CN109639964A - Image processing method, processing unit and computer readable storage medium - Google Patents
Image processing method, processing unit and computer readable storage medium Download PDFInfo
- Publication number
- CN109639964A CN109639964A CN201811416403.9A CN201811416403A CN109639964A CN 109639964 A CN109639964 A CN 109639964A CN 201811416403 A CN201811416403 A CN 201811416403A CN 109639964 A CN109639964 A CN 109639964A
- Authority
- CN
- China
- Prior art keywords
- image
- reference picture
- eigenvector
- feature vector
- model
- 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)
Abstract
The application is about a kind of image processing method, processing unit and computer storage medium.The image processing method includes: the first image obtained about present filming scene;The first image is sent to server;At least one reference picture about the first image is received, which is by the server according to the similar image with the first image semanteme got after the first image semantic analysis;And show at least one described reference picture, in order to which user shoots the present filming scene with reference at least one described described reference picture.According to the image processing method, client receives server according at least one reference picture similar with first image, semantic got after first image, semantic analysis.Client shows at least one reference picture about first image.In order to which user likes according to itself, the present filming scene is shot with reference at least one reference picture.To improve shooting effect.
Description
Technical field
The application belongs to technical field of image processing, especially image processing method, processing unit and computer-readable deposits
Storage media.
Background technique
With science and technology be showing improvement or progress day by day and smart phone it is universal, either in daily life or out on tours, when seeing
When to beautiful scenery or unique building, many users subconsciously take out mobile phone and take pictures.For personage's shooting,
Other than the shooting skill of photographer, the quality of imaged person posture and composition selection during taking pictures, and determine to shine
One key factor of piece shooting success or not.For professional person (such as model, star etc.), during taking pictures
It to obtain good shooting effect is unusual a piece of cake feelings by transformation different gestures.But it in daily life, shows various
Various kinds posture is taken pictures and is but less easy for many layman, therefore often because not arrogating to oneself in taking pictures
Long pose brings many worries to imaged person and photographer.For example, since imaged person is bad to pose, then
Entirely take pictures during, the most of the time, which can spend in, poses in this link, not only will cause shooting efficiency and at
Power it is low, and repeatedly adjustment posture be also easy that photographer and imaged person is made to have a feeling of impatience.In another example due to
Imaged person is bad to pose, therefore often moulding is single for the photo shot, and is unable to reach good shooting effect.
Summary of the invention
To overcome the problems, such as that take pictures present in the relevant technologies effect difference and time-consuming, the application disclose a kind of image procossing
Method, processing unit and computer readable storage medium, to solve the problems, such as existing process of taking pictures.
According to the embodiment of the present application in a first aspect, provide a kind of image processing method, described image processing method application
In client, comprising:
Obtain the first image about present filming scene;
The first image is sent to server;
Receive at least one reference picture about the first image, the reference picture be by the server according to
The similar image with the first image semanteme got after the first image semantic analysis;And
At least one described reference picture is shown, in order to which user is with reference at least one described described reference picture shooting
The present filming scene.
Optionally, at least one reference picture described in the display, in order to which user is with reference at least one described described ginseng
Present filming scene described in image taking is examined, simultaneously, further includes:
The acquisition parameters extracted at least one described described reference picture and/or composition parameter are shown, in order to institute
It states user and refers to the application acquisition parameters and/or the composition parameter shooting present filming scene.
Optionally, the acquisition parameters include: filter, white balance, exposure compensating number and focal length, the composition parameter packet
It includes: background, composition, personage's posture.
According to the second aspect of the embodiment of the present application, a kind of image processing method, described image processing method application are provided
In server end, comprising:
Receive the first image about present filming scene that client is sent;
Semantic analysis is carried out to the first image;
According to the semantic analysis result, the similar reference picture of semanteme of search and the first image;And
At least one described reference picture is sent to the client.
Optionally, described that semantic analysis is carried out to the first image, comprising: obtain the first image first is special
Levy vector sum second feature vector.
Optionally, described according to the semantic analysis result, search is similar with reference to figure with the semanteme of the first image
Picture, comprising:
Obtain the first eigenvector of each prestored images;
Obtain the second feature vector of each prestored images;
The mapping for establishing and storing each prestored images, the first eigenvector and the second feature vector is closed
System, the dimension of the first eigenvector are greater than the dimension of the second feature vector;
The first image and multiple pre-stored images are subjected to aspect ratio pair;And
The reference picture of at least one the first image is obtained according to comparison result.
It is optionally, described by the first image and multiple pre-stored images carry out aspect ratio pair, comprising:
By multiple second feature vectors of the multiple prestored images of second feature vector sum of the first image point
It carry out not similitude comparison;
At least one target image is obtained according to comparison result;And
The first eigenvector of the first eigenvector of the first image and at least one target image is carried out
Similarity compares;
At least one described reference picture is obtained according to comparison result.
Optionally, described at least one target image is obtained according to comparison result to include:
If the distance of the second feature vector of the second feature vector sum the first image of a prestored images is small
In given threshold, using the prestored images as the target image;
It is described to include: according at least one described reference picture of comparison result acquisition
If the distance of the first eigenvector of the first eigenvector and the first image of prestored images is small
In given threshold, using the prestored images as the reference picture.
Optionally, the first eigenvector is obtained according to the first model, the second feature is obtained according to the second model
Vector.
Optionally, first model is N layers of convolutional neural networks model, and second model is N+1 layers of convolutional Neural
Network model, second model increase a hidden layer between the N-1 layer of first model and N layers.
