WO2022028184A1

WO2022028184A1 - Photography control method and apparatus, electronic device, and storage medium

Info

Publication number: WO2022028184A1
Application number: PCT/CN2021/104515
Authority: WO
Inventors: 陈雪繁
Original assignee: RealMe重庆移动通信有限公司
Priority date: 2020-08-03
Filing date: 2021-07-05
Publication date: 2022-02-10
Also published as: CN112019739A

Abstract

Disclosed in the embodiments of the present application are a photography control method and apparatus, an electronic device, and a storage medium, the method comprising: when photographing a target photography scene, controlling an image acquisition unit to acquire at least one preview image frame of the target photography scene; inputting the preview image into an image quality evaluation model constructed on the basis of a deep learning algorithm to obtain an image quality evaluation result of the preview image; and, on the basis of the image quality evaluation result of the preview image, determining from the at least one preview image frame a target photography image of the target photography scene. Thus, it is only necessary to add to existing electronic devices an image quality evaluation model used for performing image quality evaluation on preview images acquired during the photography process and, on the basis of the image quality evaluation result, automatically selecting a preview image frame with the best image quality, thereby avoiding the problem of poor photography effects due to user causes when the user autonomously controls the photography, and improving the photography effect of the electronic device.

Description

A shooting control method, device, electronic device and storage medium

CROSS-REFERENCE TO RELATED APPLICATIONS

This application is based on the prior Chinese patent application with the application number of 202010768972.0, the filing date of August 3, 2020, and the invention-creation title of "a photographing control method, device, electronic device and storage medium", and requires the prior Chinese patent application. The priority of the Chinese patent application, the entire content of the earlier Chinese patent application is hereby incorporated by reference into this application in its entirety.

technical field

The present application relates to image processing technology, and in particular, to a shooting control method, device, electronic device, and storage medium.

Background technique

When a user shoots with an electronic device with a shooting function, the electronic device displays grid lines on the image display interface, automatically adjusts shooting parameters by recognizing the shooting scene, or assists the user to complete shooting by means of gesture shooting. However, the existing shooting control solution can only provide users with basic shooting suggestions, and still requires the user to manually operate the shooting button to shoot. Poor quality, affecting the shooting performance of electronic equipment.

SUMMARY OF THE INVENTION

In order to solve the above technical problems, the embodiments of the present application expect to provide a photographing control method, apparatus, electronic device, and storage medium.

The technical solution of the present application is realized as follows:

In a first aspect, a shooting control method is provided, the method comprising:

When shooting the target shooting scene, control the image acquisition unit to collect at least one preview image of the target shooting scene;

Inputting the preview image into an image quality evaluation model constructed based on a deep learning algorithm to obtain an image quality evaluation result of the preview image;

Based on the image quality evaluation result of the preview image, a target shot image for the target shot scene is determined from the at least one frame of preview image.

In a second aspect, a shooting control device is provided, the device comprising:

a control unit, configured to control the image acquisition unit to collect at least one preview image of the target shooting scene when shooting the target shooting scene;

an evaluation unit, configured to input the preview image into an image quality evaluation model constructed based on a deep learning algorithm to obtain an image quality evaluation result of the preview image;

A determination unit configured to determine, from the at least one frame of preview image, a target shot image for the target shot scene based on an image quality evaluation result of the preview image.

In a third aspect, an electronic device is provided, comprising: a processor and a memory configured to store a computer program executable on the processor, wherein the processor is configured to execute the aforementioned method when the computer program is executed. step.

In a fourth aspect, there is provided a computer-readable storage medium on which a computer program is stored, wherein the computer program, when executed by a processor, implements the steps of the aforementioned method.

Embodiments of the present application provide a shooting control method, device, electronic device, and storage medium. The method includes: when shooting a target shooting scene, controlling an image acquisition unit to collect at least one frame of preview image of the target shooting scene; The preview image is input into an image quality evaluation model constructed based on a deep learning algorithm, and an image quality evaluation result of the preview image is obtained; based on the image quality evaluation result of the preview image, it is determined from the at least one frame of preview image. An image is captured for the target of the target capture scene. In this way, it is only necessary to add an image quality evaluation model to the existing electronic equipment, which is used to evaluate the image quality of the collected preview images during the shooting process, and automatically select a preview image with the best image quality according to the image quality evaluation results. , to avoid the problem of poor shooting effect caused by the user when the user autonomously controls the shooting, and improve the shooting effect of the electronic device.

