WO2019114508A1

WO2019114508A1 - Image processing method, apparatus, computer readable storage medium, and electronic device

Info

Publication number: WO2019114508A1
Application number: PCT/CN2018/116592
Authority: WO
Inventors: 达剑
Original assignee: Oppo广东移动通信有限公司
Priority date: 2017-12-13
Filing date: 2018-11-21
Publication date: 2019-06-20
Also published as: CN108108415B; CN108108415A

Abstract

An image processing method, comprising: acquiring faces in images; acquiring from among the images an image set featuring the same faces; recognizing age values corresponding to the faces in the images in the image set; according to the age values, generating sorting information for the images in the image set; and according to the sorting information and the image set, generating a time sequence photo album comprising the images in the image set.

Description

Image processing method, device, computer readable storage medium, and electronic device

Cross-reference to related applications

This application claims the priority of the Chinese Patent Application, filed on Dec. 13, 2017, filed on Jan. 13,,,,,,,,,,,,,,,,,,,,,,,,,,,,, In this application.

Technical field

The present application relates to the field of computer technology, and in particular, to an image processing method, a storage medium, and an electronic device.

Background technique

With the popularity of electronic devices and the rapid development of mobile Internet, the use of electronic devices is increasing. The photo album function has become one of the commonly used applications of electronic devices, and belongs to applications with extremely high user frequency. A large number of images are stored in the album of the electronic device, and the traditional electronic device album has the functions of providing various image browsing and classification.

In the process of browsing photos in an album, the traditional method is usually classified according to image types or image modification time or creation time, and it is difficult to motivate the user to sort the classified albums or image collections. The desire to view the photos, the user's viscosity is poor.

Summary of the invention

Embodiments of the present application provide an image processing method, apparatus, storage medium, and electronic device.

An image processing method comprising:

Get the face in the image;

Obtaining a set of images having the same face in the image;

Identifying an age value corresponding to a face of the image in the set of images;

Generating ranking information for images in the set of images based on the age values;

Based on the ranking information and the set of images, an album having a sequence of images in the set of images is generated.

An image processing apparatus, the apparatus comprising:

An image collection obtaining module, configured to acquire a human face in the image; and acquire an image collection having the same human face in the image;

An age value identification module, configured to identify an age value corresponding to a face of the image in the image set;

a sorting information generating module, configured to generate, according to the age value, sorting information of an image in the image set;

And an album generating module, configured to generate, according to the sorting information and the image set, a photo album having a sequence of images in the image set.

A computer readable storage medium having stored thereon a computer program that, when executed by a processor, implements the operations of the image processing method described in any of the embodiments herein.

An electronic device comprising a memory, a processor, and a computer program stored on the memory and operable on the processor, the processor executing the computer program to implement the operation of the image processing method according to any of the embodiments of the present application .

The above image processing method, apparatus, storage medium, and electronic device acquire a face in an image; acquire an image set having the same face in the image, and identify an age value corresponding to the face of the image in the image set, and then follow each The age value of the face in the image generates sorting information for the image, and according to the sorting information and the image set, generates a time-series album including the image in the image set, so that the image in the album can be displayed according to the time series, thereby improving The user's use of viscosity also increases the flexibility to classify and display images.

DRAWINGS

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings to be used in the embodiments or the prior art description will be briefly described below. Obviously, the drawings in the following description are only It is a certain embodiment of the present application, and other drawings can be obtained according to the drawings without any creative work for those skilled in the art.

FIG. 1 is an application scenario diagram of an image processing method in an embodiment.

2 is a schematic diagram showing the internal structure of an electronic device in an embodiment.

3 is a flow chart of an image processing method in one embodiment.

4A is a schematic diagram showing an image in an album in one embodiment.

4B is a schematic diagram showing an image in an album in another embodiment.

4C is a schematic diagram of dividing an image and displaying an image in an album in one embodiment.

Figure 5 is a flow diagram of forming an image collection in one embodiment.

6 is a flow diagram of determining a master face from a face area and/or a face definition occupied by each face in an image in one embodiment.

FIG. 7 is a diagram showing, in an embodiment, when it is recognized that a plurality of faces are included in an image, whether the remaining faces in the detected image correspond to existing user identities, and if yes, the remaining faces are classified into corresponding existing user identities. In the flowchart of obtaining the user identity corresponding to the remaining faces.

Figure 8 is a flow diagram of the processing of a suspected passerby face in one embodiment.

Figure 9 is a flow diagram of the division of image sets for new images in one embodiment.

Figure 10 is a block diagram showing the structure of an image processing apparatus in an embodiment.

Figure 11 is a block diagram showing the structure of an image processing apparatus in another embodiment.

Figure 12 is a schematic illustration of an image processing circuit in one embodiment.

Detailed ways

In order to make the objects, technical solutions, and advantages of the present application more comprehensible, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It is understood that the specific embodiments described herein are merely illustrative of the application and are not intended to be limiting.

In an embodiment, as shown in FIG. 1 , an application scenario diagram of an image processing method is provided, where the application environment includes an electronic device 110 and a server 120. The terminal 110 and the server 120 are connected through a network. An image is stored in the electronic device 110, and the image may be stored in the internal memory of the electronic device 110 or may be stored in an SD (Secure Digital Memory Card) card built into the electronic device 110. Alternatively, the server 120 may also store an image, and the image stored on the server 120 may be an image stored by the electronic device in the cloud. The electronic device can process images stored locally or in the cloud, such as forming a collection of images from the image, or sorting and displaying the images in the collection of images.

2 is a schematic diagram showing the internal structure of an electronic device in an embodiment. As shown in FIG. 2, the electronic device includes a processor, a memory, and a display screen connected by a system bus. The processor is used to provide computing and control capabilities to support the operation of the entire electronic device. The memory is used to store data, programs, etc., and at least one computer program is stored on the memory, and the computer program can be executed by the processor to implement an image processing method suitable for an electronic device provided in the embodiments of the present application. The memory may include a non-volatile storage medium such as a magnetic disk, an optical disk, a read-only memory (ROM), or a random storage memory (Random-Access-Memory, RAM). For example, in one embodiment, the memory includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The computer program can be executed by a processor for implementing an image processing method provided by the various embodiments below. The internal memory provides a cached operating environment for operating system computer programs in a non-volatile storage medium. The display screen can be a touch screen, such as a capacitive screen or an electronic screen, for displaying visual information such as an image, and can also be used to detect a touch operation applied to the display screen to generate a corresponding command. The electronic device can be a cell phone, a tablet or a personal digital assistant or a wearable device.

It will be understood by those skilled in the art that the structure shown in FIG. 2 is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation of the computer device to which the solution of the present application is applied. The specific computer device may It includes more or fewer components than those shown in the figures, or some components are combined, or have different component arrangements. For example, the electronic device can further include a camera through which an image can be generated and processed.

In an embodiment, as shown in FIG. 3, an image processing method is provided. This embodiment is mainly applied to the electronic device shown in FIG. 1 as an example. The method includes:

Operation 302, obtaining a face in the image.

