup unit, for setting the pre-determined distance section when the car damage identification scene is vista shot scene
For first distance section；When the car damage identification scene is close shot photographed scene, the pre-determined distance section is set as second
Apart from section；Wherein, the average distance in the first distance section is greater than the average distance in the second distance section.
In one example, the neural network classification model is based on training sample and trains in advance, the training sample
Including the multiple images under car damage identification scene, each image has the label in proven shooting distance section.
In one example, the determination unit 41 is specifically used for obtaining preview image, using the preview image as institute
The input for stating neural network classification model determines the corresponding contingency table of the preview image using the neural network classification model
Label, the tag along sort correspond to current shooting apart from affiliated apart from section.
In one example, the determination unit 41 is specifically used for obtaining preview image and at least one sensor information；
Using the preview image feature and at least one sensor information as the input of the neural network classification model, utilize
The neural network classification model determines that the corresponding tag along sort of the preview image, the tag along sort correspond to current shooting
Apart from affiliated apart from section.
Further, at least one sensor information, comprising: gyroscope information, depth transducer information or infrared
Sensor information.
In one example, described device further include:
Blocking unit, current shooting for determining when the determination unit 41 is apart from affiliated apart from section and current
When the corresponding pre-determined distance section of car damage identification scene is inconsistent, changes shooting button form or block shooting function.
In one example, the first prompt information that first output unit 42 exports be voice messaging, text information,
At least one of image information and vibration information；And/or
The second prompt information that second output unit 43 exports is voice messaging, text information, image information and shake
At least one of dynamic information.
In one example, second output unit 43, specifically for working as the affiliated of the determination unit 41 determination
Apart from section apart from mean value be greater than pre-determined distance section apart from mean value when, export second prompt information, second prompt
Information is to remind user close；When the determination unit 41 determine it is affiliated apart from section apart from mean value be less than it is default away from
With a distance from section when mean value, the second prompt information is exported, second prompt information is to remind user to walk far.
The device provided by this specification embodiment first obtains preview image by determination unit 41, and according to described pre-
Image of looking at using neural network classification model trained in advance determine current shooting apart from affiliated apart from section, then by first
Output unit 42 works as the determining current shooting of the determination unit 41 apart from affiliated apart from section and current car damage identification field
When the corresponding pre-determined distance section of scape is consistent, the first prompt information is exported, first prompt information is for reminding user can
With shooting；Or by the second output unit 43 when the determination unit 41 determine current shooting apart from it is affiliated apart from section with
When the corresponding pre-determined distance section of current car damage identification scene is inconsistent, the second prompt information, second prompt are exported
Information is used to remind user that shooting distance is improper, and user is reminded to need close or walk remote.Therefore this specification
The device that embodiment provides, terminal obtain preview image, find shooting distance problem by intelligent algorithm, and remind in time
How user adjusts shooting distance, can control shooting distance by the interaction of terminal and user, to keep the photo of shooting full
Demand of the foot to car damage identification.
In one or more embodiments of this specification, from the letter of image, gyroscope, depth transducer and infrared sensor
Breath estimates camera posture, and is verified by the relative positional relationship of vehicle and camera to shooting distance.Verifying knot
Fruit adjusts distance to determine how prompt user, by prompting user's operation to achieve the effect that promote shooting quality.
According to the embodiment of another aspect, a kind of computer readable storage medium is also provided, is stored thereon with computer journey
Sequence enables computer execute and combines method described in Fig. 2 and Fig. 3 when the computer program executes in a computer.
According to the embodiment of another further aspect, a kind of calculating equipment, including memory and processor, the memory are also provided
In be stored with executable code, when the processor executes the executable code, realize the method in conjunction with described in Fig. 2 and Fig. 3.
Those skilled in the art are it will be appreciated that in said one or multiple examples, function described in the invention
It can be realized with hardware, software, firmware or their any combination.It when implemented in software, can be by these functions
Storage in computer-readable medium or as on computer-readable medium one or more instructions or code transmitted.