Description of drawings

FIG. 1 is a first schematic flowchart of a shooting control method in an embodiment of the present application;

2 is a first schematic diagram of a shooting preview interface in an embodiment of the present application;

3 is a second schematic diagram of a shooting preview interface in an embodiment of the present application;

FIG. 4 is a second schematic flowchart of the shooting control method in the embodiment of the present application;

FIG. 5 is a third schematic flowchart of the shooting control method in the embodiment of the present application;

6 is a schematic diagram of the construction principle of the IQA model in the embodiment of the application;

7 is a schematic diagram of the detection principle of the IQA model in the embodiment of the application;

8 is a schematic diagram of a camera frame in an embodiment of the present application;

FIG. 9 is a schematic diagram of the composition and structure of the photographing control device in the embodiment of the application;

FIG. 10 is a schematic diagram of a composition structure of an electronic device in an embodiment of the present application.

detailed description

In order to understand the features and technical contents of the embodiments of the present invention in more detail, the implementation of the embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

The embodiment of the present application provides a shooting control method, which can be applied to an electronic device with a shooting function. As shown in FIG. 1 , the method may specifically include:

Step 101: when shooting the target shooting scene, control the image acquisition unit to collect at least one frame of preview image of the target shooting scene;

Specifically, a user uses an electronic device with a shooting function to shoot a target shooting scene, and when the user performs a shooting start operation on the electronic device to generate a start control command, the electronic device starts to collect a preview image in response to the start control command.

In practical applications, the startup control instruction may be a button instruction, a touch instruction, or a voice instruction. Specifically, the method for obtaining the control instruction may include: obtaining the key command collected by the key input unit; obtaining the touch control command collected by the touch control unit; and obtaining the voice command collected by the voice collecting unit.

In practical applications, the electronic devices may be smart phones, tablet computers, portable multimedia players, virtual reality devices, wearable devices, and the like.

Step 102: Input the preview image into an image quality evaluation model constructed based on a deep learning algorithm to obtain an image quality evaluation result of the preview image;

In some embodiments, the method further includes: constructing an initial evaluation model based on a deep learning algorithm; acquiring a training sample set including at least one type of sample image; wherein each type of sample image corresponds to a shooting scene; using the training sample The initial evaluation model is trained on the set, and the trained image quality evaluation model is obtained.

Here, the initial evaluation model can use lightweight solutions for mobile terminals, such as: mobilenet-V3, faster-rcnn, yolov3, etc., and implement it based on open source libraries such as python and tensorflow lite.

Here, the training sample set can be directly obtained by using the existing image database, and using a large number of resources in the existing image database to train the model, an image quality evaluation model with wider application scenarios can be obtained. For example, ImageNet database, PASCAL VOC database, AFLW face database, LFW face database, etc.

In some embodiments, the acquiring a training sample set including at least one type of sample image includes: acquiring at least one type of sample image that satisfies a preset sample condition; wherein the preset sample condition includes at least one of the following: a human being The quality of the face feature is higher than the first threshold, and the quality of the image is higher than the second threshold; and the training sample set is composed of at least one type of sample images.

That is to say, since the selection of the training sample set will also affect the accuracy of model training, for some shooting scenes that are often used by users or the shooting scenes that users are most concerned about, such as selfie scenes or portrait shooting scenes, high quality can be selected. , that is, using high-quality sample images as a reference for model training, the trained image quality evaluation model will have higher quality evaluation standards for this type of shooting scene, and can filter out better images from the preview images. image.

In practical applications, the quality of facial features is used to characterize the performance of facial features. For example, according to a series of criteria such as face, closed eyes, smile, and light kiss, sample images with facial feature quality higher than the first threshold are selected. The first threshold is one or more thresholds obtained after quantification according to the evaluation criteria of the quality of the facial features.

Image quality is used to characterize the overall quality of the image. For example, the image quality is evaluated according to image parameters such as image clarity, 3A status (white balance, exposure time, focal length), and a sample image whose image quality is higher than the second threshold is selected. The threshold is one or more thresholds obtained after quantification according to the evaluation criteria of image quality.

In some embodiments, the method further includes: controlling the display unit to display the image quality evaluation result of the preview image.

That is to say, after evaluating the image quality of the preview image, the image quality evaluation result can also be displayed in real time to remind the user of the quality of the current preview image, as a basis for the user to adjust the shooting posture or shooting parameters. Improve shooting efficiency.

Fig. 2 is the first schematic diagram of the shooting preview interface in the embodiment of the application. As shown in Fig. 2, when the user turns on the camera to take a Selfie, the user can select a normal Selfie mode or a Selfie preferred mode. The shooting mode and the Selfie preferred mode are the modes that automatically select the best image quality generated by the technical solution of the present application. The default mode of the electronic device can be the ordinary Selfie mode, and the user can switch to the Selfie preferred mode by clicking the switch button. The default mode There is also a preferred mode for selfies. In the selfie preferred mode, the user can generate a start control command by clicking the shooting button at the bottom of the display interface, start collecting the preview image, and use the image quality evaluation model to evaluate the quality of the preview image.