Optionally, the image is an image including a human face, and the electronic device may acquire an image from a multi-user family shared album in the local device and/or other devices, and may also be an image shared by other devices, and the shared image is The multi-user family with access rights of the electronic device shares the image in the album, and recognizes and acquires the face in the image.

The electronic device may receive an image division instruction that may specify a corresponding one or more images and acquire the division instruction designation image. It is also possible to automatically acquire an image, such as an image that can be acquired in a state of being charged by the screen, or automatically acquire an image in a state where the resource usage rate on the electronic device is low and/or the power exceeds the preset power threshold. The resource usage includes, but is not limited to, one or more of CPU usage and memory usage. For the acquired image, the face in the image can be further obtained.

In one embodiment, when an image is added to the electronic device or the preset cloud, the face in the image is automatically acquired.

Operation 304, acquiring a set of images having the same face in the image.

Optionally, the image set is a set of images formed according to a face, and the image in each image set is an image containing the same face. The electronic device presets a correspondence between the face and the image set. According to the acquired face, the image set having the same face can be acquired from the image according to the preset correspondence.

In one embodiment, when it is detected that there is a change in the image contained in the image collection, the corresponding image collection may be automatically acquired. The change of the image includes an increase and/or decrease of the image, and may also include modifying the pixels in the existing image, such as performing beauty treatment on the existing image.

Operation 306, identifying an age value corresponding to a face of the image in the image collection.

The age value indicates the size of the age presented by the face in the electronically recognized image. Optionally, the electronic device may identify a face region of the face in the image in the image corresponding to each image in the image set, and identify feature data of the age-related portion of the face of the person. The part may include a plurality of pieces, and the analysis is performed according to the feature data of each part, and the age value corresponding to the face is calculated. Among them, the part includes a part such as a pupil, an eye corner, a corner of the mouth, a nose, and the like which obviously change with age. For example, a comprehensive analysis can be performed according to the characteristics of the skin texture, the pore size, the eye bag, the eye corner, and the like displayed on the image, and the age value corresponding to the face is calculated. As shown in FIG. 4A, the age value corresponding to the face in the images 1 to 4 can be identified according to the above-mentioned part, for example, the age value of the face in the image 1 in FIG. 4A is recognized as 8 years old, in the image 2 The age of the face is 16 years old, the age of the face in image 3 is 25 years old, and the age of the face in image 4 is 40 years old.

Operation 308, generating ranking information for the images in the image collection based on the age value.

Optionally, the sorting information represents information for sorting and displaying the images in the image set, and the sorting information includes an order of sorting each image in the image set in the image set, so that the order of the sorting may be performed according to the order of the sorting. Sort display. The sorting information may be sorted information according to age values from small to large, and may also be sorted information according to age values from large to small. Taking the sorting information from small to large according to the age value as an example, the smaller the age value in the image, the more the order of display of the corresponding image in the image set is in the process of playing or displaying the image in the image set. before.

Operation 310, based on the sorting information and the image set, generate a photo album having a sequence including images in the image set.

The electronic device can gather images of the same face in the image collection in one place to form a corresponding album. And based on the sorting information, the order in which the respective images are displayed in the album is generated, thereby forming an album having a time series. For example, the images in the image set may be sorted from large to small according to the age value of the same face, or sorted from large to small, and the generated time series is an album sorted by an age order or an age reversed image.

The image processing method described above, by acquiring a face in an image; acquiring an image set having the same face in the image, and identifying an age value corresponding to the face of the image in the image set, and then following the face of each image The age value generates sorting information for the image, and according to the sorting information and the image set, generates a time-series album including the image in the image set, so that the image in the album can be displayed according to the timing, providing a brand-new image classification And the way of viewing, the way to classify and view images is more, which improves the user's viscosity and improves the flexibility of classifying and displaying images.

In one embodiment, after operation 310, further comprising: playing the album with timing in a time series.

Optionally, the timing includes sequential ordering and/or reverse ordering according to age values, and the album with time series includes files of any form in a slide, a movie or an album. You can sort from large to small, or from small to large, depending on the age.

Optionally, the electronic device may receive a play instruction for the image in the album with time series, obtain the timing of the corresponding album according to the play instruction, and sequentially display the images in the album according to the timing. The electronic device can provide a virtual button for playing an image in the album, which can be triggered when a click operation on the virtual button is received. It is also possible to preset an open voice message for triggering a play command. The corresponding voice information is received by calling the voice receiving device, and when the voice information is detected to match the open voice information, the play command may also be triggered.

After receiving the play instruction for the image in the album, the timing of the pre-generated album may be acquired, and the images in the album are sequentially displayed according to the timing. For example, the image in the album may be played in a slide show according to the play command, or a short film or album may be generated for the user to view.

As shown in FIG. 4A and FIG. 4B, the album 1 includes images 1 to 4; and the album 2 includes images 5 to 7. Each album is also set with a corresponding timing. For example, the timing of the images in the album 1 is sorted in the order of the images 1 to 4, and the album 2 is sorted in the order of the images 5 to 7. The electronic device can play the images in the album with time series according to the sorting to show the evolution of the characters in an album from small to large or from large to small.

In one embodiment, operation 306 includes generating ranking information for the plurality of images based on the generation times of the plurality of images when the age values of the faces are the same among the plurality of images of the same image set.

Optionally, when there are multiple images, when the age values of the faces are the same, the generation time of the multiple images may be further compared. The plurality of images are sorted according to the generation time. For example, it can be sorted according to the generation time from small to large, or from large to small.

In one embodiment, when the ranking information is sorting information according to the age value from small to large, the images for the same age value are sorted from morning to night according to the generation time; when the sorting information is according to the age value from large to small When sorting information, the images of the same age value are sorted from late to early according to the generation time. Taking the sorting information as the sorting information according to the age value from small to large as an example, when there are three images, wherein the face is the face of the user A, and the age value of the face of the user A is determined to be 25 years old. Then, the size of the generation time of the three images can be further compared, and the sorting information of the three images is generated. And for the three images, the image with the earlier generation time is sorted first.

The ordering of image sorting is further improved by sorting images of the same age value according to the generation time.

In one embodiment, the above method further includes an operation of forming an image set, the operation being performed prior to operation 304, as shown in FIG. 5, comprising:

Operation 502, performing face recognition on the face in the image, and determining a user identity corresponding to the face in the image.

The electronic device can perform face recognition on the image, extract face feature information therein, and identify the user identity of the face according to the face feature information. For example, the electronic device may identify, for the image A, a face in which the face belongs to the user A, and the user identity is the user A.

In an embodiment, for the facial feature information extracted from the image, the facial feature information may be further detected to match the face feature information of the existing identified identity of the user, and when matching, It is determined that the user identity corresponding to the face in the image is the matched user identity. When there is no user identity matching the facial feature information, it indicates that the user identity is a newly appearing user identity. A new user identity can be created correspondingly as the user identity corresponding to the face in the image. And generating facial feature information corresponding to the newly created user identity according to the facial feature information of the face in the image. For example, the face feature information in the image can be directly used as the face feature information corresponding to the user identity.

