Reference image screening method and device, computer readable medium and electronic device
By acquiring global and local features of candidate images and combining them with image evaluation data to select benchmark images, the problem of inaccurate benchmark image selection in existing technologies is solved, achieving higher accuracy and applicability.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-08-22
- Publication Date
- 2026-05-15
AI Technical Summary
In multi-frame noise reduction, existing technologies struggle to accurately select a baseline image that reflects the image content, resulting in poor image quality improvement.
By acquiring global and local image features of candidate images and combining the image evaluation data of global and local image features, a benchmark image that can accurately reflect the image content is selected.
It improves the accuracy and applicability of benchmark image selection, ensuring that the selected images better reflect the captured content and enhance image quality.
Smart Images
Figure CN115375573B_ABST
Abstract
Description
Technical Field
[0001] This disclosure relates to the field of image capture technology, and specifically to a reference image selection method, a reference image selection device, a computer-readable medium, and an electronic device. Background Technology
[0002] As people's living standards continue to improve, image quality is becoming an increasingly important concern. When shooting images, image quality can generally be improved through multi-frame noise reduction. Multi-frame noise reduction refers to the process where, in night scenes or low-light environments, the camera continuously captures multiple photos or images from the moment the shutter is pressed until the image is formed. Different pixels with noise characteristics are identified at different frame rates, and then weighted and combined to obtain a cleaner, purer night scene or low-light image.
[0003] When using multi-frame noise reduction to improve image quality, some of the acquired image frames may have poor quality. Therefore, it is necessary to select the reference image from the multiple image frames to participate in the multi-frame noise reduction calculation. Summary of the Invention
[0004] The purpose of this disclosure is to provide a reference image screening method, reference image screening device, computer-readable medium, and electronic device, thereby improving the accuracy of reference image screening and expanding the applicability of image screening to at least a certain extent.
[0005] According to a first aspect of this disclosure, a benchmark image screening method is provided, comprising:
[0006] Acquire at least two candidate images and determine the global image features corresponding to the candidate images;
[0007] Determine the region of interest in the candidate image and the local image features corresponding to the region of interest;
[0008] Based on the global image features and the local image features, determine the image evaluation data corresponding to each candidate image;
[0009] A benchmark image is selected from the candidate images based on the image evaluation data.
[0010] According to a second aspect of this disclosure, a reference image screening apparatus is provided, comprising:
[0011] A global image feature determination module is used to acquire at least two candidate images and determine the global image features corresponding to the candidate images;
[0012] A local image feature determination module is used to determine the region of interest in the candidate image and the local image features corresponding to the region of interest.
[0013] The image evaluation data determination module is used to determine the image evaluation data corresponding to each candidate image based on the global image features and the local image features.
[0014] A benchmark image filtering module is used to filter benchmark images from the candidate images based on the image evaluation data.
[0015] According to a third aspect of this disclosure, a computer-readable medium is provided that stores a computer program thereon, which, when executed by a processor, implements the method described above.
[0016] According to a fourth aspect of this disclosure, an electronic device is provided, characterized in that it comprises:
[0017] Processor; and
[0018] Memory is used to store one or more programs, which, when executed by one or more processors, cause the one or more processors to perform the methods described above.
[0019] One embodiment of this disclosure provides a benchmark image selection method that can acquire at least two consecutively captured candidate images and determine the global image features corresponding to the candidate images. Then, when a region of interest (ROI) is determined in a candidate image, the local image features corresponding to the ROI can be determined. Furthermore, based on the global and local image features, image evaluation data corresponding to each candidate image can be determined, and benchmark images are selected from among the candidate images based on the image evaluation data. When selecting benchmark images, not only are the global image features of the candidate images considered, but also the local image features of the ROI within the candidate images. The benchmark images are selected using image evaluation data jointly determined by the global and local image features, ensuring that the selected benchmark images accurately reflect the relevant information of the captured content, thus improving the accuracy of benchmark image selection. Simultaneously, selecting benchmark images based on general global and local image features is applicable to various types of candidate images, expanding the applicability of the selection scheme.
[0020] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description
[0021] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure. It is obvious that the drawings described below are merely some embodiments of this disclosure, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort. In the drawings:
[0022] Figure 1 A schematic diagram of an exemplary system architecture to which embodiments of the present disclosure may be applied is shown;
[0023] Figure 2 The schematic diagram illustrates a flowchart of a benchmark image selection method according to an exemplary embodiment of the present disclosure;
[0024] Figure 3 This schematically illustrates a flowchart for determining global image features in an exemplary embodiment of the present disclosure;
[0025] Figure 4 This schematic diagram illustrates the principle of a jump block statistics method in an exemplary embodiment of the present disclosure;
[0026] Figure 5 The illustration schematically shows a flowchart of determining local image features in an exemplary embodiment of the present disclosure;
[0027] Figure 6 This illustration schematically depicts a process for selecting a reference image in an exemplary embodiment of the present disclosure;
[0028] Figure 7 This illustration schematically depicts a process for determining a reference image using image evaluation data in an exemplary embodiment of this disclosure.