FIG. 3 is a second schematic diagram of the shooting preview interface in the embodiment of the present application. As shown in FIG. 3 , when the user clicks the shooting button, evaluation processing of the preview image is started, and the preview interface can display a circle mark to indicate that evaluation is underway. After the image quality evaluation is completed, a telescopic progress bar is displayed on the right side of the preview interface to visually display whether the current preview image is close to the Selfie optimization threshold. If the threshold is exceeded, the user can click the shooting button again to complete the shooting, or automatically capture this frame. image and save.

Step 103: Based on the image quality evaluation result of the preview image, determine a target shot image for the target shot scene from the at least one frame of preview image.

Specifically, the preview image corresponding to the optimal image quality evaluation result may be selected as the target shot image and saved.

Here, the execution subject of steps 101 to 103 may be a processor of an electronic device, and the electronic device may be a smartphone, a wearable device (including a watch, a wristband, smart glasses, etc.), a tablet computer, a virtual reality device, and a vehicle-mounted device. Wait.

By adopting the above technical solution, it is only necessary to add an image quality evaluation model to the existing electronic equipment, which is used to evaluate the image quality of the collected preview images during the shooting process, and automatically select the one with the best image quality according to the image quality evaluation results. The frame preview image avoids the problem of poor shooting effect caused by the user when the user controls the shooting independently, and improves the shooting effect of the electronic device.

On the basis of the above-mentioned embodiment, the photographing control method provided by the embodiment of the present application is further exemplified, as shown in Figure 4, the method may specifically include:

Step 401 : when a start-up control instruction is detected, control the image acquisition unit to acquire at least one frame of preview image of the target shooting scene;

Step 402: Input the preview image into an image quality evaluation model constructed based on a deep learning algorithm to obtain an image quality evaluation result of the preview image;

Image quality is used to characterize the overall quality of the image. For example, sample images with image quality higher than the second threshold are selected according to image parameters such as image clarity, 3A status (white balance, exposure time, focal length), and the second threshold is based on the image quality. One or more thresholds obtained after quantification of quality evaluation criteria.

In some embodiments, the acquiring a training sample set including at least one type of sample images includes: acquiring at least one type of sample images that meet preset sample conditions; performing data enhancement on the sample images to obtain more sample images; of sample images to establish the training sample set.

Here, in order to increase the amount of data and alleviate the network overfitting, the sample images can be enhanced to obtain more training samples. Multiple images can be obtained on the basis of one image, and the number of samples can be enlarged, which can improve model training. efficient. Enhancement processing includes: random cropping, horizontal flipping, size scaling, hue adjustment, brightness adjustment, saturation adjustment and other preprocessing for a given sample image.

Further, when using the trained image quality evaluation model to evaluate the preview image quality, the output image quality evaluation result may also include the evaluation result of the facial feature quality and/or the evaluation result of the image quality.

That is to say, after evaluating the image quality of the preview image, the image quality evaluation result can also be displayed in real time to remind the user of the quality of the current preview image, which can be used as a basis for the user to adjust the shooting posture or shooting parameters. , to improve shooting efficiency.

Step 403: Based on the image quality evaluation result of the preview image, select the optimal image quality evaluation result;

In practical applications, when at least one frame of preview image is a frame of preview image, the optimal image quality evaluation result is the evaluation result of the frame of preview image. That is to say, the collected preview images are evaluated in sequence, and when it is detected that the evaluation result of the preview image of the current frame meets the image preference condition, the preview image of the current frame is directly used as the target image.

When at least one frame of preview images includes at least two frames of preview images, the optimal image quality evaluation result is selected according to the corresponding at least two image quality evaluation results.

Exemplarily, when the image quality evaluation result includes the evaluation result of the facial feature quality and/or the evaluation result of the image quality, the optimal facial feature quality, or the optimal image quality, or the one with the best comprehensive image quality is selected as the optimal image quality. Image quality evaluation results.

Step 404: Determine whether the optimal image quality evaluation result satisfies the image preference condition, if yes, go to Step 405; if not, go back to Step 406;

In some embodiments, the image preference condition includes at least one image quality evaluation criterion, wherein different image quality evaluation criteria are used to evaluate different evaluation results; correspondingly, it is determined whether the optimal image quality evaluation result satisfies Image optimization conditions, including: judging whether at least one evaluation result in the optimal image quality evaluation results satisfies the corresponding image quality evaluation standard; when each evaluation result satisfies the corresponding image quality evaluation standard, determining the optimal image quality evaluation result The image preference condition is satisfied; when there are some evaluation results that do not satisfy the corresponding image quality evaluation standard, it is determined that the optimal image quality evaluation result does not satisfy the image preference condition. That is to say, according to at least one image quality evaluation standard specified in the image preference condition, it is judged whether the optimal image quality evaluation result satisfies all the evaluation standards. ; otherwise, reacquire the preview image.