Operation 504, dividing the image into an image collection corresponding to the user identity to form an image collection having the same face.

The electronic device sets a corresponding image set for each user identity, and the images in the image set are images having the same user identity. The electronic device can establish a mapping relationship between different user identities and different image sets, so that images of the same user identity are divided into the same image set according to the corresponding relationship to form an image set having the same face.

In one embodiment, for an image of the identified user identity, a user identity tag can be set thereon to mark the user identity of the face in the image, and such that in subsequent processing, the flag can be determined The image is an image that has identified the user's identity, avoiding duplicate recognition. And according to the user identity mark, the image is divided into an image set corresponding to the user identity mark.

In an embodiment, when the identified user identity is an existing user identity, an image set corresponding to the existing user identity may be acquired, and the image is divided into corresponding image sets. When the identified user identity is a new user identity, a set of images corresponding to the new user identity is created, and the image is divided into the created image collection.

For example, when the user identity of the recognized image is the existing user A, the image set A corresponding to the user A is directly acquired, and the image is divided into the image set A. When the identified user A is a non-existing user, an image set A corresponding to the user A is created, and the image is divided into the image set A, so that the images in the image set A are all included in the user A. The image of the face. For example, there is an image 8 as shown in FIG. 4C. When it is recognized that the user identity corresponding to the face in the image 8 is the user identity corresponding to the image set as shown in FIG. 4B, the image 8 may be divided into the corresponding FIG. 4B. In the collection of images. And in the divided image set, the age value corresponding to the face in each image generates sorting information for the image, and generates an album with time series and displays. For example, the order may be sorted according to the age value. For example, when it is recognized that the age value corresponding to the face in the image 8 is between the image 6 and the image 7, the sort order of the image 8 is located in the image. Between 6 and image 7, a time-stamped album of image 5-image 6-image 8-image 7 as shown in Fig. 4C is generated.

In the above embodiment, by identifying the face in the image, the user to which the face belongs is determined, and then the image is divided into an image set corresponding to the user, thereby realizing the identity of the user to which the image face belongs. The image division divides the image of the face containing the same user into the same image set, which improves the flexibility of image division.

In one embodiment, operation 302 includes determining a master face of the plurality of faces when the image includes a plurality of faces.

The face in the image may include a plurality of, and when it is recognized that the face includes a plurality of faces, it may further identify which of the faces is the master face in the image. The master face of each image refers to the face of the core person in each image, and the resolution reaches a threshold value to identify the master face of the identity corresponding to the face. The core character refers to the person corresponding to the owner's face in the image or the character corresponding to the owner's face in the multi-person group image, or the number of times the image is not the master face but appears on the same image as the owner's face belonging to a different user identity. The person corresponding to the face whose threshold is set.

In one embodiment, only one face is extracted as a master face for each image. Face recognition is performed on the master face of each image obtained according to the face recognition algorithm, and the face category corresponding to the owner face of each image is obtained. The face category refers to the identity of the person corresponding to the result of the master's face recognition. Optionally, the electronic device obtains a single photo from the album, and the photo is an image, and the owner's face is extracted from the single photo. Because it is a single photo, there is only one face in the image. When the clarity of the single face has reached the threshold that can recognize the face, the face in the single photo is marked as the main face. After the face recognition of the owner's face, it is found that the identity corresponding to the owner's face is Zhang San, that is, the user identity corresponding to the owner's face is Zhang San. For other single-person or non-single-person images, the face of the core character is extracted and the sharpness reaches the threshold, and the face corresponding to the identity of the face can be identified, marked as the main face, and the face of the master face is recognized, and the owner is obtained. The identity corresponding to the face, which is the face category corresponding to the owner's face.

When the sharpness of the face of the core person in an image does not reach the sharpness threshold of the identity corresponding to the face, continue to check whether other faces in the album have similar faces that have been marked as the face of the face. Exist, when present, mark this face and the owner's face as a type of master face. Conversely, if there is a preset number of faces in the other images in the album that are similar to the faces of the core characters in the image, then the clearest one is selected as the master face from the similar faces. Marking these similar faces as a type of master face, and performing face recognition on the master face, the identity corresponding to the master face is obtained, and the identity is the user identity corresponding to the master face, and is also the core in the image. The user identity of the person's face. The preset number of times can be set to 5 times. Of course, in other embodiments, other reasonable times can be set, for example, 3 times, 4 times, 6 times, 10 times, and the like.

In one embodiment, the master face may also include a plurality of, when a plurality of faces are included, one of the master faces in the image may be acquired. The obtained master face may be determined according to a user's selection operation.

In one embodiment, for each master face, the identity of the user to which the corresponding face belongs is identified, and the same photo may be divided into an image set corresponding to the user identity of each master face. For example, when there is a face A and a face B in a photo, the face detection can be performed in the above manner, and when the face A and the face B are both the master face, the person can be further identified. The user identity A corresponding to the face A and the user identity B corresponding to the face B are simultaneously divided into the image set A corresponding to the user A and the image set B corresponding to the user.

In one embodiment, determining a master face of the plurality of faces includes determining a master face based on a face area and/or a face definition occupied by each face in the image.

The electronic device obtains an image from a local or cloud album containing images of no human face (pure natural scenery), single-person images, and multi-person group images, including single-person images and multi-person group images. human face. For a single photo, there is only one face in the image. When the clarity of the single face has reached the threshold that can recognize the face, the face in the single photo is marked as the main face. For a multi-person group image, the master face can be determined according to a combination of any one or more of a face area and/or a face definition of the face in the image.

Optionally, the face with the largest face area is obtained from the multi-person group image, and it is determined whether the clarity of the face conforms to the clarity threshold of the recognized face, and whether the face meets the preset condition. If yes, the face with the largest face area is used as the master face of the multi-person group image. If not, the face with the second largest face size is continuously obtained from the multi-person group image, and it is determined whether the face meets the preset condition. Repeat until the face area reaches the minimum threshold or gets the owner's face. The master face is obtained in order from the largest to the smallest, so that the accuracy of the master face can be improved.

For multi-person group images, the master face can also be obtained in descending order of face clarity. Optionally, the face with the largest face resolution is obtained from the multi-person group image, and the face area of the face is determined to meet the face area threshold of the recognized face, and whether the face meets the preset condition. When yes, the face with the largest face resolution is used as the master face of the multi-person group image. If not, the face of the face resolution second is continuously obtained from the multi-person group image, and the face is judged to be in conformity. The preset condition is repeated until the face resolution reaches the minimum definition threshold or the owner's face is acquired.

In one embodiment, as shown in FIG. 6, the master face is determined based on the face area and/or face definition occupied by each face in the image, including:

Operation 602, obtaining a face with the largest face area from the image, determining whether the face meets the preset condition, and if so, performing operation 604, otherwise, performing operation 606.