[0029] Figure 8 This schematically illustrates another process for selecting benchmark images in an exemplary embodiment of this disclosure;
[0030] Figure 9 This schematic diagram illustrates the composition of a reference image screening apparatus in an exemplary embodiment of the present disclosure;
[0031] Figure 10 A schematic diagram of an electronic device to which embodiments of the present disclosure may be applied is shown. Detailed Implementation
[0032] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that this disclosure will be more comprehensive and complete, and will fully convey the concept of the exemplary embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.
[0033] Furthermore, the accompanying drawings are merely illustrative of this disclosure and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor devices and / or microcontroller devices.
[0034] Figure 1 A schematic diagram of a system architecture for an exemplary application environment in which a benchmark image screening method and apparatus according to embodiments of the present disclosure can be applied is shown.
[0035] like Figure 1 As shown, system architecture 100 may include one or more of terminal devices 101, 102, and 103, a network 104, and a server 105. Network 104 serves as the medium for providing communication links between terminal devices 101, 102, and 103 and server 105. Network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables. Terminal devices 101, 102, and 103 may be various electronic devices with image processing capabilities, including but not limited to desktop computers, portable computers, smartphones, and tablets. It should be understood that... Figure 1 The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, there can be any number of terminal devices, networks, and servers. For example, server 105 could be a server cluster composed of multiple servers.
[0036] The reference image filtering method provided in this embodiment is generally executed by terminal devices 101, 102, and 103, and correspondingly, the reference image filtering device is generally disposed in terminal devices 101, 102, and 103. However, those skilled in the art will readily understand that the reference image filtering method provided in this embodiment can also be executed by server 105, and correspondingly, the reference image filtering device can also be disposed in server 105. This exemplary embodiment does not impose any special limitations on this. For example, in one exemplary embodiment, a user may continuously collect at least two candidate images through terminal devices 101, 102, and 103, and then upload the at least two candidate images to server 105. After the server determines the reference image using the reference image filtering method provided in this embodiment, it transmits the reference image to terminal devices 101, 102, and 103, etc.
[0037] In relevant benchmark image selection schemes, after acquiring multiple candidate images, the sharpness of each image is first calculated. The image with the highest sharpness is designated as the first benchmark image, and the image with the second highest sharpness is designated as the second benchmark image. Then, it is determined whether the difference between these two images exceeds a threshold. If so, the first benchmark image is considered the target image; otherwise, the sharpness difference between other images and the second benchmark image is calculated. If the difference does not exceed the threshold, the second benchmark image is set as the target image. The main evaluation metric in these schemes is image sharpness, which is relatively simple. Furthermore, the calculation of sharpness is easily affected by image content or noise, leading to low accuracy in benchmark image selection. Additionally, this scheme directly uses the sharpness information of the entire image as the selection criterion instead of focusing on regions of interest (such as faces) in the candidate images. This not only increases computational complexity but also easily causes the selected benchmark images to deviate from user expectations.
[0038] Based on one or more problems in the related technologies, this disclosure first provides a reference image filtering method. The following describes the reference image filtering method of the exemplary embodiment of this disclosure in detail, taking the execution of the method by a terminal device equipped with a reference image filtering device as an example.
[0039] The reference image selection method in this disclosure can be used in scenarios where multiple frames of images are used for composite processing. For example, when shooting night scenes or shooting in low-light environments, it is necessary to shoot multiple frames of images of the same shooting scene, and perform frame selection processing on the multiple frames to obtain the images with the best shooting effect. Then, noise reduction fusion and brightness enhancement processing are performed on the selected images. Alternatively, when shooting high dynamic range (HDR) images, it is necessary to select the images with the best effect from the captured images for fusion processing. Of course, it can also be other application scenarios that require selecting images from multiple frames. This disclosure does not specifically limit the type of application scenario.
[0040] Figure 2 A flowchart illustrating a benchmark image selection method in this exemplary embodiment is shown, which may include the following steps S210 to S240:
[0041] In step S210, at least two candidate images are acquired, and the global image features corresponding to the candidate images are determined.