Exemplarily, one image quality evaluation criterion may be that the facial feature quality is higher than the first threshold, another image quality evaluation criterion may be that the image quality is higher than the second threshold, and another image quality evaluation criterion may be the overall image quality. above the third threshold. Here, the comprehensive image quality is an evaluation result after comprehensively considering the facial feature quality and the image quality, for example, obtained by performing a weighted operation on the evaluation results of the facial feature quality and the image quality. Correspondingly, the image quality evaluation result may include at least one of the following: a face feature quality evaluation result, an image quality evaluation result, and an image comprehensive quality evaluation result.

Exemplarily, the image quality evaluation result includes a comprehensive image quality score, and the comprehensive image quality score is a quantified value of the comprehensive image quality. For example, if the comprehensive image quality score ranges from 0 to 100, the corresponding image quality evaluation standard can be that the comprehensive image quality score is greater than 95 points, and the preview image is considered to be a high-quality image if it is greater than 95 points, and the preview image is considered to be a high-quality image if it is less than or equal to 95 points. The preview image is a low-quality image.

Step 405: Determine the preview image corresponding to the optimal image quality evaluation result as the target shot image;

In some embodiments, when the optimal image quality evaluation result satisfies the image preference condition, the method further includes: generating a photographing control instruction; in response to the photographing control instruction, controlling the display unit to display the target photographed image, and save all the photographed images. Take an image of the target.

Step 406 : control the image acquisition unit to continue to acquire at least one frame of preview image of the target shooting scene, and return to step 402 .

That is to say, for the currently collected preview image, the optimal preview image is selected for the user as the image obtained by this shooting operation by reasonably setting the image preference conditions. When the current preview image does not meet the image selection conditions, it means that the current shooting posture or shooting parameters are unreasonable, and the preview image needs to be collected again for judgment.

In some embodiments, when the optimal image quality evaluation result does not meet the image preference condition, the method further includes: generating and outputting photographing prompt information to prompt the user to adjust the photographing information; wherein the photographing information includes at least one of the following: Items: the shooting parameters of the image acquisition unit, the shooting posture of the image acquisition unit, and the shooting object posture.

That is to say, the shooting information that can be adjusted by the user may specifically include multiple types. For example, the shooting parameters include: shutter, aperture, sensitivity, whether to turn on the flash, etc. The shooting posture of the image acquisition unit includes the shooting position, shooting height, and rotation angle of the camera. facial features.

In practical applications, the types of different shooting information that affect the image quality are also different. For example, the posture of the photographed object will affect the quality of facial features, and the photographing parameters and photographing posture of the image acquisition unit will affect the overall quality of the image. Therefore, the shooting information that affects the evaluation result can be adjusted according to the influence relationship between the shooting information and the evaluation result.

Specifically, when the optimal image quality evaluation result does not meet the image optimization condition, the generating and outputting the shooting prompt information includes: the image optimization condition includes at least one image quality evaluation standard, and determining the target evaluation result that does not meet the image quality evaluation standard; Based on the influence relationship between the shooting information and the evaluation result, corresponding shooting prompt information is generated for the target evaluation result.

Exemplarily, when the quality of the human face is lower than the first threshold, it is determined that the factor affecting the quality of the human face is that the subject is in a state of closed eyes, and the prompt information may prompt the user to open his eyes; or the factor affecting the quality of the human face is that the face of the person is occluded. , prompts the user to adjust the character's position.

When the image quality is lower than the second threshold, it is determined that the factor affecting the image quality is that the camera attitude deviates from the horizontal attitude range, and the prompt information can prompt the user to adjust the camera angle; or the factor affecting the image quality is the image brightness, the prompt information can prompt the user to adjust the camera angle. The user turns on the flash.

By adding such a user interaction mode, the user can be reminded of unreasonable shooting information in time. In practical applications, the prompt information may also prompt a specific adjustment category.

By adopting the above technical solution, it is only necessary to add an image quality evaluation model to the existing electronic equipment, which is used to evaluate the image quality of the collected preview image during the shooting process, and automatically select the one with the best image quality according to the image quality evaluation result. The frame preview image avoids the problem of poor shooting effect caused by the user when the user controls the shooting independently, and improves the shooting effect of the electronic device.