When the image is a multi-person group image, the face of the core person is obtained from the multi-person group image, and the resolution reaches a threshold value to identify the master face of the identity corresponding to the face. Optionally, the face with the largest face area is obtained from the multi-person group image, and then the face is determined to meet the preset condition. The preset conditions include, but are not limited to, the following: determining whether the sharpness of the face reaches a threshold value that can identify the identity corresponding to the face, and when it is further determining the shooting angle and focus of the face with the largest face area, determining whether the shooting angle is It is a positive face, and it is judged whether the shooting focus is the face with the largest face area, and it is judged whether the face with the largest face area is closest to the lens.

In operation 604, the face with the largest face area is taken as the master face of the image.

In operation 606, the face of the face area is obtained from the image, and the face is determined to meet the preset condition, and the execution is repeated until the face area reaches the minimum threshold or the owner's face is acquired.

When the judgment result is that the above preset condition is satisfied at the same time, the face with the largest face area should be the face corresponding to the core person, and the face with the largest face area is used as the master face of the multi-person group image and marked. For the owner's face.

When the judgment result is that the preset condition is not satisfied, it indicates that the face with the largest face area is not the face corresponding to the core person, and then the face whose face size is second largest is obtained from the multi-person group image, and the face is judged. Whether the pre-set condition is met, and when the judgment result satisfies the above-mentioned preset condition, the face whose face area is second should be the face corresponding to the core person, so that the face with the face area second is taken as a multi-person group photo The owner's face of the image is marked as the main face.

When the face whose face area is second is not in accordance with the preset condition, it means that the face whose face area is second is not the face corresponding to the core person. Continue to obtain the face whose face size is again from the multi-person group image, and repeat until the face reaches the minimum threshold or gets the owner's face. That is, a minimum threshold of a face area is preset. When a face whose face area is larger than the minimum threshold is not found, the face that meets the preset condition is not found, then the owner's face is stopped for the image.

In one embodiment, determining the master face according to the face area and/or the face definition occupied by each face in the image comprises: obtaining the face with the highest face resolution from the image, and determining whether the face conforms to the preset. Condition; if yes, the face with the highest face resolution is used as the master face of the image; if not, the face whose face is the second most clear is obtained from the image, and it is determined whether the face meets the preset condition and is repeatedly executed. Until the face resolution reaches the minimum sharpness threshold or gets the owner's face.

When the image is a multi-person group image, the face of the core person is obtained from the multi-person group image, and the face resolution reaches a threshold value to identify the master face of the identity corresponding to the face. Optionally, the face with the largest facial expression is obtained from the multi-person group image, and then the face is determined to meet the preset condition. The preset conditions include, but are not limited to, the following: determining whether the sharpness of the face reaches a threshold value that can identify the identity corresponding to the face, and if it is further determining the shooting angle and focus of the face with the largest face area, determining whether the shooting angle is positive The face determines whether the shooting focus is the face with the largest face area, and determines whether the face with the largest face area is closest to the lens.

When the judgment result is that the above preset condition is satisfied at the same time, the face with the highest face resolution should be the face corresponding to the core person, and the face with the highest face resolution is used as the master face of the multi-person group image, and Mark the face as the main face.

When the judgment result is that the above preset condition is not satisfied, it indicates that the face with the largest face resolution is not the face corresponding to the core person, and then the face with the face clarity is obtained from the multi-person group image, and the person is judged. Whether the face meets the preset condition, and when the judgment result is that the above-mentioned preset condition is satisfied at the same time, the face whose face resolution is second should be the face corresponding to the core character, so that the face with the second face of the face is used as the face Multiplayer takes a picture of the owner's face and marks it as the main face.

When the face whose face resolution is second is not in accordance with the preset condition, the face whose face resolution is second is not the face corresponding to the core character. Continue to obtain the face whose face resolution is again from the multi-person group image, and repeat until the face reaches the minimum definition threshold or the owner's face is acquired. That is, a minimum threshold of the face sharpness is preset. When a face whose face definition is greater than the minimum sharpness threshold is not found, the face that meets the preset condition is not found, then the owner of the image is stopped. face.

In the embodiment of the present application, since it is more difficult to determine the master face from the multi-person group image than the single-person image, it is determined whether the face area and/or the face definition are in descending order. Face, so that the master's face will not be missed. And the preset condition is set, and only the face that satisfies these preset conditions can become the master face, and the result of obtaining the master face is more accurate.

In an embodiment, the operation 602 includes: obtaining a face with the largest face area from the image, and determining whether the angle and the focus of the face meet the preset condition.

In the embodiment of the present application, the face of the core person is obtained from the multi-person group image, and the resolution reaches a threshold to identify the owner face of the identity corresponding to the face. Optionally, the face with the largest face area is obtained from the multi-person group image, and then the face is determined to meet the preset condition. The preset conditions include, but are not limited to, the following: determining whether the sharpness of the face reaches a threshold value that can identify the identity corresponding to the face, and if it is further determining the shooting angle and focus of the face with the largest face area, determining whether the shooting angle is positive The face determines whether the shooting focus is the face with the largest face area, and determines whether the face with the largest face area is closest to the lens. Only the face that meets these preset conditions can become the master's face, and the result of the acquired master face is more accurate.

In an embodiment, the operation 504 includes: when it is recognized that the image includes multiple faces, detecting whether the remaining faces in the image correspond to existing user identities, and if so, dividing the remaining faces into corresponding existing ones. In the user identity, the user identity corresponding to the remaining faces is obtained. The operation 506 includes: dividing the image into an image set corresponding to the user identity corresponding to the remaining face, and using the image set corresponding to the user identity corresponding to the remaining face as the image set.

After obtaining the master face of each image from the image, the remaining faces in each image are acquired. The remaining faces include the remaining faces of each image except for the faces marked as the main face and the suspected passerby face. There is no remaining face for a single-person image, but there are remaining faces for a multi-person group image. For the processing of the remaining faces, it is only necessary to divide the remaining faces into the existing user identity according to the recognition result obtained by performing face recognition on the remaining faces, and the user identity is not newly added for the remaining faces. The existing user identity is the user identity corresponding to the face of the core character of each image, so the remaining faces that can be separated into the existing user identity are naturally also the faces of the core characters, thereby achieving non-core The faces corresponding to the characters are assigned to these user identities. For example, the album contains 1000 images, and each image extracts only one face as the master face. Then, up to 1000 master faces are generated from the 1000 images, and even if many of the 1000 images are multi-person group images, only a maximum of 1000 master faces are generated. Then, the face recognition of the 1000 master faces is performed, and the identity corresponding to the master face is obtained. The identity is the user identity corresponding to the master face. For example, after the recognition, the 1000 master faces belong to 10 user identities respectively. The remaining faces are divided into existing user identities to obtain the user identities corresponding to the remaining faces. When some of the remaining faces cannot be separated into the existing user identity, the remaining faces are marked as suspected passers-by faces.