[0042] In an exemplary embodiment, a candidate image refers to an image acquired in a scenario where multi-frame image synthesis is performed. For example, in a scenario where multi-frame synthesis is enabled, the terminal device typically captures multiple frames of the same scene continuously from a fixed shooting angle within a certain time period. Then, it selects a subset of these images as input images for subsequent synthesis. The multiple frames of the same scene captured continuously from a fixed shooting angle within a certain time period constitute at least two consecutively acquired candidate images in this embodiment. Candidate images can be of any type; for example, they can be images in the Raw domain, RGB domain, or YUV domain. This example embodiment is not limited to these categories.
[0043] Optionally, at least two candidate images of the current scene can be continuously captured at a fixed shooting angle within a certain period of time by a terminal device in shooting mode. Alternatively, at least two candidate images of the shooting scene can be continuously captured at a fixed shooting angle within a certain period of time by other shooting devices, and then the captured at least two candidate images can be sent to the terminal device for frame selection processing. Of course, candidate images can also be obtained in other ways. This example embodiment does not impose any special limitations on the method of obtaining candidate images.
[0044] Optionally, at least two candidate images can be all candidate images in the same multi-frame synthesis task, or a portion of all candidate images. For example, for a multi-frame noise reduction task started in night scene mode, a total of 60 frames are acquired. All 60 frames can be used as candidate images. Of course, the 60 frames can also be divided into 3 parts, with 20 frames in each part as candidate images, and filtered in parallel. This example embodiment does not impose any special limitations on this.
[0045] Global image features refer to the feature information obtained by statistically analyzing the image features of all image regions in each candidate image. For example, global image features can be brightness features obtained by statistically analyzing the brightness information of all candidate images. Global image features can also be sharpness features obtained by statistically analyzing the sharpness information of all candidate images. Global image features can also be sharpness features obtained by statistically analyzing the sharpness information of candidate images. This example embodiment does not specifically limit the type of global image features.
[0046] Optionally, global image features may include one type of image feature in the candidate image, or at least two types of image features in the candidate image. That is, the global image features of the candidate image can be evaluated by at least two types of image features. For example, the global image features corresponding to the candidate image can be determined by brightness information and sharpness information in the candidate image, or the global image features corresponding to the candidate image can be determined by brightness information and sharpness information. This example embodiment does not impose any special limitation on the feature types and number of features constituting the global image features.
[0047] In step S220, the region of interest in the candidate image and the local image features corresponding to the region of interest are determined.
[0048] In an exemplary embodiment, the region of interest refers to an image region in the candidate image that can characterize the main content of the image. For example, the region of interest can be a face region in the candidate image, or an image region corresponding to a foreground object in the candidate image. Of course, it can also be the image region in the candidate image with the densest color distribution. This example embodiment does not make any special limitation on the type of region of interest.
[0049] Regions of interest (ROIs) in candidate images can be determined by object detection. For example, face bounding boxes in candidate images can be determined by face detection and used as ROIs in candidate images. Alternatively, foreground objects in candidate images can be determined by image edge detection, and the connected regions or minimum bounding boxes corresponding to the foreground objects can be used as ROIs in candidate images. This example embodiment is not limited to these methods.
[0050] Local image features refer to the feature information obtained by statistically analyzing the image features of the region of interest in each candidate image. For example, local image features can be brightness features obtained by statistically analyzing the brightness information in the region of interest, or sharpness features obtained by statistically analyzing the sharpness information in the region of interest. Of course, local image features can also be sharpness features obtained by statistically analyzing the clarity information in the region of interest. This example embodiment does not make any special limitation on the type of local image features.
[0051] Optionally, local image features may include one type of image feature in the region of interest, or at least two types of image features in the region of interest. That is, local image features of a candidate image can be determined by at least two types of image features in the region of interest. For example, local image features corresponding to a candidate image can be determined by brightness information and sharpness information in the region of interest, or local image features corresponding to a candidate image can be determined by brightness information and sharpness information in the region of interest. This example embodiment does not impose any special limitation on the feature types and number of features constituting local image features.
[0052] In step S230, image evaluation data corresponding to each candidate image is determined based on the global image features and the local image features.
[0053] In an exemplary embodiment, image evaluation data refers to data used to measure whether a candidate image can be used as a reference image. For example, the global image features and local image features of the candidate image can be weighted and calculated to serve as the image evaluation data of the candidate image. Alternatively, the global image features and local image features can be summed and the sum can serve as the image evaluation data of the candidate image. This example embodiment does not impose any special limitations on the method of determining image evaluation data based on global image features and local image features.
