1. A super-resolution chip, comprising:

the algorithm storage hardware module is used for storing the burnt initial super-resolution algorithm;

the updating interface is used for acquiring a user personalized image;

and the updating hardware module is respectively connected with the updating interface and the algorithm storage hardware module and is used for performing super-resolution learning on each user personalized image as a training sample to obtain a target super-resolution algorithm and updating the burnt initial super-resolution algorithm stored in the algorithm storage hardware module based on the target super-resolution algorithm.

2. The super-resolution chip of claim 1, wherein the update hardware module is further configured to determine an image content type of each of the user-customized images; screening out a target content type based on the image content type of each user personalized image; and performing super-resolution learning by using the user personalized image of the target content type as a training sample to obtain a target super-resolution algorithm for performing super-resolution processing on the user personalized image of the target content type.

3. The super-resolution chip of claim 2, wherein the update hardware module is further configured to count the number of images belonging to each of the image content types; and screening out target content types from the image content types based on the image quantity.

4. The super-resolution chip of claim 1, wherein the update interface is further configured to obtain a user-customized image sent by a target application; the target application is determined by the electronic equipment from each candidate application based on the use frequency of each candidate application.

5. The super-resolution chip of claim 1, further comprising:

the super-resolution module is used for updating the burnt initial super-resolution algorithm in the super-resolution chip based on the target super-resolution algorithm to obtain an updated super-resolution algorithm; under the condition that a test image needing super-resolution processing is obtained, determining the content type of the test image; determining a test super-resolution algorithm corresponding to the content type of the test image from the updated super-resolution algorithm according to the content type of the test image; and calling the test super-resolution algorithm to perform super-resolution processing on the test image.

6. The super-resolution chip of claim 1, wherein the update hardware module is further configured to obtain a first priority of the update hardware module, obtain a second priority of the algorithm storage hardware module, and update the burned initial super-resolution algorithm in the super-resolution chip based on the target super-resolution algorithm if the first priority is higher than the second priority.

7. An updating method of a super-resolution algorithm is characterized by comprising the following steps:

acquiring a user personalized image through a super-resolution chip;

performing super-resolution learning by taking the user personalized image as a training sample to obtain a target super-resolution algorithm;

and updating the burnt initial super-resolution algorithm in the super-resolution chip based on the target super-resolution algorithm.

8. The method of claim 7, wherein the super-resolution learning using the user-customized image as a training sample to obtain a target super-resolution algorithm comprises:

determining the image content type of each user personalized image;

screening out a target content type based on the image content type of each user personalized image;

and performing super-resolution learning by using the user personalized image of the target content type as a training sample to obtain a target super-resolution algorithm for performing super-resolution processing on the user personalized image of the target content type.

9. The method of claim 8, wherein the filtering out the target content type based on the image content type of each of the user-customized images comprises:

counting the number of images belonging to each of the image content types;

and screening out target content types from the image content types based on the image quantity.

10. The method of claim 7, wherein the obtaining of the user-customized image through the super-resolution chip comprises:

acquiring a user personalized image sent by a target application through a super-resolution chip; the target application is determined by the electronic equipment from each candidate application based on the use frequency of each candidate application.

11. The method of claim 7, further comprising:

after the burnt initial super-resolution algorithm in the super-resolution chip is updated based on the target super-resolution algorithm, an updated super-resolution algorithm is obtained;

under the condition that a test image needing super-resolution processing is obtained, determining the content type of the test image;

determining a test super-resolution algorithm corresponding to the content type of the test image from the updated super-resolution algorithm according to the content type of the test image;

and calling the test super-resolution algorithm through the super-resolution chip to perform super-resolution processing on the test image.

12. The method of claim 7, wherein the super-resolution chip comprises an update interface, an update hardware module and an algorithm storage hardware module; the updating interface is used for acquiring a user personalized image; the updating hardware module is used for performing super-resolution learning by taking each user personalized image as a training sample to obtain a target super-resolution algorithm; the algorithm storage hardware module is used for storing the burnt initial super-resolution algorithm;

the method further comprises the following steps:

acquiring a first priority of the updated hardware module and a second priority of the algorithm storage hardware module;

and under the condition that the first priority is higher than the second priority, updating the burnt initial super-resolution algorithm in the super-resolution chip by the updating hardware module based on the target super-resolution algorithm.