On the basis of the above-mentioned embodiment, the photographing control method provided by the embodiment of the present application is further exemplified. As shown in FIG. 5 , the method may specifically include:

Step 501: the user clicks the shooting button to start shooting;

Specifically, the camera is turned on to enter the Selfie preferred mode, and the user clicks the shooting button to start shooting.

Step 502: control the camera to capture the preview image;

Specifically, the camera captures the preview image and displays the preview image to the user in real time.

Step 503: send the preview image into an image quality assessment (Image Quality Assessment, IQA) model to obtain an image quality score;

In practical applications, the method further includes: acquiring at least one type of sample image that meets preset sample conditions; wherein, the preset sample conditions include at least one of the following: the quality of the facial features is higher than the first threshold, and the quality of the images is higher than the second threshold; use at least one type of sample images to form the training sample set; use the training sample set to perform model training to obtain an IQA model.

FIG. 6 is a schematic diagram of the construction principle of the IQA model in the embodiment of the application. As shown in FIG. 6 , open-source libraries such as python and tensorflow lite are used, and lightweight networks such as mobilenet-V3 and Image Content Model (ICM) are used. Build the model, and use the pre-acquired high-quality training sample set to train the model to obtain a trained IQA model.

FIG. 7 is a schematic diagram of the detection principle of the IQA model in the embodiment of the application. As shown in FIG. 7 , the camera APP collects the preview image frame, displays the preview image on the preview interface, sends the preview image to the processing unit, and the processing unit runs the IQA model, Perform image quality evaluation on the preview image, output the image quality evaluation result, and control the shooting function of the camera APP according to the image quality evaluation result.

In practical applications, after obtaining the evaluation result of the preview image, the IQA model can also display the evaluation result next to the corresponding preview image in real time to remind the user of the quality of the current preview image, which can be used as a way for the user to adjust the shooting posture or shooting parameters. According to this, by increasing the human-computer interaction, the shooting efficiency is improved.

Step 504: Determine whether the image quality score is greater than the score threshold, if yes, go to Step 505; if not, go back to Step 502;

Step 505: generating a shooting control instruction;

Step 506: In response to the photographing control instruction, control the display unit to display the target photographed image, and save the target photographed image.

Exemplarily, when the camera APP is in the Selfie preferred mode, the camera is controlled to collect the preview image and save the corresponding timestamp, and the IQA model scores the preview image, and the obtained facial feature scoring result is higher than the first threshold (for example, the first threshold value). 95 points, out of 100 points), issue a shooting control command, save the preview image of the frame corresponding to the timestamp, and complete a photo shoot.

FIG. 8 is a schematic diagram of a camera frame in an embodiment of the application. As shown in FIG. 8 , the camera frame includes a camera preprocessing process (Camera Process), a camera service process (Camera Sever), and a camera hardware abstraction layer (Camera Hardware Abstraction Layer, Camera HAL). The camera preprocessing process includes camera APP and algorithm processing service unit (Algorithm Process Service, APS). The camera APP specifically includes a picture processing subunit and a video recording processing subunit, as well as an APS adapter and an APS service unit that interact with the APS unit, and the camera APP can initialize the APS. The APS unit includes a preprocessing unit. The preprocessing unit includes an Image Quality Assessment (IQA) model and an Automatic scene detection (ASD) model. The IQA model is used to evaluate the image quality of the preview image. As a result, a preview image with optimal image quality can be determined according to the evaluation results, and the ASD model is used to automatically identify the shooting scene and instruct to adjust the shooting parameters of the camera. Camera Process sends a capture request (Capture Request) to Camera Sever, and receives Camera Sever's capture result (Capture Result), Camera Sever and Camera HAL interact through interface definition language, so that camera APP can work correctly with Camera hardware , so as to ensure that all the functions of the Camera can work properly.

The embodiment of the present application also provides a shooting control device, as shown in FIG. 9 , the device includes:

The control unit 901 is configured to control the image acquisition unit to collect at least one frame of preview image of the target shooting scene when shooting the target shooting scene;

The processing unit 902 is configured to input the preview image into an image quality evaluation model constructed based on a deep learning algorithm to obtain an image quality evaluation result of the preview image;

The determining unit 903 is configured to determine, based on the image quality evaluation result of the preview image, a target shot image for the target shot scene from the at least one frame of preview image.

In some embodiments, the control unit 901 is configured to control the image acquisition unit to acquire at least one frame of preview image of the target shooting scene when the start control instruction is detected;

Correspondingly, the determining unit 903 is configured to select the optimal image quality evaluation result based on the image quality evaluation result of the preview image; determine whether the optimal image quality evaluation result satisfies the image preference condition; the optimal image quality evaluation When the result satisfies the image optimization condition, determine the preview image corresponding to the optimal image quality evaluation result as the target shot image;

The control unit 901 is configured to control the image acquisition unit to continue to collect at least one frame of preview image of the target shooting scene and input it into the image quality evaluation model when the optimal image quality evaluation result does not meet the image preference condition.