[0054] In step S240, a benchmark image is selected from the candidate images based on the image evaluation data.
[0055] In an exemplary embodiment, the reference image refers to the image frame that is finally determined and participates in the multi-frame synthesis task. The reference image is a candidate image that can accurately reflect the content of the real scene among the acquired candidate images.
[0056] After determining the image evaluation data based on global and local image features, candidate images can be screened based on the image evaluation data to finally determine one or more benchmark images that meet the requirements as input data for participating in the multi-frame synthesis task.
[0057] When selecting benchmark images, not only are the global image features of candidate images considered, but also the local image features of the regions of interest in the candidate images. Benchmark images are selected using image evaluation data determined by both global and local image features, ensuring that the selected benchmark images accurately reflect the relevant information of the captured content and improving the accuracy of benchmark image selection.
[0058] Steps S210 to S240 will be described in detail below.
[0059] In an exemplary embodiment, the global image features may include any one or more combinations of first image brightness information, first image sharpness information, and first image clarity information corresponding to the candidate image. For example, the global image features may be any one of the first image brightness information, first image sharpness information, and first image clarity information; the global image features may also be any two of the first image brightness information, first image sharpness information, and first image clarity information; or the global image features may also be the first image brightness information, first image sharpness information, and first image clarity information.
[0060] In an optional embodiment, taking the combination of first image brightness information and first image sharpness information as an example, specifically, it can be achieved through... Figure 3 The steps in the document determine the global image features corresponding to the candidate images, referencing... Figure 3 As shown, it can specifically include:
[0061] Step S310: Divide the candidate image into regions according to the preset first image block size, and determine the first image block corresponding to the candidate image;
[0062] Step S320: Statistically analyze the brightness information in the first image block to determine the brightness information of the first image;
[0063] Step S330: Statistically analyze the sharpness information in the first image block to determine the sharpness information of the first image.
[0064] The first image brightness information refers to the statistical brightness information of the candidate image. For example, the brightness information corresponding to all pixels in the candidate image can be calculated, and the sum or average value of the brightness information of all pixels can be used as the first image brightness information of the candidate image. Of course, the candidate image can also be divided into regions, and the sum or average value of the brightness information of some image blocks in the candidate image can be used as the first image brightness information of the candidate image. This example embodiment does not make any special limitations on this.
[0065] Image sharpness refers to the metric data that reflects the clarity of an image plane and the sharpness of image edges. If the image sharpness is increased, the contrast of details on the image plane is also higher, making it look clearer. For example, under high sharpness, not only are wrinkles and blemishes on a person's face clearer, but the bulges or depressions of facial muscles can also be represented more realistically.
[0066] The first image sharpness information refers to the statistical information of the sharpness of the candidate image. For example, the candidate image can be divided into regions, and the sharpness information of each image block in the candidate image can be calculated. The sum or average of the sharpness information of all image blocks can be used as the first image sharpness information of the candidate image.
[0067] The first image block size refers to the parameter used to divide the candidate image into regions to facilitate the statistical analysis of the brightness or sharpness information of the first image in the candidate image. For example, the first image block size can be 16*16, and the specific unit can be set according to the candidate image. For example, the first image block size can be 16 pixels * 16 pixels, or 16 millimeters * 16 millimeters. Of course, the first image block size can also be 4*4, 8*8, etc. The specific setting can be customized according to the actual use. This example embodiment does not make any special limitation on this.
[0068] Candidate images can be divided into multiple image blocks according to the size of the first image block, namely the first image block. When calculating the brightness information or sharpness information of the first image, the calculation can be performed based on the pixel information in the first image block.
[0069] Optionally, a preset first jump block interval can be obtained, and then the brightness information in the first image block can be statistically analyzed based on the first jump block interval to determine the brightness information of the first image; the sharpness information in the first image block can be statistically analyzed based on the first jump block interval to determine the sharpness information of the first image.
[0070] The first skip block interval refers to a pre-set parameter used to collect global image features in the first image block. For example, the first skip block interval can be 1, that is, for the first image block in the candidate image, information can be collected every 1 image block. The first skip block interval can also be 2, 3, etc. The specific value can be customized according to the actual application situation. This embodiment is not limited to this.