13. An electronic device, comprising the super-resolution chip of any one of claims 1 to 6.

14. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method according to any one of claims 7 to 12.

Background

When the electronic device processes the picture or the video picture, the content source is directly stretched or shrunk to the resolution ratio corresponding to the screen on the display link, and then the content source is output and displayed on the screen. However, after simple stretching processing, it is found that the originally clear picture is enlarged and then has an edge blurring effect, so that the stretching display affects the viewing effect of the user, and the picture of the content source is distorted.

With the development of computer technology, super-resolution technology has appeared, that is, the resolution of the original image is improved by hardware or software method. In the traditional super-resolution technology, a super-resolution algorithm is generally burned into a super-resolution chip, and then the super-resolution chip calls the super-resolution algorithm to perform super-resolution processing on an image. However, in the conventional super-resolution technology, only one set of super-resolution algorithm is usually burned in the super-resolution chip, and the super-resolution algorithm in the super-resolution chip cannot be updated.

Disclosure of Invention

The embodiment of the application provides a super-resolution algorithm updating method and device, electronic equipment and a computer readable storage medium, which can update a super-resolution algorithm in a super-resolution chip.

A super-resolution algorithm updating method comprises the following steps:

acquiring a user personalized image through a super-resolution chip;

performing super-resolution learning by taking the user personalized image as a training sample to obtain a target super-resolution algorithm;

and updating the burnt initial super-resolution algorithm in the super-resolution chip based on the target super-resolution algorithm.

An updating device of super-resolution algorithm comprises:

the image acquisition module is used for acquiring a user personalized image through the super-resolution chip;

the algorithm acquisition module is used for performing super-resolution learning by taking the user personalized image as a training sample to obtain a target super-resolution algorithm;

and the updating module is used for updating the burnt initial super-resolution algorithm in the super-resolution chip based on the target super-resolution algorithm.

A super-resolution chip, comprising:

the algorithm storage hardware module is used for storing the burnt initial super-resolution algorithm;

the updating interface is used for acquiring a user personalized image;

and the updating hardware module is respectively connected with the updating interface and the algorithm storage hardware module and is used for performing super-resolution learning on each user personalized image as a training sample to obtain a target super-resolution algorithm and updating the burnt initial super-resolution algorithm stored in the algorithm storage hardware module based on the target super-resolution algorithm.

An electronic device comprises the super-resolution chip.

A super-resolution chip comprising a memory and a processor, the memory having stored therein a computer program, which, when executed by the processor, causes the processor to perform the steps of the updating method of the super-resolution algorithm as described above.

A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method as described above.

The updating method and device of the super-resolution algorithm, the electronic equipment, the super-resolution chip and the computer readable storage medium acquire the user personalized image through the super-resolution chip; the method comprises the steps of performing super-resolution learning by taking a user personalized image as a training sample to obtain a target super-resolution algorithm meeting the user personalized requirements, updating an initial super-resolution algorithm burnt in a super-resolution chip based on the newly obtained target super-resolution algorithm, optimizing and improving the burnt super-resolution algorithm in the super-resolution chip, and meeting the personalized requirements of the user on super-resolution processing of the image.

Drawings

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.

FIG. 1 is a flow chart of a method for updating a super resolution algorithm in one embodiment;

FIG. 2 is a flowchart illustrating the steps of performing super-resolution learning using a user-customized image as a training sample to obtain a target super-resolution algorithm in one embodiment;

FIG. 3 is a flowchart of a super resolution algorithm updating method in another embodiment;

FIG. 4 is a diagram illustrating a method for updating the super-resolution algorithm in one embodiment;

FIG. 5 is a schematic flow chart illustrating the updating of the super resolution algorithm in another embodiment;

FIG. 6 is a block diagram showing an updating apparatus of the super-resolution algorithm in one embodiment;

fig. 7 is a schematic diagram of an internal structure of a super-resolution chip in one embodiment.

Detailed Description

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

It will be understood that, as used herein, the terms "first," "second," and the like may be used herein to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish one element from another. For example, a first priority may be referred to as a second priority, and similarly, a second priority may be referred to as a first priority, without departing from the scope of the present application. The first priority and the second priority are both priorities, but they are not the same priority.