The master face and the remaining faces in the image are classified by the above operations, and the corresponding user identity is assigned to those that can be classified. Then, according to the identity of the user that the master face and the remaining faces in the image can be separated, the image is divided into image sets corresponding to the user identity. Specifically, if the owner's face and the remaining faces of an image are respectively classified into three categories, the image will appear in the three categories at the same time. For example, an image consists of three faces, namely Zhang San, Li Si, and Wang Wu. Among them, the owner's face of this image is Zhang San, and the remaining two faces find a similar master face in the existing master's face, so the two remaining faces are divided into the user's identity of Li Si and Wang Wu. in. Then the image will eventually be displayed in the Zhang San image collection, and will also be displayed in the Li Si image collection, and of course the image will be displayed in the Wang Wu image collection. Because the characters corresponding to these three identities are the core figures of the owner's face.

In the embodiment of the present application, first, the owner's face of the image is acquired, and then the face of the master is recognized by the face, and the user identity corresponding to the owner's face is obtained. Since the master's face is the face of the core character in each image, the face of the non-core character does not act as the master face, so the user identity only includes the classification of the face of the core character in the image. Then, the remaining faces are obtained from the image, the face of the remaining faces in the image is recognized, and the remaining faces are divided into the existing user identity to obtain the user identity corresponding to the remaining faces. According to the identity of the user that can be separated by the owner's face and the remaining faces in the image, the image is divided into image collections corresponding to the user's identity. The image collection is distinguished by the user identity that can be separated by the master face and the remaining faces in the image. Because the user identity only includes the classification of the face of the core character in the image, naturally there will not be a large number of non-cores. A collection of images of people.

In an embodiment, as shown in FIG. 7, when it is recognized that the image includes multiple faces, whether the remaining faces in the image are detected to correspond to the existing user identity, and if so, the remaining faces are classified into corresponding In some user identities, the user identity corresponding to the remaining faces is obtained, including:

In operation 702, the remaining faces in the image are acquired, and the remaining faces are sequentially recognized by the face.

After obtaining the master face of each image from the multi-person group image, the remaining faces in each image are acquired. The remaining faces include the remaining faces of each image except for the faces marked as the main face and the suspected passerby face. For the remaining faces, face recognition is performed on the remaining faces in each image in turn, and the identity corresponding to the remaining faces is obtained. The remaining faces are divided into existing user identities to obtain the user identities corresponding to the remaining faces. When some of the remaining faces cannot be separated into the existing user identity, the remaining faces are marked as suspected passers-by faces.

Operation 704, determining whether the remaining faces belong to the existing user identity, and if so, performing operation 706, otherwise, performing operation 708.

In operation 706, the remaining faces are divided into user identities, and the user identity corresponding to the remaining faces is obtained.

Operation 708, marking the remaining faces as suspected passers-by faces.

Optionally, it is determined whether there is an existing user identity in the remaining faces. According to the identified identity of the remaining faces, it is determined whether the identity corresponding to the remaining faces is the same as the existing owner's face identity. When the same, the remaining faces are assigned to the corresponding user identity, and the user identity is the remaining face. The corresponding user identity.

When the identity corresponding to the remaining faces does not have the same identity in the existing master face, it indicates that the remaining faces do not belong to the existing user identity, and then the remaining faces are marked as suspected passers-by faces. And the suspected passerby faces of the same identity are marked as one type.

In the embodiment of the present application, for the classification process of the remaining faces, the remaining faces are classified into the existing user identity according to the identity result obtained by performing face recognition on the remaining faces, and the remaining faces are not newly added. Increase the user's identity, so naturally the face corresponding to the non-core person will not be divided into a user identity. In this way, only the image collection of the core person will be displayed after the image is classified. It avoids the image collection of a large number of non-core characters in the traditional method, does not meet the actual needs of users, and reduces the effectiveness of information.

In one embodiment, as shown in FIG. 8, the above method further includes an operation of processing the suspected passerby face, the operation being performed after marking the remaining face as a suspected passerby face, including:

Operation 802 calculates the number of times the suspected passerby face and the owner's face belonging to different user identities appear on the same image in the album.

When the identity corresponding to the remaining faces does not have the same identity in the existing master face, it indicates that the remaining faces do not belong to the existing user identity, and then the remaining faces are marked as suspected passers-by faces. After marking the remaining faces as suspected passers-by faces, the number of times each suspected passerby face and the owner face belonging to a different user identity appear on the same image are calculated in all the images in the album, that is, the same identity is calculated. The number of times the suspected passerby face was photographed with the owner's face of different identities. For example, the number of times the suspected passerby face of the identity A is photographed with the owner's face of different identities is calculated. Specifically, it is assumed that the owner's face has 10 master faces such as Zhang San, Li Si, and Wang Wu. Then count the number of times the suspected passerby face with the identity A is taken with any of the 10 master faces. There is an image in the album that is a group photo of A and Zhang San, Li Si and Wang Wu. Then the number of suspected passers-by faces with A is taken together with the owner's face of different identities three times. When there is another image in the album, it is a photo of A and Zhang San. One image is a photo of A and Wang Wu. Then, the number of suspected passers-by faces with the identity A and the owner's face of different identities at this time is 5 times.

Operation 804, determining whether the number of times has reached a set threshold, and if so, performing operation 806, otherwise, performing operation 808.

The threshold value may be set to a minimum of 5 times when the suspected passerby face and the owner's face belonging to different user identities appear on the same image in the album. Of course, in other embodiments, other reasonable ones may be set. The number of times, for example, 6 times, 7 times, 8 times, 9 times, 10 times, and the like.

At operation 806, the suspected passerby face is added as the main face, and the user identity corresponding to the suspected passerby face is obtained.

Operation 808, keeping the suspected passerby face.

Judging whether the number of times the suspected passerby face and the owner's face belonging to different user identities appear on the same image has reached a set threshold (for example, 5 times), and when it reaches 5 times, the suspected passerby face is added as the main face. Get the identity of the user corresponding to the suspected passerby face. In this embodiment, the number of suspected passers-by faces with the identity A and the owner's face of different identities are taken 5 times, and the suspected passerby face with the identity A is added as the main face, and the suspected passerby is obtained. The user identity corresponding to the face. When it has not reached 5 times, it will remain as a suspected passerby face. Waiting for the new image in the next album to recalculate the number of times the suspected passerby face appears on the same image as the owner's face belonging to a different user identity.

In the embodiment of the present application, for each type of suspected passerby face in the album, the number of times each suspected passerby face and the owner face belonging to a different user identity appear on the same image are calculated. When the number of times reaches the set threshold, the suspected passerby face is added as the main face, and the user identity corresponding to the suspected passerby face is obtained. The suspected passerby face in the album has also been processed so that it can be added to the main face, which effectively avoids extracting only one face as the master face for each image, resulting in missing core characters. .