[0071] After dividing the candidate image into multiple image blocks according to the first image block size, the first image block can be determined. By only counting the brightness or sharpness information of a subset of image blocks within the first image block, skip-block statistics can be performed to determine the brightness or sharpness information of the candidate image corresponding to the first image. For example, the first skip-block interval can be set to 1, meaning that skip-block statistics can be performed every other image block within the first image block. If the candidate image is divided into 100 first image blocks according to the first image block size, then after skip-block statistics are performed every other image block, 5 images can be determined within the first image block. Zero image blocks, and only counting the brightness or sharpness information of these 50 image blocks, can be considered as skipping block statistics; alternatively, skipping block statistics can be performed by arranging multiple image blocks at intervals. If the candidate image is divided into 100 first image blocks according to the first image block size, then by skipping blocks at intervals of 3 image blocks, 25 image blocks can be identified in the first image blocks, and only counting the brightness or sharpness information of these 25 image blocks can be considered as skipping block statistics; of course, the specific skipping block method can be customized according to the size of the candidate image or the preset first image block size, and this example embodiment is not limited to this.
[0072] Figure 4 The schematic diagram illustrates the principle of a jump block statistics method in an exemplary embodiment of the present disclosure.
[0073] refer to Figure 4As shown, the candidate image 400 or the region of interest 400 in the candidate image can be divided into regions according to the first image block size or the second image block size. For example, the first image block size or the second image block size can be set to 16*16, and the candidate image 400 or the region of interest 400 in the candidate image can be divided into 5*5 image blocks, each with a size of 16*16.
[0074] Optionally, the first jump block interval can be set to 1, which will result in image block 401 that needs to be statistically analyzed for image brightness or image sharpness, and image block 402 that does not need to be statistically analyzed for image brightness or image sharpness. Of course, the jump block interval can also be set to 2, 3, etc., and can be customized according to the actual use case. This example embodiment is not limited to this.
[0075] By dividing the candidate image into regions and performing block-skipping statistics within the determined first image block, the brightness and sharpness information of the first image of the candidate image can be determined. This can effectively reduce the amount of computation in the selection of the benchmark image, and improve the selection efficiency of the benchmark image while ensuring the accuracy of the selection.
[0076] Optionally, multiple target image blocks can be determined in the first image block by skipping blocks. When determining the brightness information of the first image, multiple pixels can be determined from the target image blocks. For example, the average value of the brightness information of each pixel in a 4*4 area near the center of the target image block can be used as the brightness information of the target image block. Alternatively, the average value of the brightness information of the pixels at the four corners of the target image block can be used as the brightness information of the target image block. By selecting some pixels in the target image block to determine the brightness information, the amount of calculation in the reference image screening process can be further reduced, and the screening efficiency of the reference image can be improved. Of course, the average value or median value of the brightness information of all pixels in the target image block can also be used as the brightness information of the target image block. This example embodiment does not impose any special limitations on this.
[0077] When determining the first image sharpness information, the pixels in the target image block can be binarized to determine the grayscale information of the target image block, and the image sharpness information corresponding to the target image block can be determined by relevant statistical functions. For example, the grayscale variance function can be used to determine the image sharpness information of the target image block. Specifically, the image sharpness information of the target image block can be calculated according to the relation (1):
[0078] D(f)=∑ y ∑ x (|f(x,y)-f(x,y-1)|+|f(x,y)-f(x+1,y)|) (1)
[0079] Where D(f) represents the image sharpness information in each target image block, x represents the horizontal coordinate of a pixel in the target image block, y represents the vertical coordinate of a pixel in the target image block, and f(x,y) represents the grayscale value of the pixel at coordinates (x,y) in the target image block. Of course, this is merely an illustrative example; this embodiment can also calculate the image sharpness information corresponding to the target image block in other ways, and this example embodiment does not impose any special limitations on this.
[0080] In one exemplary embodiment, the local image features may include any one or a combination of two of the following: second image brightness information, second image sharpness information, and second image clarity information corresponding to the region of interest. For example, the local image features may be any one of the second image brightness information, second image sharpness information, and second image clarity information; the local image features may also be any two of the second image brightness information, second image sharpness information, and second image clarity information; or the local image features may also be the second image brightness information, second image sharpness information, and second image clarity information.
[0081] In an optional embodiment, taking the combination of second image brightness information and second image sharpness information as an example, specifically, it can be achieved through... Figure 5 The steps in the document determine the local image features corresponding to the region of interest, referencing... Figure 5 As shown, it can specifically include:
[0082] Step S510: Divide the region of interest into regions according to the preset second image block size, and determine the second image block corresponding to the region of interest;
[0083] Step S520: Statistically analyze the brightness information in the second image block to determine the brightness information of the second image;
[0084] Step S530: Statistically analyze the sharpness information in the second image block to determine the sharpness information of the second image.