Optionally, described image processing method, further includes:
Establish training set and verifying collection;
According to the training set and verifying collection, first model is obtained based on the convolutional neural networks model training
With second model.
Optionally, described image processing method, further includes: the image concentrated to the training set and the verifying carries out
Pretreatment.
Optionally, described image processing method, further includes: receive feedback information, and according to feedback information optimization described the
One model and second model.
According to a third aspect of the embodiments of the present invention, a kind of image processing apparatus, described image processing unit application are provided
In client, comprising:
First shooting module, for obtaining the first image about present filming scene;
First sending module, for sending the first image to server;
First receiving module, for receiving at least one reference picture about the first image, which is
By the server according to the similar image with the first image semanteme got after the first image semantic analysis;
Second shooting module, for showing at least one described reference picture, in order to user with reference to it is described at least one
The reference picture shoots the present filming scene.
Optionally, second shooting module is also used to show and extract at least one described described reference picture
Acquisition parameters and/or composition parameter, in order to which the user refers to the application acquisition parameters and/or composition parameter shooting institute
State present filming scene.
Optionally, the acquisition parameters include: filter, white balance, exposure compensating number and focal length, the composition parameter packet
It includes: background, composition, personage's posture.
According to a fourth aspect of the embodiments of the present invention, a kind of image processing apparatus is provided, which is characterized in that described image
Processing unit is applied to server end, comprising:
Second receiving module, for receiving the first image about present filming scene of client transmission;
Semantic module, for carrying out semantic analysis to the first image;
Reference picture obtains module, for searching for the semantic phase with the first image according to the semantic analysis result
As reference picture;
Second sending module, for sending at least one described reference picture to the client.
Optionally, described that semantic analysis is carried out to the first image, comprising: obtain the first image first is special
Levy vector sum second feature vector.
Optionally, described according to the semantic analysis result, search is similar with reference to figure with the semanteme of the first image
Picture, comprising: obtain the first eigenvector of each prestored images;
Obtain the second feature vector of each prestored images;
The mapping for establishing and storing each prestored images, the first eigenvector and the second feature vector is closed
System, the dimension of the first eigenvector are greater than the dimension of the second feature vector;
The first image and multiple pre-stored images are subjected to aspect ratio pair;And
The reference picture of at least one the first image is obtained according to comparison result.
It is optionally, described by the first image and multiple pre-stored images carry out aspect ratio pair, comprising:
By multiple second feature vectors of the multiple prestored images of second feature vector sum of the first image point
It carry out not similitude comparison;
At least one target image is obtained according to comparison result;And
The first eigenvector of the first eigenvector of the first image and at least one target image is carried out
Similarity compares;
At least one described reference picture is obtained according to comparison result.
Optionally, described at least one target image is obtained according to comparison result to include:
If the distance of the second feature vector of the second feature vector sum the first image of a prestored images is small
In given threshold, using the prestored images as the target image;
It is described to include: according at least one described reference picture of comparison result acquisition
If the distance of the first eigenvector of the first eigenvector and the first image of prestored images is small
In given threshold, using the prestored images as the reference picture.
Optionally, the first eigenvector is obtained according to the first model, the second feature is obtained according to the second model
Vector.
Optionally, first model is N layers of convolutional neural networks model, and second model is N+1 layers of convolutional Neural
Network model, second model increase a hidden layer between the N-1 layer of first model and N layers.
According to a fifth aspect of the embodiments of the present invention, a kind of computer readable storage medium is provided, it is described computer-readable
Storage medium is stored with computer instruction, and the computer instruction, which is performed, realizes image procossing described in any of the above embodiments
Method.
According to a sixth aspect of the embodiments of the present invention, a kind of computer readable storage medium is provided, it is described computer-readable
Storage medium is stored with computer instruction, and the computer instruction, which is performed, realizes image procossing described in any of the above embodiments
Method.
According to a seventh aspect of the embodiments of the present invention, a kind of electronic equipment is provided characterized by comprising
Processor;
Memory for storage processor executable instruction;
Wherein, the processor is configured to executing image processing method described in above-mentioned any one.
According to the eighth aspect of the invention, a kind of electronic equipment is provided, comprising:
Processor;
Memory for storage processor executable instruction;
Wherein, the processor is configured to executing image processing method described in above-mentioned any one.
The technical solution that embodiments herein provides can include the following benefits:
In the image processing method, ginseng of at least one of client reception server transmission about first image
Examine image.The reference picture be by server according to first image, semantic analysis after get with first image, semantic
Similar image.Client is shown from least one received reference picture about first image of server.In order to
Family is liked according to itself, shoots the present filming scene with reference at least one reference picture.To improve shooting effect.
In the image processing method, second feature vector is carried out to first image and multiple prestored images
Similitude compares and the similarity of first eigenvector compares, since the second feature vector sum of first image is multiple pre-
The dimension for storing multiple second feature vectors of image is less than the first eigenvector and at least one target of first image
The first eigenvector of image, so multiple the second of the multiple prestored images of second feature vector sum of first image
The calculation amount that the similitude of feature vector compares is smaller, while the similarity comparison of first eigenvector is the target in acquisition
It is calculated between image and first image, avoids calculating the fisrt feature between all prestored images and first image
The similarity of vector, to further decrease calculation amount.
It should be understood that above general description and following detailed description be only it is exemplary and explanatory, not
The application can be limited.
Detailed description of the invention
Fig. 1 a is the flow chart of image processing method shown according to an exemplary embodiment.
Fig. 1 b is the flow chart of image processing method shown according to an exemplary embodiment.