In one embodiment, as shown in FIG. 9, the above method further includes an operation of dividing the image set of the newly added image, and the operation may be performed according to the identity of the user that can be separated according to the owner's face and the remaining faces in the image. The image is executed after being divided into the image collection corresponding to the user identity, including:

Operation 902, obtaining a new image.

The images in the album are not static, but are dynamically changed. For example, they can be added as the user takes photos continuously, or they can download photos from the cloud to the album. Or the album is originally in the cloud, the image in the cloud album has been added. Get these new images.

In operation 904, the newly added image is classified according to the image processing method before the newly added image, and the newly added image is divided into the image corresponding to the user identity according to the user identity that the owner face and the remaining face in the newly added image can be separated. set.

The new image is classified according to the image processing method before the newly added image. Specifically, the owner's face is first acquired from the newly added image. When the newly added image is a single-person image, the only one face in the image is acquired, and this is determined. Whether the definition of the face has reached the threshold that can recognize the face. When the definition of the face of the single person has reached a threshold value for recognizing the face, the face in the single-person photo is marked as the main face, and after the face recognition is performed on the master face, the master face is obtained. The corresponding user identity.

When the clarity of the single face does not reach the sharpness threshold of the identity corresponding to the face, then continue to check whether other faces in the album have similar faces that have been marked as the face of the face, when present , the single person face and the owner's face are marked as a master face. Conversely, if there is a preset number of faces similar to the single face in other images in the album, then the most clear one is selected as the master face from the similar faces, and the similar persons are The face is marked as a type of master face, and the face of the master face is recognized, and the identity corresponding to the master face is obtained, and the identity is the user identity corresponding to the single face.

When the clarity of the single person's face does not reach the sharpness threshold of the identity corresponding to the face, and there are no other faces in the album that are already similar and have been marked as the face of the face, and the album The number of faces in the other images in which the face is similar to the single face has not reached the preset number of times. It is indicated that the blurred single person face is not the face of the core person, and the fuzzy single face that satisfies the above three conditions is marked as a suspected passerby face.

When the newly added image is a multi-person group image, the master face is acquired from the multi-person group image in descending order of the face area, and is repeatedly executed until the face reaches the minimum threshold or the owner's face is acquired. Obtain the remaining faces in the newly added image, perform face recognition on the remaining faces, and divide the remaining faces into the existing user identity to obtain the user identity corresponding to the remaining faces. When some of the remaining faces cannot be separated into the existing user identity, the remaining faces are marked as suspected passers-by faces.

According to the identity of the user that can be separated by the owner's face and the remaining faces in the image, the image is divided into image collections corresponding to the user's identity.

Operation 906, recalculating the number of times the suspected passerby face and the owner's face belonging to different user identities appear on the same image.

Since the image is newly added, it is necessary to recalculate the number of times the suspected passerby face and the owner's face belonging to different user identities appear on the same image.

Operation 908, determining whether the number of times has reached a set threshold, and if so, performing operation 910; otherwise, performing operation 912.

In operation 910, the suspected passerby face is added as the main face, and the user identity corresponding to the suspected passerby face is obtained, and the image containing the suspected passerby face is divided into the image set corresponding to the user identity.

Operation 912 remains as a suspected passerby face.

Judging whether the number of suspected passers-by faces and the owner's face belonging to different user identities appear on the same image reaches the set threshold. When the set threshold is reached, the suspected passerby face is added as the main face, and the suspected passerby face is obtained. The corresponding user identity is used to divide the image containing the suspected passerby face into the image set corresponding to the user identity. When the set threshold is not reached, it remains as a suspected passerby face. When the new image is added in the next album, the suspected passerby face is recalculated to the number of times the owner's face belonging to a different user identity appears on the same image.

In the embodiment of the present application, when the image is added, the new image is classified according to the image processing method before the new image is added, and the user identity that can be separated by the owner face and the remaining face in the newly added image is used. The added image is divided into a collection of images corresponding to the user's identity. Then recalculate the number of times the suspected passerby face and the owner's face belonging to different user identities appear on the same image. Because of the addition of images, the number of suspected passers-by faces and the owner's face belonging to different user identities may have been updated on the same image, that is, the suspected passerby face may also become the master's face. Therefore, it is possible to update the suspected passerby face to the main face in real time by recalculating.

In an embodiment, an image processing method is also provided, which is applied to the electronic device in FIG. 1 as an example, and includes:

(1) The electronic device obtains the face of the image from a local or cloud album.

(2) Get the owner's face from the image. This includes obtaining the master's face from a single-person image and a multi-person group photo. In the multi-person group image, the master face is obtained in order of the face area from large to small. The face with the largest face area is obtained from the multi-person group image, and it is determined whether the face meets the preset condition. When yes, the face with the largest face area is used as the master face of the multi-person group image; if not, the face with the second largest face size is continuously obtained from the multi-person group image, and it is determined whether the face meets the preset condition. Repeat until the face area reaches the minimum threshold or gets the owner's face.

(3) Perform face recognition on the owner's face of the image to obtain the user identity corresponding to the owner's face.

(4) Acquiring the remaining faces in the multi-person group image, and performing face recognition on the remaining faces in turn. Determine whether the remaining faces belong to the existing user identity. If yes, divide the remaining faces into the user identity to obtain the user identity corresponding to the remaining faces. If not, mark the remaining faces as suspected passers-by faces.

(5) Calculate the number of times the suspected passerby face and the owner's face belonging to different user identities appear on the same image in the album. Whether the number of judgments reaches the set threshold; if so, the suspected passerby face is added as the main face, and the user identity corresponding to the suspected passerby face is obtained; if not, the suspected passerby face is maintained.

(6) According to the identity of the user that can be separated by the master face and the remaining faces in the image, the image is divided into image sets corresponding to the user identity.

(7) Obtain the face of the newly added image.

(8) classify the newly added image according to the image processing method before the newly added image, and divide the newly added image into the image corresponding to the user identity according to the identity of the user that can be separated by the owner face and the remaining faces in the newly added image. set. Recalculating the number of times the suspected passerby face and the owner's face belonging to different user identities appear on the same image; whether the number of judgments reaches the set threshold; if so, adding the suspected passerby face to the main face, and obtaining the suspected passerby face The corresponding user identity divides the image containing the suspected passerby face into the image set corresponding to the user identity, and if not, remains as the suspected passerby face.

(9) Acquiring an image set having the same face in the image; and identifying an age value corresponding to the face of the image in the image set.

(10) generating sorting information for the image in the image set based on the age value; generating a time-series album including the image in the image set based on the sorting information and the image set.

(11) Play the album with time series in time series.

The above image processing method divides each image into an image set corresponding to the user's identity of the face in which the image is classified according to the user identity to which the face belongs. And acquiring an image set having the same face; identifying an age value corresponding to the face of the image in the image set; generating sorting information for the image in the image set according to the age value; generating the included image set according to the sorting information and the image set The image has a sequence of albums, and the time-stamped albums are played in time series, which is convenient for helping users to record image stories of different ages and increasing the viscosity of the user.