[0085] The second image brightness information refers to the brightness statistics of the region of interest in the candidate image. For example, the brightness information of all pixels in the region of interest can be calculated, and the sum or average value of the brightness information of all pixels can be used as the second image brightness information of the region of interest. Alternatively, the region of interest can be divided into regions, and the sum or average value of the brightness information of some image blocks in the region of interest can be used as the second image brightness information of the candidate image. This example embodiment does not impose any special limitations on this.
[0086] The second image sharpness information refers to the statistical information on the sharpness of the region of interest in the candidate image. For example, the region of interest can be divided into regions, and the sharpness information of each image block in the region of interest can be calculated. The sum or average of the sharpness information of all image blocks can be used as the second image sharpness information of the candidate image.
[0087] The second image block size refers to the parameter used to divide the region of interest (ROI) to facilitate the statistical analysis of the brightness or sharpness information of the second image within the ROI. For example, the second image block size can be 16*16, and the specific unit can be set according to the candidate image. For instance, the second image block size can be 16 pixels * 16 pixels, or 16 millimeters * 16 millimeters. Of course, the second image block size can also be 4*4, 8*8, etc. The specific setting can be customized according to the actual use. This example embodiment does not impose any special limitations on this.
[0088] It is understood that the second image block size can be set to the same image block size as the first image block size, or it can be set to a different image block size than the first image block size. This example embodiment does not impose any special limitations on this.
[0089] Optionally, a preset second jump block interval can be obtained, and then the brightness information in the second image block can be statistically analyzed based on the second jump block interval to determine the brightness information of the second image; the sharpness information in the second image block can be statistically analyzed based on the second jump block interval to determine the sharpness information of the second image.
[0090] It is understood that the method for determining the brightness information and sharpness information of the second image in this embodiment is the same as the method for determining the brightness information and sharpness information of the first image in the previous embodiment. The method for calculating the jump block statistics is also the same as the method for calculating the jump block statistics of the first image brightness information and sharpness information. For details, please refer to the above embodiments, which will not be repeated here.
[0091] It is understood that the first hop block interval and the second hop block interval can be set to the same interval size or different interval sizes. For example, the first hop block interval used for statistical analysis of global image features can be set to 2, and the second hop block interval used for statistical analysis of local image features can be set to 1. The specific settings can be made according to the application situation, and this embodiment does not impose any special limitations on this.
[0092] It should be noted that in the embodiments of this disclosure, the "first" and "second" in "first image brightness information" and "second image brightness information", "first image sharpness information" and "second image sharpness information", "first image block size" and "second image block size", "first jump block interval" and "second jump block interval" are only used to distinguish the image brightness information, image sharpness information, image block size and jump block interval corresponding to the global candidate image and the region of interest in the candidate image, and have no special meaning, and should not impose any special limitations on this example embodiment.
[0093] In an exemplary embodiment, the region of interest may include a face image region. For example, if the candidate image already contains face bounding box information, the face bounding box information can be directly used as the face image region of the candidate image, i.e., the region of interest of the candidate image. If the candidate image does not contain face bounding box information, face detection can be performed on the candidate image to determine the face image region in the candidate image. This example embodiment is not limited thereto.
[0094] In an exemplary embodiment, the image evaluation data may include image brightness data and image sharpness data. The determination of image evaluation data for each candidate image based on global image features and local image features can be achieved through the following steps: obtaining preset first weight data and second weight data; calculating the image brightness data by weighting the first image brightness information of the candidate image and the second image brightness information of the region of interest based on the first weight data; and calculating the image sharpness data by weighting the first image sharpness information of the candidate image and the second image sharpness information of the region of interest based on the second weight data.
[0095] The first weight data refers to the weight data used to calculate the first image brightness information of the candidate image and the second image brightness information of the region of interest. For example, the first weight data can be 0.5 and 0.5, or it can be 0.4 and 0.6. The first weight data can be customized according to the actual use case. This example embodiment does not make any special limitation on this.
[0096] The second weight data refers to the weight data used to weight the first image sharpness information of the candidate image and the second image sharpness information of the region of interest. It can be understood that the value of the second weight data can be set to the same weight as the first weight data, or it can be set to a different weight. The specific setting can be customized according to the actual use case. This example embodiment does not make any special limitation on this.
[0097] It should be noted that the terms "first" and "second" in "first weight data" and "second weight data" in this example embodiment are only used to distinguish the weight data for weighted calculation of image brightness information and image sharpness information, and have no special meaning, and should not impose any special limitations on this example embodiment.
[0098] Figure 6 The illustration schematically shows a process diagram for selecting a reference image in an exemplary embodiment of the present disclosure.