Fig. 2 is the flow chart of image processing method shown according to an exemplary embodiment.
Fig. 3 a is the schematic diagram of image processing apparatus shown according to an exemplary embodiment.
Fig. 3 b is the schematic diagram of image processing apparatus shown according to an exemplary embodiment.
Fig. 4 is a kind of block diagram of device for executing image processing method shown according to an exemplary embodiment.
Fig. 5 is a kind of block diagram of device for executing image processing method shown according to an exemplary embodiment.
Specific embodiment
Example embodiments are described in detail here, and the example is illustrated in the accompanying drawings.Following description is related to
When attached drawing, unless otherwise indicated, the same numbers in different drawings indicate the same or similar elements.Following exemplary is implemented
Embodiment described in example does not represent all embodiments consistent with the application.On the contrary, they are only and such as institute
The example of the consistent device and method of some aspects be described in detail in attached claims, the application.
Fig. 1 a is the flow chart of image processing method shown according to an exemplary embodiment, which answers
For client, specifically includes the following steps:
In step S111, the first image about present filming scene is obtained.
In this step, when occurring present filming scene in camera, client shoots current shooting at regular intervals
Scene obtains the first image about present filming scene.
In step S112, the first image is sent to server.
In this step, first image shot in step S111 is sent to server by client.
In step S113, at least one reference picture about the first image is received, which is by institute
Server is stated according to the similar image with the first image semanteme got after the first image semantic analysis.
In this step, reference picture of at least one of client reception server transmission about first image.It should
Reference picture is by server according to the figure similar with first image, semantic got after first image, semantic analysis
Picture.
In step S114, at least one described reference picture is shown, in order to described at least one described in user's reference
Reference picture shoots the present filming scene.
In this step, client is shown from least one received reference picture about first image of server.
In order to which user likes according to itself, the present filming scene is shot with reference at least one reference picture.
According to the embodiment of the present application, reference of at least one of client reception server transmission about first image
Image.The reference picture be by server according to first image, semantic analysis after get with the first image, semantic phase
As image.Client is shown from least one received reference picture about first image of server.In order to user
Liked according to itself, shoots the present filming scene with reference at least one reference picture.To improve shooting effect.
Fig. 1 b is the flow chart of image processing method shown according to an exemplary embodiment, which answers
For server end, specifically includes the following steps:
In step S121, the first image about present filming scene that client is sent is received.
In this step, server receives the first image about present filming scene of client shooting.
In step S122, semantic analysis is carried out to the first image.
In this step, after server receives first image, semantic analysis is carried out to first image.
In step S123, according to the semantic analysis result, the similar reference of semanteme of search and the first image
Image.
In this step, according in step S122 to first image carry out semantic analysis as a result, searching in the server
The similar reference picture of the semanteme of rope and first image.
In step S124, at least one described reference picture is sent to the client.
In this step, the reference picture of at least one of step S123 first image is sent to client.It should
Reference picture provides reference for the subsequent shooting photo of client.
According to the embodiment of the present application, after server receives first image, semantic analysis is carried out to first image.
According to first image carry out semantic analysis as a result, searching for the similar ginseng of semanteme with first image in the server
Examine image.By at least one, the reference picture of first image is sent to client to server.The reference picture is after client
Continuous shooting photo provides reference.To improve shooting effect.
Fig. 2 is the flow chart of image processing method shown according to an exemplary embodiment, including client and server
End.Specifically includes the following steps:
In step s 201, client obtains the first image about present filming scene.
In step S202, user end to server sends the first image.
In step S203, server receives the first image about present filming scene that client is sent.
In step S204, server obtains the first eigenvector and second feature vector of the first image.
In step S205, server obtains the first eigenvector and second feature vector of each prestored images.
In step S206, each prestored images, the first eigenvector and described are established and stored to server
The mapping relations of two feature vectors, the dimension of the first eigenvector are greater than the dimension of the second feature vector.
In step S207, the first image and multiple pre-stored images are carried out aspect ratio pair by server.
In step S208, server obtains the reference picture of at least one the first image according to comparison result.
In step S209, server sends at least one described reference picture to the client.
In step S210, client receives at least one reference picture about the first image, the reference picture
It is by the server according to the similar figure with the first image semanteme got after the first image semantic analysis
Picture.
In step S211, client shows at least one described reference picture, in order to which user is with reference to described at least one
A reference picture shoots the present filming scene.
The present embodiment describes the flow chart that client and server end interacts.Step S201- step S202 and step
Rapid S210- step S211 is executed in client executing, step S203- step S209 in server end.Step S201- step
S202 is consistent with step S111- step S112, and step 203 is consistent with step S121, and step S209 is consistent with step S124, this
In just repeat no more.
In step S204, server carries out semantic analysis to first image, obtains the fisrt feature of first image
Vector sum second feature vector.
In step S205-S206, in server end, it is pre-stored multiple prestored images.Server is to multiple pre-
It stores image and carries out semantic analysis, obtain the first eigenvector and second feature vector of each prestored images.Server is built
The mapping relations of each prestored images, its first eigenvector and its second feature vector, first eigenvector are found
Dimension be greater than the second feature vector dimension.First eigenvector indicates more fine granularity relative to second feature vector
Characteristic information.Such as second feature vector can be obtained based on first eigenvector.