In one embodiment, as shown in fig. 1, a method for updating a super-resolution algorithm is provided, and this embodiment is exemplified by applying the method to a super-resolution chip of an electronic device. The electronic device may be a terminal or a server, the terminal may be but is not limited to various personal computers, notebook computers, smart phones, tablet computers and portable wearable devices, and the server may be implemented by an independent server or a server cluster formed by a plurality of servers. It is understood that the method can also be applied to a system comprising a terminal and a server, and is implemented through the interaction of the terminal and the server. In this embodiment, the method includes the steps of:

and 102, acquiring a user personalized image through a super-resolution chip.

Super Resolution (SR) refers to a process of recovering a high resolution image (HR) from a low resolution image (LR) or an image sequence. A high resolution image means that the image has a high pixel density and can provide more detail that tends to play a key role in the application.

The super-resolution chip is a chip for performing super-resolution processing on an image. The super-resolution chip stores a super-resolution algorithm, and the super-resolution algorithm is called to perform super-resolution processing on the image, so that the resolution of the image is improved. The super-resolution chip is integrated in the electronic equipment.

The user-customized image is an image representing the user's personalized needs. The user personalization requirements may include at least portrait processing requirements and landscape processing requirements, etc. Wherein, the portrait processing requirements include face beautifying requirements, background blurring requirements, and the like. Landscape processing requirements such as increased filter requirements, brightness adjustment requirements, etc. The user personalized image is a self-portrait image, and the characterized user personalized requirement is face beautifying processing of the self-portrait image. The user personalized image is a landscape photo, and the characterized user personalized requirement is to adjust the brightness of the landscape photo so that the landscape photo is more realistic.

And acquiring the user personalized image from each application through the super-resolution chip. The application may be a third-party application or a system application. Third party applications are related software developed by other organizations or individuals than the software composer for deficiencies in the functionality of certain software or applications. Such as music applications, social applications, video applications, etc. downloaded by the user. The system application is an application program carried by an operating system of the electronic equipment. The system application is, for example, an album application, a desktop application, etc. of the operating system of the electronic device. The desktop application may obtain a ui (user interface) image.

And 104, performing super-resolution learning by taking the user personalized image as a training sample to obtain a target super-resolution algorithm.

The super-resolution learning refers to learning of super-resolution processing by a super-resolution chip. The target super-resolution algorithm refers to a super-resolution algorithm obtained by performing super-resolution learning on the user personalized image.

Specifically, the electronic equipment controls the super-resolution chip to take the personalized images of the users as training samples, and a machine learning algorithm is adopted to perform super-resolution learning on the training samples to obtain a target super-resolution algorithm. The machine Learning algorithm at least includes one of machine Learning (rotate Learning), Learning from teaching or Learning by Learning, Learning by deduction, analytical Learning (analytical Learning), supervised Learning (unsupervised Learning), unsupervised Learning (unsupervised Learning), and the like.

And 106, updating the burnt initial super-resolution algorithm in the super-resolution chip based on the target super-resolution algorithm.

The initial super-resolution algorithm refers to a burned super-resolution algorithm in a super-resolution chip. The number of the initial super-resolution algorithms is not limited, and may be one or more. Burning refers to burning data into a storage medium by a burning machine or the like. The super-resolution chip comprises an algorithm storage hardware module, and the algorithm storage hardware module stores a burned initial super-resolution algorithm. The MTP module is burned with a set of initial super-resolution algorithm set obtained based on learning before the super-resolution chip leaves a factory, wherein the initial super-resolution algorithm set obtained based on learning is obtained after super-resolution learning is carried out according to a trained graph set.

The electronic equipment controls the super-resolution chip to update the burnt initial super-resolution algorithm in the super-resolution chip based on the target super-resolution algorithm. In one embodiment, the super-resolution chip replaces the burned initial super-resolution algorithm with the target super-resolution algorithm to obtain an updated super-resolution algorithm, which is the target super-resolution algorithm. In another embodiment, the super-resolution chip adds a target super-resolution algorithm on the basis of the burned initial super-resolution algorithm to obtain an updated super-resolution algorithm. In other embodiments, the super-resolution chip may further replace the target super-resolution algorithm with a designated super-resolution algorithm in the burned initial super-resolution algorithm to obtain an updated super-resolution algorithm. Wherein, the specified super-resolution algorithm can be set according to the requirement. The method for updating the programmed initial super-resolution algorithm by the super-resolution chip is not limited herein.