[0099] refer to Figure 6 As shown, in step S610, at least two candidate images are acquired continuously.
[0100] Step S620: Determine whether the candidate image contains a region of interest. If it does, proceed to step S650; otherwise, proceed to step S630.
[0101] Step S630: Divide the candidate image into regions according to the preset first image block size;
[0102] Step S640: Jump block statistics of global image features corresponding to candidate images. These global image features may include first image brightness information and first image sharpness information.
[0103] Step S650: Divide the region of interest into regions according to the preset second image block size;
[0104] Step S660: Jump block statistics of local image features corresponding to the region of interest. These local image features may include second image brightness information and second image sharpness information.
[0105] Step S670: Determine image evaluation data based on global image features and local image features. The image evaluation data may include image brightness data and image sharpness data. For example, image brightness data can be obtained by weighting first image brightness information and second image brightness information using first weight data; image sharpness data can be obtained by weighting first image sharpness information and second image sharpness information using second weight data.
[0106] Step S680: Based on the image evaluation data, a benchmark image is selected from the candidate images, and the current process ends.
[0107] In an exemplary embodiment, before screening the benchmark image among the candidate images based on image evaluation data, the candidate images may be preliminarily screened through the following steps, which may specifically include: determining the brightness difference between the image brightness data of any target candidate image and the image brightness data of other candidate images; if the brightness difference is determined to be greater than or equal to a brightness difference threshold, the target candidate image is removed from the candidate images.
[0108] By calculating the brightness difference between any candidate image and other candidate images, if the brightness difference is greater than or equal to the brightness difference threshold, the target candidate image can be considered an erroneous image among all candidate images. If it participates in the subsequent multi-frame synthesis task, it may lead to a deterioration of the multi-frame synthesis result. Therefore, it needs to be removed to further ensure the accuracy of the baseline image selection.
[0109] In one exemplary embodiment, it can be achieved through Figure 7 The steps described herein enable the selection of a benchmark image from among candidate images based on image evaluation data, with reference to... Figure 7 As shown, it can specifically include:
[0110] Step S710: Sort the candidate images according to the image sharpness data of the candidate images, and determine the sharpest candidate image corresponding to the maximum image sharpness data and the second sharpest candidate image corresponding to the second maximum image sharpness data.
[0111] Step S720: If the second image sharpness information corresponding to the first region of interest in the sharpest candidate image is the maximum value among the second image sharpness information corresponding to all regions of interest, then the sharpest candidate image is used as the reference image; or
[0112] Step S730: If the second image sharpness information corresponding to the second region of interest in the sub-sharp candidate image is the maximum value among the second image sharpness information corresponding to all regions of interest, then the sub-sharp candidate image is used as the reference image.
[0113] Among them, the sharpest candidate image refers to the candidate image with the highest image sharpness information among the sorted candidate images, and the second sharpest candidate image refers to the candidate image with the second highest image sharpness information among the sorted candidate images. For example, the candidate images can be sorted in descending order of image sharpness information, and the candidate image with the highest image sharpness information can be taken as the sharpest candidate image, and the candidate image with the second highest image sharpness information can be taken as the second sharpest candidate image. Of course, the candidate images can also be sorted in ascending order of image sharpness information, and this example embodiment does not make any special limitation on this.
[0114] The first region of interest (ROI) refers to the region of interest in the sharpest candidate image, and the second region of interest refers to the region of interest in the next sharpest candidate image. It should be noted that the terms "first" and "second" in the "first region of interest" and "second region of interest" in this embodiment are merely used to distinguish the corresponding regions of interest in the sharpest and next sharpest candidate images, and have no special meaning, nor should they impose any special limitations on this example embodiment.
[0115] First, candidate images are initially screened using image brightness data from the image evaluation data. Then, the candidate images are ranked using image sharpness data from the image evaluation data to determine the sharpest and second-sharpest candidate images. Finally, the final benchmark image is determined based on the image sharpness information of the regions of interest in the sharpest and second-sharpest candidate images. By using both image brightness data and image sharpness data as evaluation indicators for screening benchmark images, compared with a single evaluation indicator, the interference of image content or image noise on the screening results can be effectively eliminated, further ensuring the accuracy of the screening benchmark image and improving the robustness of the screening results.
[0116] In an exemplary embodiment, if it is determined that there is no region of interest in the candidate image, the image evaluation data corresponding to each candidate image can be determined based on the global image features. That is, when it is determined that there is no region of interest in the candidate image, the global image features of the candidate image can be directly used as the evaluation index of each candidate image.