In step S207, first image and multiple prestored images are carried out aspect ratio pair by server.Specifically,
Server carries out multiple second feature vectors of the multiple prestored images of second feature vector sum of first image respectively
Similitude compare, such as the multiple prestored images of second feature vector sum by calculating first image multiple second
Hamming distance between feature vector come obtain first image the multiple prestored images of second feature vector sum it is multiple
The similitude of second feature vector.And by the of the first eigenvector of first image and at least one target image
One feature vector carries out similarity comparison, such as the first eigenvector by calculating first image and at least one mesh
Hamming distance between the first eigenvector of logo image come obtain first image first eigenvector and this at least one
The similarity of the first eigenvector of a target image.
In step S208, according to multiple the of the multiple prestored images of second feature vector sum of first image
The comparison result of two feature vectors, if the second feature of the second feature vector sum of prestored images first image
The Hamming distance of vector is less than given threshold, using the prestored images as target image.It is special according to the first of first image
The comparison result of the first eigenvector of vector sum at least one target image is levied, if the first of a prestored images
The Hamming distance of the first eigenvector of feature vector and first image is less than given threshold, using the prestored images as
The reference picture of first image.
In step S210-S211, client receives at least one reference picture about first image.This is with reference to figure
It seem by server according to the image similar with first image, semantic got after first image, semantic analysis.Client
End shows at least one reference picture, in order to which user shoots present filming scene with reference at least one reference picture.Client
End also shows the acquisition parameters extracted at least one reference picture and/or composition parameter, in order to which user is with reference to application
The acquisition parameters and/or composition parameter shoot present filming scene.The acquisition parameters include: filter, white balance, exposure compensating
Several and focal length, the composition parameter include: background, composition, personage's posture.
According to the embodiment of the present application, second feature vector sum the is carried out to first image and multiple prestored images
The similitude of one feature vector compares, since the dimension of second feature vector is less than the dimension of first eigenvector, so this
The calculation amount that the similarity of multiple second feature vectors of the multiple prestored images of second feature vector sum of one image compares
It is smaller, while the similarity comparison of first eigenvector is carried out after the target image of acquisition, it is only necessary to target image
Carry out similarity comparison with the first eigenvector of the first image, thus avoid all prestored images and the first image it
Between calculate the similarity of the first eigenvector, this kind of processing mode reduces calculation amount, so as to which reference picture is anti-in time
It feeds client, in order to which user shoots present filming scene with reference at least one reference picture.
In an optional embodiment of the application, the image processing method, further includes: establish the first model and
Two models.Specifically, the training set and verifying collection of image are established.The image concentrated to the training set and the verifying carries out image
The normalization of size and dimension pre-processes.According to the training set, based on convolutional neural networks model training obtain the first model and
Second model.According to the training set, the feedback information obtained when verifying to first model and second model is received,
And first model and second model are optimized according to the feedback information.First model is N layers of convolutional neural networks model,
Second model is N+1 layers of convolutional neural networks model, which increases between the N-1 layer of first model and N layers
A hidden layer is added.The first eigenvector of first image and multiple prestored images is obtained according to first model.Root
The second feature vector of first image and multiple prestored images is obtained according to second model.In one embodiment, should
The neuron excitation function of hidden layer is sigmoid activation primitive.Sigmoid activation primitive is by first image and multiple pre-
The feature vector of storage image normalizes to (0,1) section.
Fig. 3 a is the schematic diagram of image processing apparatus shown according to an exemplary embodiment, which answers
It for client, specifically includes: the first shooting module 311, the first sending module 312, the first receiving module 313 and second count
Take the photograph module 314.
First shooting module 311, for obtaining the first image about present filming scene.
First sending module 312, for sending the first image to server.
First receiving module 313, for receiving at least one reference picture about the first image, this is with reference to figure
It seem by the server according to the similar figure with the first image semanteme got after the first image semantic analysis
Picture.
Second shooting module 314, for showing at least one described reference picture, in order to which user's reference is described at least
One reference picture shoots the present filming scene.
In embodiments herein, the first shooting module 311, for when occurring present filming scene in camera, visitor
Family end shoots present filming scene at regular intervals, obtains the first image about present filming scene.First sending module
312, for sending first image to server.First receiving module 313, for receive server transmission at least one
Reference picture about first image.The reference picture is got after being analyzed by server according to first image, semantic
Image similar with first image, semantic.Second shooting module 314, for show from server it is received at least one
Reference picture about first image.In order to which user likes according to itself, shot with reference at least one reference picture
The present filming scene.
In an optional embodiment of the application, the second shooting module 314 is also used to show at least one ginseng
Examine the acquisition parameters such as the filter extracted in image, white balance, exposure compensating number and focal length and/or background, composition, Ren Wuzi
The compositions parameter such as gesture shoots present filming scene using the acquisition parameters and/or composition parameter in order to which user refers to.
Fig. 3 b is the schematic diagram of image processing apparatus shown according to an exemplary embodiment, which answers
For server end, specifically include: the second receiving module 321, semantic module 322, reference picture obtain 323 and of module
Second sending module 324.
Second receiving module 321, for receiving the first image about present filming scene of client transmission.
Semantic module 322, for carrying out semantic analysis to the first image.
Reference picture obtains module 323, for according to the semantic analysis result, the language of search and the first image
The similar reference picture of justice.
Second sending module 324, for sending at least one described reference picture to the client.
In embodiments herein, the second receiving module 321, for receive client shooting about current shooting
First image of scene.Semantic module 322, for carrying out semantic analysis to first image.Reference picture obtains module
323, for according to the semantic analysis result, the similar reference picture of semanteme of search and first image.Second sending module
324, for sending at least one reference picture to client.