In the embodiment, the user personalized image is obtained through the super-resolution chip; the method comprises the steps of performing super-resolution learning by taking a user personalized image as a training sample to obtain a target super-resolution algorithm meeting the user personalized requirements, updating an initial super-resolution algorithm burnt in a super-resolution chip based on the newly obtained target super-resolution algorithm, optimizing and improving the burnt super-resolution algorithm in the super-resolution chip, and meeting the personalized requirements of the user on super-resolution processing of the image.

In another embodiment, the super-resolution chip can also send the user personalized image to a designated server; performing super-resolution learning by using the user personalized image as a training sample through an appointed server to obtain a target super-resolution algorithm, and sending the target super-resolution algorithm to a super-resolution chip of the electronic equipment; and updating the burnt initial super-resolution algorithm in the super-resolution chip based on the target super-resolution algorithm.

In one embodiment, as shown in fig. 2, performing super-resolution learning using the user personalized image as a training sample to obtain a target super-resolution algorithm, including:

at step 202, the image content type of each user-customized image is determined.

The image content type refers to a type to which the content of the user-customized image corresponds. The image content types at least include a portrait type and a landscape type. Further, the portrait type may specifically include a face type, a whole-body portrait type, a multi-person snap type, and the like. The landscape type may specifically include a mountain type, a sea type, a forest type, and the like.

Specifically, the electronic device controls the super-resolution chip to acquire the image content type of each user personalized image. In one embodiment, the electronic device control central processing unit respectively processes each user personalized image by using an image content type analysis model, determines the image content type of each user personalized image, and transmits the image content type as the attribute information of the corresponding user personalized image and the user personalized image to the super-resolution chip. In another embodiment, the electronic device controls the super-resolution chip to process each user personalized image by using an image content type analysis model, and determines the image content type of each user personalized image.

And step 204, screening out target content types based on the image content types of the personalized images of the users.

The target content type refers to a content type of the user-customized image for super resolution learning. The target content type may be at least one of image content types.

And the electronic equipment controls the super-resolution chip to screen out a target content type from the image content types based on the image content types of the user personalized images. In one embodiment, the super-resolution chip may count the number of images belonging to each image content type, and take the image content type with the highest number of images as the target content type. In another embodiment, the super-resolution chip may randomly select one or more image content types as the target content type. In another embodiment, the super-resolution chip may further use an image content type specified by a user as a target content type from among the image content types. The manner of filtering out the target content type is not limited herein.

And step 206, performing super-resolution learning by using the user personalized image of the target content type as a training sample to obtain a target super-resolution algorithm for performing super-resolution processing on the user personalized image of the target content type.

The super-resolution chip performs super-resolution learning by using the user personalized image of the target content type as a training sample, so that the obtained target super-resolution algorithm can perform super-resolution processing on the image of the target content type more accurately, the user personalized requirements of the image belonging to the target content type can be met, and the image of the target content type with higher resolution and more clearness is obtained.

In this embodiment, the super-resolution chip screens out a target content type, performs super-resolution learning using a user personalized image of the target content type as a training sample, obtains a target super-resolution algorithm for a scene of the target content type, and can perform super-resolution processing on the image of the scene of the target content type more accurately, so as to obtain a clearer image of the target content type with higher resolution.

In one embodiment, screening out the target content type based on the image content type of the personalized image of each user comprises: counting the number of images belonging to each image content type; and screening out target content types from the image content types based on the quantity of the images.

In one embodiment, the super-resolution chip counts the number of images belonging to each image content type, and takes the image content type with the highest number of images as the target content type. It can be understood that the image content type with the highest number of images represents that the type is the type closest to the personalized requirements of the user, and then the image content type with the highest number of images is taken as the target content type, and the target content type can be subjected to super-resolution learning, so that a target super-resolution algorithm for performing super-resolution processing on the image of the target content type is obtained, the personalized requirements of the user can be solved more accurately, and the image with higher resolution and clearer resolution belonging to the target content type is obtained.

In another embodiment, the super-resolution chip counts the number of images belonging to each image content type, and takes the image content type with the second highest number of images as the target content type. In another embodiment, the super-resolution chip counts the number of images belonging to each image content type, and takes the image content type with the lowest number of images as the target content type. The method for screening the target content type from the image content types by the super-resolution chip based on the number of the images can be set as required, and is not limited herein.