In some embodiments, the image preference condition includes at least one image quality evaluation criterion, wherein different image quality evaluation criteria are used to evaluate different evaluation results; the determining unit 903 is configured to determine at least one of the optimal image quality evaluation results Whether an evaluation result satisfies the corresponding image quality evaluation standard; when each evaluation result satisfies the corresponding image quality evaluation standard, it is determined that the optimal image quality evaluation result satisfies the image optimization condition; when there are some evaluation results that do not meet the corresponding image quality evaluation standard When the image quality evaluation standard is used, it is determined that the optimal image quality evaluation result does not satisfy the image optimization condition.

In some embodiments, the determining unit 903 is configured to generate a photographing control instruction when the optimal image quality evaluation result satisfies the image preference condition; in response to the photographing control instruction, control the display unit to display the target photographed image, and Save the target shot image.

In some embodiments, the determining unit 903 is configured to generate and output photographing prompt information when the optimal image quality evaluation result does not satisfy the image preference condition, so as to prompt the user to adjust the photographing information; wherein the photographing information includes the following: At least one item: the shooting parameters of the image acquisition unit, the shooting posture of the image acquisition unit, and the shooting object posture.

In some embodiments, the control unit 901 is configured to control the image acquisition unit to acquire at least one frame of preview image of the target shooting scene when a start control instruction is detected;

The determining unit 903 is configured to select an optimal image quality evaluation result based on the image quality evaluation result of the preview image; and determine the preview image corresponding to the optimal image quality evaluation result as the target shot image.

In some embodiments, the processing unit 902 is configured to, after obtaining the image quality evaluation result of the preview image, control the display unit to display the image quality evaluation result of the preview image.

In some embodiments, the apparatus further includes: a construction unit (not shown in FIG. 8 ), configured to construct an initial evaluation model based on a deep learning algorithm; obtain a training sample set including at least one type of sample images; wherein each type of The sample image corresponds to a shooting scene; the initial evaluation model is trained by using the training sample set, and the trained image quality evaluation model is obtained.

In some embodiments, the construction unit is configured to obtain at least one type of sample image that satisfies a preset sample condition; wherein the preset sample condition includes at least one of the following: the facial feature quality is higher than a first threshold, the image quality higher than the second threshold; using at least one type of sample images to form the training sample set.

Based on the hardware implementation of each unit in the above-mentioned photographing control device, the embodiment of the present application also provides an electronic device. As shown in FIG. 10 , the electronic device includes: a processor 1001 and a computer program configured to store a computer program that can be run on the processor. memory 1002;

Wherein, the processor 1001 is configured to execute the method steps in the foregoing embodiments when running a computer program.

Of course, in practical application, as shown in FIG. 9 , various components in the electronic device are coupled together through a bus system 1003 . It will be appreciated that the bus system 1003 is used to implement the connection communication between these components. In addition to the data bus, the bus system 1003 also includes a power bus, a control bus, and a status signal bus. However, for the sake of clarity, the various buses are labeled as bus system 1003 in FIG. 9 .

In practical applications, the above-mentioned processor may be an application specific integrated circuit (ASIC, Application Specific Integrated Circuit), a digital signal processing device (DSPD, Digital Signal Processing Device), a programmable logic device (PLD, Programmable Logic Device), a field programmable At least one of a programmable gate array (Field-Programmable Gate Array, FPGA), a controller, a microcontroller, and a microprocessor. It can be understood that, for different devices, the electronic device used to implement the above processor function may also be other, which is not specifically limited in the embodiment of the present application.

The above-mentioned memory can be a volatile memory (volatile memory), such as random access memory (RAM, Random-Access Memory); or a non-volatile memory (non-volatile memory), such as read-only memory (ROM, Read-Only Memory) Memory), flash memory (flash memory), hard disk (HDD, Hard Disk Drive) or solid-state drive (SSD, Solid-State Drive); or a combination of the above types of memory, and provide instructions and data to the processor.

Using the above technical electronic equipment, it is only necessary to add an image quality evaluation model to the existing electronic equipment, which is used to evaluate the image quality of the collected preview images during the shooting process, and automatically select the best image quality according to the image quality evaluation results. A frame preview image avoids the problem of poor shooting effect caused by the user when the user controls the shooting independently, and improves the shooting effect of the electronic device.