In an optional embodiment of the present invention, semantic module 322, for obtaining the of first image
One feature vector and second feature vector.Reference picture obtains module 323, and first for obtaining each prestored images is special
Levy vector.Obtain the second feature vector of each prestored images.Establish and store each prestored images, first spy
Levy the mapping relations of second feature vector described in vector sum, the dimension of the first eigenvector be greater than the second feature to
The dimension of amount.The first image and multiple pre-stored images are subjected to aspect ratio pair.And according to comparison result obtain to
The reference picture of a few the first image.
In an optional embodiment of the present invention, reference picture obtains module 323, for by first image
Multiple second feature vectors of the multiple prestored images of second feature vector sum carry out similitude comparison respectively.According to comparison
As a result at least one target image is obtained, for example, if the second feature vector sum of prestored images first image
The Hamming distance of second feature vector is less than given threshold, using the prestored images as target image.And by first figure
The first eigenvector of picture and the first eigenvector of at least one target image carry out similarity comparison.According to comparison result
At least one reference picture is obtained, for example, if the first of the first eigenvector of a prestored images and first image
The Hamming distance of feature vector is less than given threshold, using the prestored images as reference picture.
In an optional embodiment of the present invention, reference picture obtains module 323, obtains the according to the first model
One feature vector.Reference picture obtains module 323, obtains second feature vector according to the second model.First model is N layers
Convolutional neural networks model, second model are N+1 layers of convolutional neural networks model, and second model is in first model
A hidden layer is increased between N-1 layers and N layers, N is more than or equal to 8.In one embodiment, the neuron of the hidden layer motivates letter
Number is sigmoid activation primitive.Sigmoid activation primitive is by the feature vector of first image and multiple prestored images
Normalize to (0,1) section.
Fig. 4 is a kind of block diagram of device 1200 for executing image processing method shown according to an exemplary embodiment.Example
Such as, interactive device 1200 can be mobile phone, computer, digital broadcasting terminal, messaging device, game console, put down
Panel device, Medical Devices, body-building equipment, personal digital assistant etc..
Referring to Fig. 4, device 1200 may include following one or more components: processing component 1202, memory 1204,
Power supply module 1206, multimedia component 1208, audio component 1210, the interface 1212 of input/output (I/ O), sensor module
1214 and communication component 1216.
The integrated operation of the usual control device 1200 of processing component 1202, such as with display, telephone call, data communication,
Camera operation and record operate associated operation.Processing component 1202 may include one or more processors 1220 to hold
Row instruction, to perform all or part of the steps of the methods described above.In addition, processing component 1202 may include one or more
Module, convenient for the interaction between processing component 1202 and other assemblies.For example, processing component 1202 may include multimedia mould
Block, to facilitate the interaction between multimedia component 1208 and processing component 1202.
Memory 1204 is configured as storing various types of data to support the operation in equipment 1200.These data
Example include any application or method for being operated on device 1200 instruction, contact data, telephone directory number
According to, message, picture, video etc..Memory 1204 can by any kind of volatibility or non-volatile memory device or it
Combination realize, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM) can
Erasable programmable read-only memory (EPROM) (EPROM), programmable read only memory (PROM), read-only memory (ROM), magnetic memory,
Flash memory, disk or CD.
Power supply module 1206 provides electric power for the various assemblies of device 1200.Power supply module 1206 may include power supply pipe
Reason system, one or more power supplys and other with for device 1200 generate, manage, and distribute the associated component of electric power.
Multimedia component 1208 includes the screen of one output interface of offer between described device 1200 and user.
In some embodiments, screen may include liquid crystal display (LCD) and touch panel (TP).If screen includes touch surface
Plate, screen may be implemented as touch screen, to receive input signal from the user.Touch panel includes one or more touchings
Sensor is touched to sense the gesture on touch, slide, and touch panel.The touch sensor can not only sense touch or cunning
The boundary of movement, but also detect duration and pressure associated with the touch or slide operation.In some embodiments
In, multimedia component 1208 includes a front camera and/or rear camera.When equipment 1200 is in operation mode,
Such as in a shooting mode or a video mode, front camera and/or rear camera can receive external multi-medium data.Often
A front camera and rear camera can be a fixed optical lens system or have focal length and optical zoom energy
Power.
Audio component 1210 is configured as output and/or input audio signal.For example, audio component 1210 includes one
Microphone (MIC), when device 1200 is in operation mode, when such as call mode, recording mode, and voice recognition mode, Mike
Wind is configured as receiving external audio signal.The received audio signal can be further stored in memory 1204 or warp
It is sent by communication component 1216.In some embodiments, audio component 1210 further includes a loudspeaker, for exporting audio
Signal.
I/O interface 1212 provides interface, above-mentioned peripheral interface module between processing component 1202 and peripheral interface module
It can be keyboard, click wheel, button etc..These buttons may include, but are not limited to: home button, volume button, start button and
Locking press button.
Sensor module 1214 includes one or more sensors, for providing the state of various aspects for device 1200
Assessment.For example, sensor module 1214 can detecte the state that opens/closes of equipment 1200, the relative positioning of component, example
Such as the display and keypad that the component is device 1200, sensor module 1214 can be with detection device 1200 or device
The position change of 1200 1 components, the existence or non-existence that user contacts with device 1200,1200 orientation of device or acceleration/
The temperature change slowed down with device 1200.Sensor module 1214 may include proximity sensor, be configured to do not appointing
What detected the presence of nearby objects when physical contact.Sensor module 1214 can also include optical sensor, such as CMOS or
Ccd image sensor, for being used in imaging applications.In some embodiments, which can also include
Acceleration transducer, gyro sensor, Magnetic Sensor, pressure sensor or temperature sensor.