Above-described specific embodiment has carried out further the purpose of the present invention, technical scheme and beneficial effects
It is described in detail, it should be understood that being not intended to limit the present invention the foregoing is merely a specific embodiment of the invention
Protection scope, all any modification, equivalent substitution, improvement and etc. on the basis of technical solution of the present invention, done should all
Including within protection scope of the present invention.
Claims (20)
1. a method of control shooting distance, which comprises
Terminal obtains preview image, and according to preview image utilization neural network classification model determination trained in advance
The current shooting of terminal is apart from affiliated apart from section；
When the current shooting apart from affiliated apart from section pre-determined distance section phase corresponding with current car damage identification scene
When meeting, the terminal exports the first prompt information, and first prompt information is for reminding user that can shoot；
When the current shooting apart from it is affiliated apart from section pre-determined distance section corresponding with current car damage identification scene not
When being consistent, the terminal exports the second prompt information, and second prompt information is used to remind user that shooting distance is improper,
And user is reminded to need close or walk remote.
2. the method for claim 1, wherein the car damage identification scene includes vista shot scene and close shot shooting field
Scape；
The method also includes:
When the car damage identification scene is vista shot scene, the pre-determined distance section is set as first distance section；
When the car damage identification scene is close shot photographed scene, the pre-determined distance section is set as second distance section；
Wherein, the average distance in the first distance section is greater than the average distance in the second distance section.
3. the method for claim 1, wherein the neural network classification model is based on training sample and trains in advance,
The training sample includes the multiple images under car damage identification scene, and each image is with proven shooting distance section
Label.
4. the method for claim 1, wherein described utilize neural network point trained in advance according to the preview image
Class model determines the current shooting of the terminal apart from affiliated apart from section, comprising:
Using the preview image as the input of the neural network classification model, determined using the neural network classification model
The corresponding tag along sort of the preview image, the tag along sort correspond to the current shooting of the terminal apart from affiliated distance
Section.
5. the method for claim 1, wherein the method also includes: the terminal also obtains the terminal at least
A kind of sensor information；
The current shooting for determining the terminal using neural network classification model trained in advance according to the preview image
Apart from affiliated apart from section, comprising:
Using at least one sensor information of the preview image and the terminal as the defeated of the neural network classification model
Enter, determines that the corresponding tag along sort of the preview image, the tag along sort correspond to using the neural network classification model
The current shooting of the terminal is apart from affiliated apart from section.
6. method as claimed in claim 5, wherein at least one sensor information, comprising: gyroscope information, depth
Sensor information or infrared sensor information.
7. the method for claim 1, wherein the method also includes:
When the current shooting apart from it is affiliated apart from section pre-determined distance section corresponding with current car damage identification scene not
When being consistent, the terminal changes shooting button form or blocks shooting function.
8. the method for claim 1, wherein first prompt information is voice messaging, text information, image information
At least one of with vibration information；And/or
Second prompt information is at least one of voice messaging, text information, image information and vibration information.
9. such as method described in any item of the claim 1 to 8, wherein when the current shooting apart from affiliated apart from section
When corresponding with current car damage identification scene pre-determined distance section is inconsistent, the terminal exports the second prompt information, packet
It includes:
When it is described it is affiliated apart from section apart from mean value be greater than pre-determined distance section apart from mean value when, the terminal output the
Two prompt informations, second prompt information is to remind user close；
When it is described it is affiliated apart from section apart from mean value be less than pre-determined distance section apart from mean value when, the terminal output the
Two prompt informations, second prompt information is to remind user to walk far.
10. a kind of device for controlling shooting distance, described device include:
Determination unit utilizes neural network classification mould trained in advance for obtaining preview image, and according to the preview image
Type determines current shooting apart from affiliated apart from section；
First output unit, the current shooting for determining when the determination unit is apart from affiliated apart from section and current vehicle
When the corresponding pre-determined distance section of setting loss scene is consistent, the first prompt information is exported, first prompt information is for mentioning
Awake user can shoot；
Second output unit, the current shooting for determining when the determination unit is apart from affiliated apart from section and current vehicle
When the corresponding pre-determined distance section of setting loss scene is inconsistent, the second prompt information is exported, second prompt information is used for
It reminds user's shooting distance improper, and user is reminded to need close or walk remote.