In one embodiment, as shown in FIG. 10, an image processing apparatus is provided, the apparatus comprising:

The image set obtaining module 1002 is configured to acquire a face in the image; and acquire an image set having the same face in the image.

The age value identification module 1004 is configured to identify an age value corresponding to a face of the image in the image collection.

The sorting information generating module 1006 is configured to generate sorting information for the images in the image set according to the age value.

The album generating module 1008 is configured to generate a time-series album including images in the image collection according to the sorting information and the image set.

In an embodiment, the image set obtaining module 1002 is further configured to perform face recognition on a face in the image, determine a user identity corresponding to the face in the image, and divide the image into an image set corresponding to the user identity to form an image set. A collection of images with the same face.

In an embodiment, the image set obtaining module 1002 is further configured to determine a master face of the plurality of faces when the image includes multiple faces.

In one embodiment, the image set acquisition module 1002 is further configured to determine a master face based on a face area and/or a face definition occupied by each face in the image.

In an embodiment, the image set obtaining module 1002 is further configured to obtain a face with the largest face area from the image, and determine whether the face meets the preset condition; if yes, the face with the largest face area is used as the image owner. Face; if not, continue to obtain the face whose face area is the second largest from the image, determine whether the face meets the preset condition, and repeat until the face area reaches the minimum area threshold or obtains the owner's face.

In an embodiment, the image set obtaining module 1002 is further configured to obtain a face with the highest face resolution from the image, and determine whether the face meets the preset condition; if yes, the face with the highest face resolution is used as the image. The master's face; if not, continue to obtain the face whose face is the second most clear from the image, determine whether the face meets the preset condition, and repeat until the face resolution reaches the minimum sharpness threshold or obtains the master's face. .

In one embodiment, the ranking information generating module 1006 is further configured to generate, when the age values of the faces are the same among the plurality of images in the same image set, the ranking information for the plurality of images according to the generation time of the plurality of images.

In one embodiment, as shown in FIG. 11, yet another image processing apparatus is provided, the apparatus further comprising:

The playing module 1010 is configured to play the album with time series in time series, the sequence includes sorting by the age value and/or sorting in reverse order, and the album with the time series includes any one of a slide, a movie or an album.

In one embodiment, the image collection acquisition module 1002 is further for acquiring images from a multi-user family shared album in the local machine and/or other devices.

The division of each module in the above image processing apparatus is for illustrative purposes only. In other embodiments, the image processing apparatus may be divided into different modules as needed to complete all or part of the functions of the image processing apparatus.

In one embodiment, a computer readable storage medium is provided having stored thereon a computer program that, when executed by a processor, implements the operations of the image processing methods provided by the various embodiments described above.

An electronic device comprising a memory, a processor, and a computer program stored on the memory and operable on the processor, the processor performing the operations of the image processing methods provided by the various embodiments described above when executing the computer program.

The embodiment of the present application also provides a computer program product. A computer program product comprising instructions which, when run on a computer, cause the computer to perform the operations of the image processing methods provided by the various embodiments described above.

An embodiment of the present application further provides an electronic device. The above electronic device includes an image processing circuit, and the image processing circuit can be implemented by using hardware and/or software components, and can include various processing units defining an ISP (Image Signal Processing) pipeline. Figure 12 is a schematic illustration of an image processing circuit in one embodiment. As shown in FIG. 12, for convenience of explanation, only various aspects of the image processing technique related to the embodiment of the present application are shown.

As shown in FIG. 12, the image processing circuit includes an ISP processor 1240 and a control logic 1250. The image data captured by imaging device 1210 is first processed by ISP processor 1240, which analyzes the image data to capture image statistics that can be used to determine and/or control one or more control parameters of imaging device 1210. Imaging device 1210 can include a camera having one or more lenses 1212 and image sensors 1214. Image sensor 1214 may include a color filter array (such as a Bayer filter) that may acquire light intensity and wavelength information captured with each imaging pixel of image sensor 1214 and provide a set of primitives that may be processed by ISP processor 1240 Image data. A sensor 1220, such as a gyroscope, can provide acquired image processing parameters, such as anti-shake parameters, to the ISP processor 1240 based on the sensor 1220 interface type. The sensor 1220 interface may utilize a SMIA (Standard Mobile Imaging Architecture) interface, other serial or parallel camera interfaces, or a combination of the above.

In addition, image sensor 1214 can also transmit raw image data to sensor 1220, sensor 1220 can provide raw image data to ISP processor 1240 based on sensor 1220 interface type, or sensor 1220 can store raw image data into image memory 1230.

The ISP processor 1240 processes the raw image data pixel by pixel in a variety of formats. For example, each image pixel may have a bit depth of 12, 10, 12, or 14 bits, and the ISP processor 1240 may perform one or more image processing operations on the raw image data, collecting statistical information about the image data. Among them, image processing operations can be performed with the same or different bit depth precision.

ISP processor 1240 can also receive image data from image memory 1230. For example, sensor 1220 interface transmits raw image data to image memory 1230, which is then provided to ISP processor 1240 for processing. Image memory 1230 can be part of a memory device, a storage device, or a separate dedicated memory within an electronic device, and can include DMA (Direct Memory Access) features.

When receiving raw image data from the image sensor 1214 interface or from the sensor 1220 interface or from the image memory 1230, the ISP processor 1240 can perform one or more image processing operations, such as time domain filtering. The processed image data can be sent to image memory 1230 for additional processing prior to being displayed. The ISP processor 1240 can also receive processing data from the image memory 1230, processing the processed data in image data in the original domain and in the RGB and YCbCr color spaces. The processed image data can be output to display 1280 for viewing by a user and/or further processed by a graphics engine or GPU (Graphics Processing Unit). Additionally, the output of ISP processor 1240 can also be sent to image memory 1230, and display 1280 can read image data from image memory 1230. In one embodiment, image memory 1230 can be configured to implement one or more frame buffers. Additionally, the output of ISP processor 1240 can be sent to encoder/decoder 1270 to encode/decode image data. The encoded image data can be saved and decompressed before being displayed on the display 1280 device.

The ISP processor 1240 processes the image data by performing VFE (Video Front End) processing and CPP (Camera Post Processing) processing on the image data. The VFE processing of the image data may include correcting the contrast or brightness of the image data, modifying the digitally recorded illumination state data, performing compensation processing on the image data (such as white balance, automatic gain control, gamma correction, etc.), and performing image data. Filter processing, etc. CPP processing of image data may include scaling the image, providing a preview frame and a recording frame to each path. Among them, CPP can use different codecs to process preview frames and record frames. The image data processed by the ISP processor 1240 can be sent to the beauty module 1260 to perform a cosmetic process on the image before being displayed. The beauty module 1260 can process the beauty of the image data, including: whitening, freckle, dermabrasion, face-lifting, acne, eye enlargement, and the like. The beauty module 1260 can be a CPU (Central Processing Unit), a GPU, a coprocessor, or the like in the mobile terminal. The processed data of the beauty module 1260 can be sent to the encoder/decoder 1270 to encode/decode the image data. The encoded image data can be saved and decompressed before being displayed on the display 1280 device. The beauty module 1260 can also be located between the encoder/decoder 1270 and the display 1280, that is, the beauty module performs cosmetic processing on the imaged image. The encoder/decoder 1270 described above may be a CPU, a GPU, a coprocessor, or the like in a mobile terminal.