Communication component 1216 is configured to facilitate the communication of wired or wireless way between device 1200 and other equipment.
Device 1200 can access the wireless network based on communication standard, such as WiFi, carrier network (such as 2G, 3G, 4G or 5G), or
Their combination.In one exemplary embodiment, communication component 1216 receives via broadcast channel and comes from external broadcasting management
The broadcast singal or broadcast related information of system.In one exemplary embodiment, the communication component 1216 further includes near field
(NFC) module is communicated, to promote short range communication.For example, radio frequency identification (RFID) technology, infrared number can be based in NFC module
It is realized according to association (IrDA) technology, ultra wide band (UWB) technology, bluetooth (BT) technology and other technologies.
In the exemplary embodiment, device 1200 can be by one or more application specific integrated circuit (ASIC), number
Signal processor (DSP), digital signal processing appts (DSPD), programmable logic device (PLD), field programmable gate array
(FPGA), controller, microcontroller, microprocessor or other electronic components are realized, for executing the above method.
In the exemplary embodiment, a kind of non-transitorycomputer readable storage medium including instruction, example are additionally provided
It such as include the memory 1204 of instruction, above-metioned instruction can be executed by the processor 1220 of device 1200 to complete the above method.Example
Such as, the non-transitorycomputer readable storage medium can be ROM, random access memory (RAM), CD-ROM, tape,
Floppy disk and optical data storage devices etc..
Fig. 5 is a kind of block diagram of device 1300 for executing image processing method shown according to an exemplary embodiment.Example
Such as, device 1300 may be provided as a server.Referring to Fig. 5, device 1300 includes processing component 1322, is further wrapped
One or more processors, and the memory resource as representated by memory 1332 are included, it can be by processing component for storing
The instruction of 1322 execution, such as application program.The application program stored in memory 1332 may include one or one with
On each correspond to one group of instruction module.In addition, processing component 1322 is configured as executing instruction, it is above-mentioned to execute
Information list display methods.
Device 1300 can also include that a power supply module 1326 be configured as the power management of executive device 1300, and one
A wired or wireless network interface 1350 is configured as device 1300 being connected to network and input and output (I/O) interface
1358.Device 1300 can be operated based on the operating system for being stored in memory 1332, such as Windows ServerTM, Mac
OS XTM, UnixTM, LinuxTM, FreeBSDTM or similar.
Those skilled in the art after considering the specification and implementing the invention disclosed here, will readily occur to the application's
Other embodiments.This application is intended to cover any variations, uses, or adaptations of the application, these modifications, purposes
Or adaptive change follow the application general principle and including the application it is undocumented in the art known in often
Knowledge or conventional techniques.The description and examples are only to be considered as illustrative, and the true scope and spirit of the application are by following
Claim point out.
It should be understood that the application is not limited to the precise structure that has been described above and shown in the drawings, and
And various modifications and changes may be made without departing from the scope thereof.Scope of the present application is only limited by the accompanying claims.
Claims (10)
1. a kind of image processing method, which is characterized in that described image processing method is applied to client, comprising:
Obtain the first image about present filming scene;
The first image is sent to server;
At least one reference picture about the first image is received, which is by the server according to described
The similar image with the first image semanteme got after the analysis of one image, semantic;And
At least one described reference picture is shown, in order to which user works as with reference to described at least one described described reference picture shooting
Preceding photographed scene.
2. image processing method according to claim 1, which is characterized in that at least one is with reference to figure described in the display
Picture, in order to which user shoots the present filming scene with reference at least one described described reference picture, simultaneously, further includes:
The acquisition parameters extracted at least one described described reference picture and/or composition parameter are shown, in order to the use
Family refers to the application acquisition parameters and/or composition parameter shoots the present filming scene;
Preferably, the acquisition parameters include: filter, white balance, exposure compensating number and focal length, and the composition parameter includes: back
Scape, composition, personage's posture.
3. a kind of image processing method, which is characterized in that described image processing method is applied to server end, comprising:
Receive the first image about present filming scene that client is sent;
Semantic analysis is carried out to the first image;
According to the semantic analysis result, the similar reference picture of semanteme of search and the first image;And
At least one described reference picture is sent to the client.
4. image processing method according to claim 3, which is characterized in that described to carry out semantic point to the first image
Analysis, comprising: obtain the first eigenvector and second feature vector of the first image;
Preferably, described according to the semantic analysis result, the similar reference picture of semanteme of search and the first image, packet
It includes:
Obtain the first eigenvector of each prestored images;
Obtain the second feature vector of each prestored images;
Establish and store the mapping relations of each prestored images, the first eigenvector and the second feature vector, institute
The dimension for stating first eigenvector is greater than the dimension of the second feature vector;
The first image and multiple pre-stored images are subjected to aspect ratio pair;And
The reference picture of at least one the first image is obtained according to comparison result;
It is preferably, described by the first image and multiple pre-stored images carry out aspect ratio pair, comprising:
By multiple second feature vectors of the multiple prestored images of second feature vector sum of the first image respectively into
Row similitude compares;
At least one target image is obtained according to comparison result;And
The first eigenvector of the first image is similar with the progress of the first eigenvector of at least one target image
Degree compares;
At least one described reference picture is obtained according to comparison result;
Preferably, described at least one target image is obtained according to comparison result to include:
It is set if the distance of the second feature vector of the second feature vector sum the first image of a prestored images is less than
Threshold value is determined, using the prestored images as the target image;
It is described to include: according at least one described reference picture of comparison result acquisition
It is set if the distance of the first eigenvector of the first eigenvector and the first image of prestored images is less than
Threshold value is determined, using the prestored images as the reference picture;
Preferably, the first eigenvector is obtained according to the first model, the second feature vector is obtained according to the second model;
Preferably, first model is N layers of convolutional neural networks model, and second model is N+1 layers of convolutional neural networks
Model, second model increase a hidden layer between the N-1 layer of first model and N layers;
Preferably, described image processing method, further includes:
Establish training set and verifying collection;
According to the training set and verifying collection, first model and described is obtained based on the convolutional neural networks model training
Second model;
Preferably, described image processing method, further includes: the image concentrated to the training set and the verifying is located in advance
Reason;
Preferably, described image processing method, further includes: receive feedback information, and first mould is optimized according to feedback information
Type and second model.