11. device as claimed in claim 10, wherein the car damage identification scene includes vista shot scene and close shot shooting
Scene；
Described device further include:
Setup unit, for setting the pre-determined distance section as the when the car damage identification scene is vista shot scene
One apart from section；When the car damage identification scene is close shot photographed scene, the pre-determined distance section is set as second distance
Section；Wherein, the average distance in the first distance section is greater than the average distance in the second distance section.
12. device as claimed in claim 10, wherein the neural network classification model is based on training sample and instructs in advance
Practice, the training sample includes the multiple images under car damage identification scene, and each image has proven shooting distance area
Between label.
13. device as claimed in claim 10, wherein the determination unit is specifically used for obtaining preview image, will be described pre-
It lookes at input of the image as the neural network classification model, determines the preview image using the neural network classification model
Corresponding tag along sort, the tag along sort correspond to current shooting apart from affiliated apart from section.
14. device as claimed in claim 10, wherein the determination unit, specifically for acquisition preview image and at least
A kind of sensor information；Using the preview image and at least one sensor information as the neural network classification model
Input, determine the corresponding tag along sort of the preview image, the tag along sort pair using the neural network classification model
Should in current shooting apart from affiliated apart from section.
15. device as claimed in claim 14, wherein at least one sensor information, comprising: gyroscope information, depth
Spend sensor information or infrared sensor information.
16. device as claimed in claim 10, wherein described device further include:
Blocking unit, the current shooting for determining when the determination unit is apart from affiliated fixed apart from section and current vehicle
When the corresponding pre-determined distance section of damage scene is inconsistent, changes shooting button form or block shooting function.
17. device as claimed in claim 10, wherein the first prompt information of the first output unit output is voice letter
At least one of breath, text information, image information and vibration information；And/or
Second prompt information of the second output unit output is voice messaging, text information, image information and vibration information
At least one of.
18. the device as described in any one of claim 10 to 17, wherein second output unit is specifically used for working as institute
State that determination unit determines it is affiliated apart from section apart from mean value be greater than pre-determined distance section apart from mean value when, output second
Prompt information, second prompt information is to remind user close；When the determination unit determine it is affiliated apart from section
Apart from mean value be less than pre-determined distance section apart from mean value when, export the second prompt information, second prompt information to
User is reminded to walk far.
19. a kind of computer readable storage medium, is stored thereon with computer program, when the computer program in a computer
When execution, computer perform claim is enabled to require the method for any one of 1-9.
20. a kind of calculating equipment, including memory and processor, executable code, the processing are stored in the memory
When device executes the executable code, the method for any one of claim 1-9 is realized.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810969478.3A CN108989684A (en) | 2018-08-23 | 2018-08-23 | The method and apparatus for controlling shooting distance |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810969478.3A CN108989684A (en) | 2018-08-23 | 2018-08-23 | The method and apparatus for controlling shooting distance |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108989684A true CN108989684A (en) | 2018-12-11 |
Family
ID=64547198
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810969478.3A Pending CN108989684A (en) | 2018-08-23 | 2018-08-23 | The method and apparatus for controlling shooting distance |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108989684A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110910628A (en) * | 2019-12-02 | 2020-03-24 | 支付宝(杭州)信息技术有限公司 | Interactive processing method and device for vehicle damage image shooting and electronic equipment |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040263625A1 (en) * | 2003-04-22 | 2004-12-30 | Matsushita Electric Industrial Co., Ltd. | Camera-linked surveillance system |