Embodiments of the present application further provide a computer-readable storage medium for storing a computer program.

Optionally, the computer-readable storage medium can be applied to any electronic device in the embodiments of the present application, and the computer program enables the computer to execute the corresponding processes implemented by the processors in the various methods of the embodiments of the present application, in order to It is concise and will not be repeated here.

In the several embodiments provided in this application, it should be understood that the disclosed apparatus and method may be implemented in other manners. The device embodiments described above are only illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components may be combined, or Can be integrated into another system, or some features can be ignored, or not implemented. In addition, the coupling, or direct coupling, or communication connection between the components shown or discussed may be through some interfaces, and the indirect coupling or communication connection of devices or units may be electrical, mechanical or other forms. of.

The unit described above as a separate component may or may not be physically separated, and the component displayed as a unit may or may not be a physical unit, that is, it may be located in one place or distributed to multiple network units; Some or all of the units may be selected according to actual needs to achieve the purpose of the solution in this embodiment.

In addition, each functional unit in each embodiment of the present invention may all be integrated into one processing module, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integration The unit can be implemented either in the form of hardware or in the form of hardware plus software functional units. Those of ordinary skill in the art can understand that all or part of the steps of implementing the above method embodiments can be completed by program instructions related to hardware, the aforementioned program can be stored in a computer-readable storage medium, and when the program is executed, execute Including the steps of the above method embodiment; and the aforementioned storage medium includes: a mobile storage device, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk and other various A medium on which program code can be stored.

The methods disclosed in the several method embodiments provided in this application can be arbitrarily combined under the condition of no conflict to obtain new method embodiments.

The features disclosed in the several product embodiments provided in this application can be combined arbitrarily without conflict to obtain a new product embodiment.

The features disclosed in several method or device embodiments provided in this application can be combined arbitrarily without conflict to obtain new method embodiments or device embodiments.

The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention. should be included within the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Industrial Applicability

Claims

A shooting control method, wherein the method comprises:

When shooting the target shooting scene, control the image acquisition unit to collect at least one preview image of the target shooting scene;

Inputting the preview image into an image quality evaluation model constructed based on a deep learning algorithm to obtain an image quality evaluation result of the preview image;

Based on the image quality evaluation result of the preview image, a target shot image for the target shot scene is determined from the at least one frame of preview image.
The method according to claim 1, wherein the controlling the image acquisition unit to acquire at least one frame of preview image of the target shooting scene comprises:

When the start control instruction is detected, the image acquisition unit is controlled to acquire at least one frame of preview image of the target shooting scene;

The determining, based on the image quality evaluation result of the preview image, a target shot image for the target shot scene from the at least one frame of preview image includes:

Based on the image quality evaluation result of the preview image, select the optimal image quality evaluation result;

Determine whether the optimal image quality evaluation result satisfies the image optimization condition;

When the optimal image quality evaluation result satisfies the image preference condition, determine that the preview image corresponding to the optimal image quality evaluation result is the target shot image;

When the optimal image quality evaluation result does not meet the image preference condition, the image acquisition unit is controlled to continue to collect at least one frame of preview image of the target shooting scene, and input it into the image quality evaluation model.
The method according to claim 2, wherein the image preference condition includes at least one image quality evaluation criterion, wherein different image quality evaluation criteria are used to evaluate different evaluation results;

The judging whether the optimal image quality evaluation result satisfies the image preference condition includes:

Judging whether at least one of the optimal image quality evaluation results satisfies the corresponding image quality evaluation standard;

When each evaluation result satisfies the corresponding image quality evaluation standard, it is determined that the optimal image quality evaluation result satisfies the image optimization condition;

When there are some evaluation results that do not meet the corresponding image quality evaluation criteria, it is determined that the optimal image quality evaluation result does not meet the image preference condition.
The method according to claim 2, wherein when the optimal image quality evaluation result satisfies the image preference condition, the method further comprises:

Generate shooting control instructions;

In response to the photographing control instruction, the display unit is controlled to display the target photographed image and save the target photographed image.
The method according to claim 2, wherein when the optimal image quality evaluation result does not satisfy the image preference condition, the method further comprises:

Shooting prompt information is generated and output to prompt the user to adjust the shooting information; wherein the shooting information includes at least one of the following: shooting parameters of the image acquisition unit, shooting posture of the image acquisition unit, and shooting object posture.
The method according to claim 1, wherein the controlling the image acquisition unit to acquire at least one frame of preview image of the target shooting scene comprises:

When the start control instruction is detected, the image acquisition unit is controlled to acquire at least one frame of preview image of the target shooting scene;

The determining, based on the image quality evaluation result of the preview image, a target shot image for the target shot scene from the at least one frame of preview image includes:

Based on the image quality evaluation result of the preview image, select the optimal image quality evaluation result;

It is determined that the preview image corresponding to the optimal image quality evaluation result is the target shot image.
The method according to claim 1, wherein after obtaining the image quality evaluation result of the preview image, the method further comprises:

The display unit is controlled to display the image quality evaluation result of the preview image.
The method of claim 1, wherein the method further comprises:

Build an initial evaluation model based on a deep learning algorithm;

Obtain a training sample set containing at least one type of sample image; wherein each type of sample image corresponds to one shooting scene;

The initial evaluation model is trained by using the training sample set to obtain the trained image quality evaluation model.
The method according to claim 8, wherein the acquiring a training sample set comprising at least one type of sample images comprises:

Acquiring at least one type of sample image that satisfies a preset sample condition; wherein the preset sample condition includes at least one of the following: the facial feature quality is higher than a first threshold, and the image quality is higher than a second threshold;

The training sample set is composed of at least one type of sample images.
A shooting control device, wherein the device comprises:

a control unit, configured to control the image acquisition unit to collect at least one preview image of the target shooting scene when shooting the target shooting scene;

a processing unit, configured to input the preview image into an image quality evaluation model constructed based on a deep learning algorithm to obtain an image quality evaluation result of the preview image;

A determination unit configured to determine, from the at least one frame of preview image, a target shot image for the target shot scene based on an image quality evaluation result of the preview image.
The device according to claim 10, wherein the control unit is configured to control the image acquisition unit to acquire at least one frame of preview image of the target shooting scene when a start control instruction is detected;

The determining unit is configured to select an optimal image quality evaluation result based on the image quality evaluation result of the preview image; determine whether the optimal image quality evaluation result satisfies the image preference condition; the optimal image quality evaluation result satisfies In the case of the image optimization condition, determining that the preview image corresponding to the optimal image quality evaluation result is the target shot image;

The control unit is configured to control the image acquisition unit to continue to collect at least one frame of preview image of the target shooting scene when the optimal image quality evaluation result does not meet the image preference condition, and input it into the image in the quality assessment model.
The apparatus according to claim 11, wherein the image preference condition includes at least one image quality evaluation criterion, wherein different image quality evaluation criteria are used to evaluate different evaluation results;

The determining unit is configured to determine whether at least one evaluation result in the optimal image quality evaluation results satisfies the corresponding image quality evaluation standard; when each evaluation result satisfies the corresponding image quality evaluation standard, determine the optimal image quality evaluation result. The optimal image quality evaluation result satisfies the image preference condition; when there are some evaluation results that do not satisfy the corresponding image quality evaluation standard, it is determined that the optimal image quality evaluation result does not satisfy the image preference condition.
The apparatus according to claim 11, wherein the determining unit is configured to generate a shooting control instruction when the optimal image quality evaluation result satisfies the image optimization condition;

In response to the photographing control instruction, the display unit is controlled to display the target photographed image and save the target photographed image.
The apparatus according to claim 11, wherein the determining unit is configured to generate and output photographing prompt information to prompt the user to adjust the photographing information when the optimal image quality evaluation result does not satisfy the image preference condition; Wherein, the shooting information includes at least one of the following: shooting parameters of the image acquisition unit, shooting posture of the image acquisition unit, and shooting object posture.
The device according to claim 10, wherein the control unit is configured to control the image acquisition unit to acquire at least one frame of preview image of the target shooting scene when a start control instruction is detected;

The determining unit is configured to select an optimal image quality evaluation result based on the image quality evaluation result of the preview image; and determine the preview image corresponding to the optimal image quality evaluation result as the target shot image.
The apparatus according to claim 10, wherein the processing unit is configured to, after obtaining the image quality evaluation result of the preview image, control the display unit to display the image quality evaluation result of the preview image.
The apparatus of claim 10, wherein the apparatus further comprises:

a construction unit, configured to construct an initial evaluation model based on a deep learning algorithm; obtain a training sample set including at least one type of sample image; wherein, each type of sample image corresponds to a shooting scene; use the training sample set to train the initial evaluation model to obtain the trained image quality evaluation model.
The apparatus according to claim 17, wherein the construction unit is configured to obtain at least one type of sample image that satisfies a preset sample condition; wherein the preset sample condition includes at least one of the following: high-quality facial features At the first threshold, the image quality is higher than the second threshold; and at least one type of sample image is used to form the training sample set.
An electronic device comprising: a processor and a memory configured to store a computer program executable on the processor,

Wherein, the processor is configured to execute the steps of the method of any one of claims 1 to 9 when running the computer program.
A computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the method of any one of claims 1 to 9.