The statistics determined by the ISP processor 1240 can be sent to the control logic 1250 unit. For example, the statistical data may include image sensor 1214 statistical information such as auto exposure, auto white balance, auto focus, flicker detection, black level compensation, lens 1212 shading correction, and the like. Control logic 1250 can include a processor and/or a microcontroller that executes one or more routines (such as firmware) that can determine control parameters of imaging device 1210 and ISP processing based on received statistical data. Control parameters of the device 1240. For example, control parameters of imaging device 1210 may include sensor 1220 control parameters (eg, gain, integration time for exposure control), camera flash control parameters, lens 1212 control parameters (eg, focus or zoom focal length), or a combination of these parameters. The ISP control parameters may include gain levels and color correction matrices for automatic white balance and color adjustment (eg, during RGB processing), as well as lens 1212 shading correction parameters.

The image processing method as above can be realized by using the image processing technique of FIG.

Any reference to a memory, storage, database or other medium used herein may include non-volatile and/or volatile memory. Suitable non-volatile memories can include read only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM), which acts as an external cache. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronization. Link (Synchlink) DRAM (SLDRAM), Memory Bus (Rambus) Direct RAM (RDRAM), Direct Memory Bus Dynamic RAM (DRDRAM), and Memory Bus Dynamic RAM (RDRAM).

The above-mentioned embodiments are merely illustrative of several embodiments of the present application, and the description thereof is more specific and detailed, but is not to be construed as limiting the scope of the claims. It should be noted that a number of variations and modifications may be made by those skilled in the art without departing from the spirit and scope of the present application. Therefore, the scope of the invention should be determined by the appended claims.

Claims

An image processing method comprising:

Get the face in the image;

Obtaining a set of images having the same face in the image;

Identifying an age value corresponding to a face of the image in the set of images;

Generating sorting information for images in the set of images based on the age values; and

Based on the ranking information and the set of images, an album having a sequence of images in the set of images is generated.
The method according to claim 1, wherein the acquiring a face in the image comprises:

When a plurality of faces are included in the image, a master face of the plurality of faces is determined.
The method according to claim 2, wherein the determining a master face of the plurality of faces comprises:

The master face is determined based on the face area and/or face definition occupied by each face in the image.
The method according to claim 3, wherein the determining the master face based on the face area and/or the face definition occupied by each face in the image comprises:

Obtaining a face with the largest face area from the image, and determining whether the face meets a preset condition;

If yes, the face with the largest face area is used as the master face of the image;

If not, the face of the face area is obtained from the image, and the face is determined to meet the preset condition, and the execution is repeated until the face area reaches the minimum area threshold or the owner's face is obtained. ;and / or

Obtaining a face with the highest facial expression from the image, and determining whether the face meets a preset condition;

If yes, the face with the highest face resolution is used as the master face of the image;

If not, proceed to obtain a face whose face is second in sharpness from the image, determine whether the face meets a preset condition, and repeat execution until the face resolution reaches a minimum definition threshold or acquires a master. The face is up.
The method according to claim 4, wherein the face having the largest face area is obtained from the image, and determining whether the face meets a preset condition comprises:

Obtaining a face with the largest face area from the image, and determining whether the angle and focus of the face conform to a preset condition.
The method according to any one of claims 1 to 5, wherein the generating the ranking information of the image in the image set according to the age value comprises:

When the age values of the faces are the same among the plurality of images in the same image set, the ranking information for the plurality of images is generated according to the generation time of the plurality of images.
The method according to any one of claims 1 to 5, further comprising, after the generating an album having a sequence of images in the image set according to the sorting information and the image set, further comprising :

Playing the album with time series according to the timing, the sequence includes sequentially sorting and/or reverse sorting according to an age value, and the album with time series includes any one of a slide, a short film or an album. file.
The method according to any one of claims 1 to 5, wherein the acquiring a face in the image comprises:

Get the faces in the image from a multi-user family shared album.
The method according to any one of claims 1 to 5, further comprising: before the acquiring the image set having the same face in the image, further comprising:

Identifying a face in the image to determine a user identity corresponding to the face in the image; and

The image is divided into image sets corresponding to the user identity to form an image collection having the same face.
The method according to claim 9, wherein the dividing the image into an image set corresponding to the user identity to form an image set having the same face comprises:

And when the identified user identity is an existing user identity, acquiring an image set corresponding to the existing user identity, and dividing the image into a corresponding image set to form an image set having the same face ;

When the identified user identity is a new user identity, create an image set corresponding to the new user identity, and divide the image into the created image set to form an image set having the same face. .
The method of claim 2, wherein the method further comprises:

Obtaining a master face of the image, performing face recognition on the master face, and obtaining a user identity corresponding to the master face.
The method of claim 11 further comprising:

Acquiring the remaining faces in the image, and performing face recognition on the remaining faces in turn;

When the remaining human face belongs to an existing user identity, the remaining human face is divided into the corresponding existing user identity, and the user identity corresponding to the remaining human face is obtained;

When the remaining face does not belong to an existing user identity, the remaining face is marked as a suspected passerby face.
The method of claim 12, further comprising:

Calculating the number of times the suspected passerby face and the owner's face belonging to different user identities appear on the same image in the album;

When the number of times reaches the set threshold, the suspected passerby face is added as the main face, and the user identity corresponding to the suspected passerby face is obtained.
The method of claim 13 further comprising:

Obtaining the master face and the remaining faces in the newly added image, and dividing the newly added image into the image set corresponding to the user identity according to the user identity that the master face and the remaining faces in the newly added image can be separated into, Recalculating the number of times the suspected passerby face and the owner's face belonging to different user identities appear on the same image in the album, and when the number of times reaches the set threshold, the suspected passerby face is added as the main face. Obtaining the identity of the user corresponding to the suspected passerby face.
An image processing apparatus, characterized in that the apparatus comprises:

An image collection obtaining module, configured to acquire a human face in the image; and acquire an image collection having the same human face in the image;

An age value identification module, configured to identify an age value corresponding to a face of the image in the image set;

a sorting information generating module, configured to generate, according to the age value, sorting information of an image in the image set;

And an album generating module, configured to generate, according to the sorting information and the image set, a photo album having a sequence of images in the image set.
A computer readable storage medium having stored thereon a computer program, wherein the computer program is executed by a processor to perform the operations of the image processing method according to any one of claims 1 to 14.
An electronic device comprising a memory and a processor, wherein the memory stores a computer program, wherein the instructions are executed by the processor, causing the processor to perform any of claims 1 to 14 The operation of the image processing method described in the item.