| CN106952303A (en) * | 2017-03-09 | 2017-07-14 | 北京旷视科技有限公司 | Vehicle distance detecting method, device and system |
| CN107194323A (en) * | 2017-04-28 | 2017-09-22 | 阿里巴巴集团控股有限公司 | Car damage identification image acquiring method, device, server and terminal device |
| CN107635095A (en) * | 2017-09-20 | 2018-01-26 | 广东欧珀移动通信有限公司 | Shoot method, apparatus, storage medium and the capture apparatus of photo |
| CN107909622A (en) * | 2017-11-30 | 2018-04-13 | 上海联影医疗科技有限公司 | Model generating method, the scanning planing method of medical imaging and medical image system |
| CN108174108A (en) * | 2018-03-08 | 2018-06-15 | 广州三星通信技术研究有限公司 | The method and apparatus and mobile terminal for effect of taking pictures are adjusted in the terminal |
-
2018
- 2018-08-23 CN CN201810969478.3A patent/CN108989684A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040263625A1 (en) * | 2003-04-22 | 2004-12-30 | Matsushita Electric Industrial Co., Ltd. | Camera-linked surveillance system |
| CN106952303A (en) * | 2017-03-09 | 2017-07-14 | 北京旷视科技有限公司 | Vehicle distance detecting method, device and system |
| CN107194323A (en) * | 2017-04-28 | 2017-09-22 | 阿里巴巴集团控股有限公司 | Car damage identification image acquiring method, device, server and terminal device |
| CN107635095A (en) * | 2017-09-20 | 2018-01-26 | 广东欧珀移动通信有限公司 | Shoot method, apparatus, storage medium and the capture apparatus of photo |
| CN107909622A (en) * | 2017-11-30 | 2018-04-13 | 上海联影医疗科技有限公司 | Model generating method, the scanning planing method of medical imaging and medical image system |
| CN108174108A (en) * | 2018-03-08 | 2018-06-15 | 广州三星通信技术研究有限公司 | The method and apparatus and mobile terminal for effect of taking pictures are adjusted in the terminal |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110910628A (en) * | 2019-12-02 | 2020-03-24 | 支付宝(杭州)信息技术有限公司 | Interactive processing method and device for vehicle damage image shooting and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9146106B2 (en) | Laser receiver using a smart device | |
| JP5109836B2 (en) | Imaging device | |
| CN112258567B (en) | Visual positioning method and device for object grabbing point, storage medium and electronic equipment | |
| US7391886B1 (en) | Digital camera with image tracking system | |
| US9332208B2 (en) | Imaging apparatus having a projector with automatic photography activation based on superimposition | |
| WO2013099271A1 (en) | Dimension measuring method, electronic device with camera, and program for electronic device with camera | |
| CN104754216A (en) | Photographing method and device | |
| CN104104867A (en) | Method for controlling image photographing device for photographing and device thereof | |
| CN103747180A (en) | Photo shooting method and photographing terminal | |
| CN109478227A (en) | Calculate the iris in equipment or the identification of other physical feelings | |
| JP6395010B2 (en) | Information provision system for billing location evaluation | |
| JP2020161174A (en) | Information processing device and recognition support method | |
| KR102186597B1 (en) | Target size measurement system and method | |
| US20170352132A1 (en) | Method and apparatus for correcting tilt of subject ocuured in photographing, mobile terminal, and storage medium | |
| CN106197366A (en) | Distance information processing method and device | |
| JP2015050693A (en) | Camera installation simulator and computer program thereof | |
| CN109040592A (en) | The method and apparatus for controlling shooting angle | |
| JP6723720B2 (en) | Display system, information processing method and program | |
| CN105180802A (en) | Identification method and device of object size information | |
| CN108989684A (en) | The method and apparatus for controlling shooting distance | |
| KR101924302B1 (en) | Projection mapping system and apparatus | |
| US20220070365A1 (en) | Mixed reality image capture and smart inspection | |
| JP6936828B2 (en) | Construction drawing creation support system | |
| CN103927728B (en) | A kind for the treatment of method and apparatus of image | |
| CN105426886B (en) | A kind of face verification method and face verification device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| REG | Reference to a national code |
Ref country code: HK Ref legal event code: DE Ref document number: 40001300 Country of ref document: HK |
|
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181211 |
|
| RJ01 | Rejection of invention patent application after publication |