5. a kind of image processing apparatus, which is characterized in that described image processing unit is applied to client, comprising:
First shooting module, for obtaining the first image about present filming scene;
First sending module, for sending the first image to server;
First receiving module, for receiving at least one reference picture about the first image, which is by institute
Server is stated according to the similar image with the first image semanteme got after the first image semantic analysis;
Second shooting module, for showing at least one described reference picture, in order to described at least one described in user's reference
Reference picture shoots the present filming scene;
Preferably, second shooting module is also used to show the shooting extracted at least one described described reference picture
Parameter and/or composition parameter, in order to which the user is described current with reference to the application acquisition parameters and/or the shooting of composition parameter
Photographed scene;
Preferably, the acquisition parameters include: filter, white balance, exposure compensating number and focal length, and the composition parameter includes: back
Scape, composition, personage's posture.
6. a kind of image processing apparatus, which is characterized in that described image processing unit is applied to server end, comprising:
Second receiving module, for receiving the first image about present filming scene of client transmission;
Semantic module, for carrying out semantic analysis to the first image;
Reference picture obtains module, similar with the semanteme of the first image for searching for according to the semantic analysis result
Reference picture;
Second sending module, for sending at least one described reference picture to the client;
Preferably, described that semantic analysis is carried out to the first image, comprising: to obtain the first eigenvector of the first image
With second feature vector.
Preferably, described according to the semantic analysis result, the similar reference picture of semanteme of search and the first image, packet
It includes: obtaining the first eigenvector of each prestored images;
Obtain the second feature vector of each prestored images;
Establish and store the mapping relations of each prestored images, the first eigenvector and the second feature vector, institute
The dimension for stating first eigenvector is greater than the dimension of the second feature vector;
The first image and multiple pre-stored images are subjected to aspect ratio pair;And
The reference picture of at least one the first image is obtained according to comparison result;
It is preferably, described by the first image and multiple pre-stored images carry out aspect ratio pair, comprising:
By multiple second feature vectors of the multiple prestored images of second feature vector sum of the first image respectively into
Row similitude compares;
At least one target image is obtained according to comparison result;And
The first eigenvector of the first image is similar with the progress of the first eigenvector of at least one target image
Degree compares;
At least one described reference picture is obtained according to comparison result;
Preferably, described at least one target image is obtained according to comparison result to include:
It is set if the distance of the second feature vector of the second feature vector sum the first image of a prestored images is less than
Threshold value is determined, using the prestored images as the target image;
It is described to include: according at least one described reference picture of comparison result acquisition
It is set if the distance of the first eigenvector of the first eigenvector and the first image of prestored images is less than
Threshold value is determined, using the prestored images as the reference picture;
Preferably, the first eigenvector is obtained according to the first model, the second feature vector is obtained according to the second model;
Preferably, first model is N layers of convolutional neural networks model, and second model is N+1 layers of convolutional neural networks
Model, second model increase a hidden layer between the N-1 layer of first model and N layers.
7. a kind of computer readable storage medium, which is characterized in that the computer-readable recording medium storage has computer to refer to
It enables, the computer instruction is performed realization such as the described in any item image processing methods of claim 1 to 2.
8. a kind of computer readable storage medium, which is characterized in that the computer-readable recording medium storage has computer to refer to
It enables, the computer instruction is performed realization such as the described in any item image processing methods of claim 3 to 4.
9. a kind of electronic equipment characterized by comprising
Processor;
Memory for storage processor executable instruction;
Wherein, the processor is configured to perform claim requires image processing method described in 1 to 2 any one.