In another embodiment, determining the image content type of each user-personalized image comprises: determining the image content type of each user personalized image in a preset time period; screening out target content types based on the image content types of the personalized images of the users, wherein the screening comprises the following steps: counting the type use frequency of each image content type in a preset time period; and screening out the target content type from the image content types based on the use frequency of each type. The preset time period can be set according to needs. The type use frequency refers to a frequency with which the image content type is used. The higher the type use frequency of the image content type is, the more the user likes to use the image of the image content type, and the image content type can more represent the personalized requirements of the user.

Alternatively, the super-resolution chip may use the image content type with the highest type use frequency as the target content type, or may use the image content type with the second highest type use frequency as the target content type, without being limited thereto.

In one embodiment, acquiring a user-customized image through a super-resolution chip includes: acquiring a user personalized image sent by a target application through a super-resolution chip; the target application is determined by the electronic device from the candidate applications based on the frequency of use of the candidate applications.

The candidate application is an application installed in the electronic device for determining the target application. The target application is an application program which is determined from the candidate applications and used for super-resolution learning of the produced user personalized image.

The candidate application may be a third party application or may be a system application. Third party applications are related software developed by other organizations or individuals than the software composer for deficiencies in the functionality of certain software or applications. Such as music applications, social applications, video applications, etc. downloaded by the user. The system application is an application program carried by an operating system of the electronic equipment. The system application is, for example, an album application, a desktop application, etc. of the operating system of the electronic device. Similarly, the target application determined from the candidate applications may be a third party application or a system application.

It is understood that, during the process of using the electronic device by the user, the electronic device may record the usage of each candidate application by the user, including the usage frequency, the usage time, and the like of each candidate application. The electronic device determines a target application from the candidate applications based on the frequency of use of the candidate applications. In one embodiment, the electronic device counts the use frequency of each candidate application, and determines the candidate application with the highest use frequency as the target application. In another embodiment, the electronic device counts the frequency of use of each candidate application, and determines the candidate application with the second highest frequency of use as the target application. In another embodiment, the electronic device counts the use frequency of each candidate application in a preset time period, and determines the candidate application with the highest use frequency in the preset time period as the target application. The manner in which the electronic device determines the target application is not limited, and is not limited herein.

And acquiring the user personalized image sent by the target application through the super-resolution chip. Optionally, under the condition that the target application sends the user personalized image in real time, the user personalized image sent by the target application is obtained in real time through the super-resolution chip; and under the condition that the target application sends the user personalized image at regular time, the user personalized image sent by the target application is obtained at regular time through the super-resolution chip.

In this embodiment, the super-resolution chip is used to obtain the user personalized image sent by the target application, and the target application is determined based on the use frequency of each candidate application, so that the personalized requirement of the user can be more accurately judged according to the use frequency of the user to each candidate application, and the user personalized image meeting the personalized requirement of the user can be more accurately obtained.

In one embodiment, as shown in fig. 3, the method further comprises:

step 302, after the burned initial super-resolution algorithm in the super-resolution chip is updated based on the target super-resolution algorithm, an updated super-resolution algorithm is obtained.

The super-resolution chip obtains an updated super-resolution algorithm, and the updated super-resolution algorithm comprises a target super-resolution algorithm.

And 304, determining the content type of the test image under the condition of acquiring the test image needing super-resolution processing.

The test image is an image that needs to be super-resolution processed. The test image content type is a type to which the image content of the test image corresponds. The test image content types at least comprise a portrait type and a landscape type. Further, the portrait type may specifically include a face type, a whole-body portrait type, a multi-person snap type, and the like. The landscape type may specifically include a mountain type, a sea type, a forest type, and the like.

The super-resolution chip acquires the content type of the test image under the condition of acquiring the test image needing super-resolution processing. In one embodiment, the electronic device control central processing unit processes the test image by using the image content type analysis model, determines the test image content type of the test image, and transmits the test image content type as the attribute information of the test image together with the test image to the super-resolution chip. In another embodiment, the electronic device controls the super-resolution chip to process the test image by using the image content type analysis model, and determines the test image content type of the test image.

And step 306, determining a test super-resolution algorithm corresponding to the content type of the test image from the updated super-resolution algorithm according to the content type of the test image.

The test super-resolution algorithm is a super-resolution algorithm corresponding to the content type of the test image, and can perform super-resolution processing on the image of the content type of the test image more accurately.

After the super-resolution chip obtains the test image content type of the test image, the test image content type is matched with the types processed by the super-resolution algorithms in the updated super-resolution algorithm, and when the matching is successful, the test super-resolution algorithm matched with the test image content type is obtained.