10. a kind of electronic equipment characterized by comprising
Processor;
Memory for storage processor executable instruction;
Wherein, the processor is configured to perform claim requires image processing method described in 3 to 4 any one.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811416403.9A CN109639964A (en) | 2018-11-26 | 2018-11-26 | Image processing method, processing unit and computer readable storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811416403.9A CN109639964A (en) | 2018-11-26 | 2018-11-26 | Image processing method, processing unit and computer readable storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109639964A true CN109639964A (en) | 2019-04-16 |
Family
ID=66069133
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811416403.9A Pending CN109639964A (en) | 2018-11-26 | 2018-11-26 | Image processing method, processing unit and computer readable storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109639964A (en) |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110336945A (en) * | 2019-07-09 | 2019-10-15 | 上海泰大建筑科技有限公司 | A kind of intelligence assisted tomography patterning process and system |
CN110868543A (en) * | 2019-11-25 | 2020-03-06 | 三星电子(中国)研发中心 | Intelligent photographing method and device and computer readable storage medium |
CN111429377A (en) * | 2020-03-31 | 2020-07-17 | Oppo广东移动通信有限公司 | Image processing method, terminal and storage medium |
CN112383703A (en) * | 2020-10-23 | 2021-02-19 | 马上游科技股份有限公司 | Method and system for automatic framing and photographing |
CN112565599A (en) * | 2020-11-27 | 2021-03-26 | Oppo广东移动通信有限公司 | Image shooting method and device, electronic equipment, server and storage medium |
CN113411505A (en) * | 2021-08-19 | 2021-09-17 | 深圳康易世佳科技有限公司 | Photographing control method and device and storage medium |
CN113555088A (en) * | 2020-04-23 | 2021-10-26 | 上海联影智能医疗科技有限公司 | Image processing method, device, system and storage medium |
CN114827445A (en) * | 2021-01-29 | 2022-07-29 | 华为技术有限公司 | Image processing method and related device |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20160112630A1 (en) * | 2014-10-15 | 2016-04-21 | Microsoft Corporation | Camera capture recommendation for applications |
CN106909924A (en) * | 2017-02-18 | 2017-06-30 | 北京工业大学 | A kind of remote sensing image method for quickly retrieving based on depth conspicuousness |
CN107229625A (en) * | 2016-03-23 | 2017-10-03 | 北京搜狗科技发展有限公司 | It is a kind of to shoot treating method and apparatus, a kind of device for being used to shoot processing |
-
2018
- 2018-11-26 CN CN201811416403.9A patent/CN109639964A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20160112630A1 (en) * | 2014-10-15 | 2016-04-21 | Microsoft Corporation | Camera capture recommendation for applications |
CN107229625A (en) * | 2016-03-23 | 2017-10-03 | 北京搜狗科技发展有限公司 | It is a kind of to shoot treating method and apparatus, a kind of device for being used to shoot processing |
CN106909924A (en) * | 2017-02-18 | 2017-06-30 | 北京工业大学 | A kind of remote sensing image method for quickly retrieving based on depth conspicuousness |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110336945A (en) * | 2019-07-09 | 2019-10-15 | 上海泰大建筑科技有限公司 | A kind of intelligence assisted tomography patterning process and system |
CN110868543A (en) * | 2019-11-25 | 2020-03-06 | 三星电子(中国)研发中心 | Intelligent photographing method and device and computer readable storage medium |
CN111429377A (en) * | 2020-03-31 | 2020-07-17 | Oppo广东移动通信有限公司 | Image processing method, terminal and storage medium |
CN111429377B (en) * | 2020-03-31 | 2023-06-06 | Oppo广东移动通信有限公司 | Image processing method, terminal and storage medium |
CN113555088A (en) * | 2020-04-23 | 2021-10-26 | 上海联影智能医疗科技有限公司 | Image processing method, device, system and storage medium |
CN112383703A (en) * | 2020-10-23 | 2021-02-19 | 马上游科技股份有限公司 | Method and system for automatic framing and photographing |
CN112565599A (en) * | 2020-11-27 | 2021-03-26 | Oppo广东移动通信有限公司 | Image shooting method and device, electronic equipment, server and storage medium |
WO2022111072A1 (en) * | 2020-11-27 | 2022-06-02 | Oppo广东移动通信有限公司 | Image photographing method and apparatus, and electronic device, server and storage medium |
CN114827445A (en) * | 2021-01-29 | 2022-07-29 | 华为技术有限公司 | Image processing method and related device |
CN114827445B (en) * | 2021-01-29 | 2023-09-01 | 华为技术有限公司 | Image processing method and related device |
CN113411505A (en) * | 2021-08-19 | 2021-09-17 | 深圳康易世佳科技有限公司 | Photographing control method and device and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109639964A (en) | Image processing method, processing unit and computer readable storage medium | |
TWI775091B (en) | Data update method, electronic device and storage medium thereof | |
WO2021196401A1 (en) | Image reconstruction method and apparatus, electronic device and storage medium | |
WO2021031609A1 (en) | Living body detection method and device, electronic apparatus and storage medium | |
WO2018228168A1 (en) | Image processing method and related product | |
CN110188236A (en) | A kind of recommended method of music, apparatus and system | |
WO2016127671A1 (en) | Image filter generating method and device | |
US20210224607A1 (en) | Method and apparatus for neutral network training, method and apparatus for image generation, and storage medium | |
WO2017092127A1 (en) | Video classification method and apparatus | |
TWI757668B (en) | Network optimization method and device, image processing method and device, storage medium | |
CN106231152B (en) | Transmit the method and device of picture | |
TWI778313B (en) | Method and electronic equipment for image processing and storage medium thereof | |
CN109168062A (en) | Methods of exhibiting, device, terminal device and the storage medium of video playing | |
WO2020103353A1 (en) | Multi-beam selection method and device | |
WO2017219497A1 (en) | Message generation method and apparatus | |
CN108154466A (en) | Image processing method and device | |
CN113707134B (en) | Model training method and device for model training | |
CN112188091B (en) | Face information identification method and device, electronic equipment and storage medium | |
CN109034106B (en) | Face data cleaning method and device | |
CN109086752A (en) | Face identification method, device, electronic equipment and storage medium | |
CN108848303A (en) | Shoot reminding method and device | |
CN104573642B (en) | Face identification method and device | |
CN112669231B (en) | Image processing method, training method, device and medium of image processing model | |
CN113807498B (en) | Model expansion method and device, electronic equipment and storage medium | |
CN107480773B (en) | Method and device for training convolutional neural network model and storage medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190416 |