And 308, calling a test super-resolution algorithm through a super-resolution chip to perform super-resolution processing on the test image.

The super-resolution chip calls a test super-resolution algorithm, and can perform super-resolution processing on the test image of the content type of the test image to obtain an image with higher resolution and clearer resolution.

In this embodiment, the super-resolution chip determines the content type of the test image when acquiring the test image that needs to be super-resolution processed, determines the test super-resolution algorithm corresponding to the content type of the test image from the updated super-resolution algorithm according to the content type of the test image, and calls the test super-resolution algorithm, so that the super-resolution processing can be more accurately performed on the test image belonging to the content type of the test image, and an image with higher resolution and clearer resolution can be obtained.

In one embodiment, the super-resolution chip comprises an updating interface, an updating hardware module and an algorithm storage hardware module; the updating interface is used for acquiring a user personalized image; the updating hardware module is used for performing super-resolution learning by taking the personalized images of the users as training samples to obtain a target super-resolution algorithm; the algorithm storage hardware module is used for storing the burnt initial super-resolution algorithm; the method further comprises the following steps: acquiring a first priority for updating the hardware module and a second priority for storing the hardware module by an algorithm; and under the condition that the first priority is higher than the second priority, updating the burnt initial super-resolution algorithm in the super-resolution chip by the updating hardware module based on the target super-resolution algorithm.

The updating interface, the updating hardware module and the algorithm storage hardware module are all hardware modules in the super-resolution chip. The super-resolution chip acquires the user personalized image through the updating interface and also can acquire a test image needing super-resolution processing through the updating interface. And the updating interface transmits the acquired user personalized images to the updating hardware module, the updating hardware module performs super-resolution learning on each user personalized image as a training sample to obtain a target super-resolution algorithm, and then the target super-resolution algorithm is transmitted to the algorithm storage hardware module to update the burnt initial super-resolution algorithm stored in the algorithm storage hardware module. Wherein, the algorithm storage hardware module can be an MTP module. The burnt initial super-resolution algorithm in the MTP module belongs to factory default setting.

The first priority is a priority level of updating the hardware module. The second priority is the priority of the algorithm storage hardware module. The first priority and the second priority can be set according to the needs of the user. For example, if the first priority is three-level and the second priority is two-level, the first priority is higher than the second priority, and the priority for updating the hardware module is higher. For another example, if the first priority is two-level and the second priority is four-level, the first priority is lower than the second priority, and the priority for updating the hardware module is lower.

And under the condition that the first priority is higher than the second priority, the priority degree of updating the hardware module is higher, namely the updating of the burnt initial super-resolution algorithm in the algorithm storage hardware module is more important, and the burnt initial super-resolution algorithm in the super-resolution chip is updated by the updating hardware module based on the target super-resolution algorithm. Optionally, the updating manner may be to add the target super-resolution algorithm on the basis of the initial super-resolution algorithm, or to replace the initial super-resolution algorithm with the target super-resolution algorithm, which is not limited to this.

And under the condition that the first priority is lower than or equal to the second priority, the burnt initial super-resolution algorithm in the algorithm storage hardware module is not updated, or other operations are performed according to the user instruction.

In the embodiment, an updating interface and an updating hardware module are added in the super-resolution chip, and the user personalized image can be acquired through the updating interface; and performing super-resolution learning by using the update hardware module and taking the personalized images of the users as training samples to obtain a target super-resolution algorithm, so that under the condition that the first priority is higher than the second priority, the update hardware module updates the burnt initial super-resolution algorithm in the super-resolution chip based on the target super-resolution algorithm to update the burnt initial super-resolution algorithm in the super-resolution chip.

In one embodiment, as shown in FIG. 4, the super-resolution chip 402 includes a super-resolution module, an algorithm storage hardware module, an update hardware module, and an update interface. The super-resolution chip is connected with an Access Point (AP) end through an updating interface, and the Access end can acquire user personalized images including third-party applications and system applications. The system application comprises an album application, a video application and the like which are carried by an operating system of the electronic equipment. And the updating hardware module is respectively connected with the updating interface and the algorithm storage hardware module, receives the user personalized image transmitted by the updating interface, performs super-resolution learning by taking the user personalized image as a training sample to obtain a target super-resolution algorithm, and transmits the target super-resolution algorithm to the algorithm storage hardware module to update the burnt initial super-resolution algorithm. And the algorithm storage hardware module is used for storing the burned initial super-resolution algorithm.

When the super-resolution chip 402 acquires a test image to be super-resolution processed, the super-resolution module calls a super-resolution algorithm in the algorithm storage hardware module to perform super-resolution processing on the test image, so that a clearer image with higher resolution can be obtained.

Fig. 5 is a flowchart illustrating updating of the super-resolution algorithm in another embodiment. And 502, producing a super-resolution chip. The super-resolution chip executes step 504, and the super-resolution chip burns the initial super-resolution algorithm. The server executes step 506, the server learns the super-resolution to obtain a target super-resolution algorithm, and sends the target high-resolution algorithm to a super-resolution chip in the electronic device. The super-resolution chip executes step 508 and updates the super-resolution algorithm. Next, step 510, the super-resolution chip is integrated into the electronic device, and then the super-resolution chip in the electronic device may execute step 512 to perform super-resolution processing on the test image, so as to obtain a high-resolution test image 514.

It should be understood that, although the steps in the flowcharts of fig. 1 to 3 are shown in sequence as indicated by the arrows, the steps are not necessarily performed in sequence as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least some of the steps in fig. 1-3 may include multiple sub-steps or multiple stages that are not necessarily performed at the same time, but may be performed at different times, and the order of performing the sub-steps or stages is not necessarily sequential, but may be performed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.

Fig. 6 is a block diagram illustrating a structure of an updating apparatus of the super-resolution algorithm according to an embodiment. As shown in fig. 6, there is provided an updating apparatus of a super-resolution algorithm, including: an image acquisition module 602, an algorithm acquisition module 604, and an update module 606, wherein:

an image obtaining module 602, configured to obtain a user-customized image through a super-resolution chip.

And the algorithm acquisition module 604 is used for performing super-resolution learning by using the user personalized image as a training sample to obtain a target super-resolution algorithm.

And an updating module 606, configured to update the burned initial super-resolution algorithm in the super-resolution chip based on the target super-resolution algorithm.

The updating device of the super-resolution algorithm acquires the user personalized image through the super-resolution chip; the method comprises the steps of performing super-resolution learning by taking a user personalized image as a training sample to obtain a target super-resolution algorithm meeting the user personalized requirements, updating an initial super-resolution algorithm burnt in a super-resolution chip based on the newly obtained target super-resolution algorithm, optimizing and improving the burnt super-resolution algorithm in the super-resolution chip, and meeting the personalized requirements of the user on super-resolution processing of the image.

In one embodiment, the algorithm obtaining module 604 is further configured to determine an image content type of each user personalized image; screening out a target content type based on the image content type of each user personalized image; and performing super-resolution learning by taking the user personalized image of the target content type as a training sample to obtain a target super-resolution algorithm for performing super-resolution processing on the user personalized image of the target content type.

In one embodiment, the algorithm obtaining module 604 is further configured to count the number of images belonging to each image content type; and screening out target content types from the image content types based on the quantity of the images.

In one embodiment, the image obtaining module 602 is further configured to obtain, through the super-resolution chip, a user-customized image sent by the target application; the target application is determined by the electronic device from the candidate applications based on the frequency of use of the candidate applications.

In one embodiment, the apparatus further includes a super-resolution processing module, configured to obtain an updated super-resolution algorithm after updating the burned initial super-resolution algorithm in the super-resolution chip based on the target super-resolution algorithm; determining the content type of a test image of the test image under the condition of acquiring the test image needing super-resolution processing; determining a test super-resolution algorithm corresponding to the content type of the test image from the updated super-resolution algorithm according to the content type of the test image; and calling a test super-resolution algorithm through a super-resolution chip to perform super-resolution processing on the test image.

In one embodiment, the super-resolution chip comprises an updating interface, an updating hardware module and an algorithm storage hardware module; the updating interface is used for acquiring a user personalized image; the updating hardware module is used for performing super-resolution learning by taking the personalized images of the users as training samples to obtain a target super-resolution algorithm; and the algorithm storage hardware module is used for storing the burnt initial super-resolution algorithm. The update module 606 is further configured to obtain a first priority for updating the hardware module, and a second priority for storing the hardware module by using an algorithm; and under the condition that the first priority is higher than the second priority, updating the burnt initial super-resolution algorithm in the super-resolution chip by the updating hardware module based on the target super-resolution algorithm.

The division of each module in the update apparatus of the super-resolution algorithm is only used for illustration, and in other embodiments, the update apparatus of the super-resolution algorithm may be divided into different modules as needed to complete all or part of the functions of the update apparatus of the super-resolution algorithm.

For specific definition of the updating means of the super-resolution algorithm, reference may be made to the above definition of the updating method of the super-resolution algorithm, which is not described herein again. The modules in the super-resolution algorithm updating device can be wholly or partially realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.

The application provides a super-resolution chip which comprises an algorithm storage hardware module, an updating interface and an updating hardware module. And the algorithm storage hardware module is used for storing the burnt initial super-resolution algorithm. And the updating interface is used for acquiring the personalized image of the user. And the updating hardware module is respectively connected with the updating interface and the algorithm storage hardware module and is used for performing super-resolution learning on the personalized images of the users as training samples to obtain a target super-resolution algorithm and updating the burnt initial super-resolution algorithm stored in the algorithm storage hardware module based on the target super-resolution algorithm.

In one possible embodiment, the update interface may be an I/O interface, a USB (Universal Serial Bus) interface, or the like. The algorithm storage hardware module may be a variety of memories. The update hardware module may be a processor such as a DSP (Digital Signal Processing), an MCU (micro controller Unit), and a CPU (central Processing Unit).

In one embodiment, the update hardware module is further configured to determine an image content type of each user-customized image; screening out a target content type based on the image content type of each user personalized image; and performing super-resolution learning by taking the user personalized image of the target content type as a training sample to obtain a target super-resolution algorithm for performing super-resolution processing on the user personalized image of the target content type.

In one embodiment, the update hardware module is further configured to count the number of images belonging to each image content type; and screening out target content types from the image content types based on the quantity of the images.

In one embodiment, the update interface is further configured to obtain a user-customized image sent by the target application; the target application is determined by the electronic device from the candidate applications based on the frequency of use of the candidate applications.

In one embodiment, the super-resolution chip further comprises a super-resolution module, configured to obtain an updated super-resolution algorithm after updating an initial super-resolution algorithm burned in the super-resolution chip based on a target super-resolution algorithm; determining the content type of a test image of the test image under the condition of acquiring the test image needing super-resolution processing; determining a test super-resolution algorithm corresponding to the content type of the test image from the updated super-resolution algorithm according to the content type of the test image; and calling a test super-resolution algorithm to perform super-resolution processing on the test image.

In an embodiment, the update hardware module is further configured to obtain a first priority of updating the hardware module, store a second priority of the hardware module by using an algorithm, and update the burned initial super-resolution algorithm in the super-resolution chip based on the target super-resolution algorithm when the first priority is higher than the second priority.

The application also provides an electronic device, wherein the electronic device comprises a super-resolution chip, and the steps of the method described in the embodiment of the application can be realized through the super-resolution chip.

Fig. 7 is a schematic diagram of an internal structure of a super-resolution chip in one embodiment. As shown in fig. 7, the super-resolution chip includes a processor and a memory connected by a system bus. The processor is used for providing calculation and control capability and supporting the operation of the whole super-resolution chip. The memory may be an algorithm storage hardware module, which may be an MTP (Multiple-Time Programmable) module, the non-volatile storage medium stores a computer program, which is executable by the processor for implementing a super-resolution algorithm updating method provided in the following embodiments.

The implementation of each module in the updating apparatus of the super-resolution algorithm provided in the embodiment of the present application may be in the form of a computer program. The computer program can be run on a super-resolution chip on a terminal or a server. The program modules constituted by the computer program may be stored on a memory of a super-resolution chip of the electronic device.

Which when executed by a processor, performs the steps of the method described in the embodiments of the present application.

The embodiment of the application also provides a computer readable storage medium. One or more non-transitory computer-readable storage media containing computer-executable instructions that, when executed by one or more processors, cause the processors to perform the steps of the update method of the super resolution algorithm.

A computer program product comprising instructions which, when run on a computer, cause the computer to perform a method of updating a super resolution algorithm.

Any reference to memory, storage, database, or other medium used herein may include non-volatile and/or volatile memory. Non-volatile memory 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 external cache memory. 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), double data rate SDRAM (DDR SDRAM), Enhanced SDRAM (ESDRAM), synchronous Link (Synchlink) DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and bus dynamic RAM (RDRAM).

The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the present application. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.