[0117] Optionally, candidate images can be sorted based on their image sharpness data to determine the sharpest candidate image corresponding to the highest image sharpness data, and the sharpest candidate image can be used as the reference image.
[0118] Figure 8 This schematically illustrates another process for selecting benchmark images in an exemplary embodiment of the present disclosure.
[0119] refer to Figure 8 As shown, in step S810, the image brightness data and image sharpness data corresponding to multiple candidate images can be determined, i.e., image evaluation data.
[0120] Step S820: Determine whether the image brightness data of the target candidate image matches other candidate images. Specifically, determine whether the measurement difference between the image brightness data of the target candidate image and the image brightness data of other candidate images is greater than or equal to the brightness difference threshold. If they match, proceed to step S830; otherwise, proceed to step S840.
[0121] Step S830: Sort the candidate images according to image sharpness information;
[0122] Step S840: Remove the target candidate images and proceed to step S830;
[0123] Step S850: Determine the sharpest candidate image and the second sharpest candidate image;
[0124] Step S860: Determine whether there is a face image region in the candidate image. If there is, proceed to step S870; otherwise, proceed to step S880.
[0125] Step S870: If the image sharpness information of the face image region in the sharpest candidate image is the maximum value, then the sharpest candidate image is used as the reference image; if the image sharpness information of the face image region in the second sharpest candidate image is the maximum value, then the second sharpest candidate image is used as the reference image, and the current process ends.
[0126] Step S880: Use the sharpest candidate image as the reference image and end the current process.
[0127] In summary, this exemplary embodiment can acquire at least two consecutively captured candidate images and determine the global image features corresponding to the candidate images. Then, when a region of interest (ROI) is determined within a candidate image, the local image features corresponding to that ROI can be determined. Furthermore, based on the global and local image features, image evaluation data corresponding to each candidate image can be determined, and a benchmark image is selected from among the candidate images based on this image evaluation data. When selecting benchmark images, not only are the global image features of the candidate images considered, but also the local image features of the ROI within the candidate images. The benchmark images are selected using image evaluation data jointly determined by the global and local image features, ensuring that the selected benchmark images accurately reflect the relevant information of the captured content, thus improving the accuracy of benchmark image selection. Simultaneously, selecting benchmark images based on general global and local image features is applicable to various types of candidate images, expanding the applicability of the selection scheme.
[0128] It should be noted that the above figures are merely illustrative of the processes included in the method according to exemplary embodiments of this disclosure, and are not intended to be limiting. It is readily understood that the processes shown in the above figures do not indicate or limit the temporal order of these processes. Furthermore, it is readily understood that these processes may be executed synchronously or asynchronously, for example, in multiple modules.
[0129] Further reference Figure 9 As shown, this example embodiment also provides a benchmark image screening device 900, including a global image feature determination module 910, a local image feature determination module 920, an image evaluation data determination module 930, and a benchmark image screening module 940. Wherein:
[0130] The global image feature determination module 910 can be used to acquire at least two candidate images and determine the global image features corresponding to the candidate images;
[0131] The local image feature determination module 920 can be used to determine the region of interest in the candidate image and the local image features corresponding to the region of interest;
[0132] The image evaluation data determination module 930 can be used to determine the image evaluation data corresponding to each candidate image based on the global image features and the local image features;
[0133] The benchmark image filtering module 940 can be used to filter benchmark images from the candidate images based on the image evaluation data.
[0134] In one exemplary embodiment, the global image features may include any one or a combination of first image brightness information, first image sharpness information, and first image clarity information.
[0135] In an exemplary embodiment, the global image features may include first image brightness information and first image sharpness information, and the global image feature determination module 910 may be used to:
[0136] The candidate image is divided into regions according to a preset first image block size to determine the first image block corresponding to the candidate image;
[0137] The brightness information in the first image block is statistically analyzed to determine the brightness information of the first image.
[0138] The sharpness information in the first image block is statistically analyzed to determine the sharpness information of the first image.
[0139] In an exemplary embodiment, the global image feature determination module 910 can be used to:
[0140] Get the preset first jump block interval;
[0141] Based on the first jump block interval, the brightness information in the first image block is statistically analyzed to determine the brightness information of the first image;
[0142] Based on the first jump block interval, the sharpness information in the first image block is statistically analyzed by jump block analysis to determine the first image sharpness information.
[0143] In one exemplary embodiment, the local image features may include any one or a combination of two or more of the following: second image brightness information, second image sharpness information, and second image clarity information.
[0144] In an exemplary embodiment, the local image features may include second image brightness information and second image sharpness information, and the local image feature determination module 920 may be used to:
[0145] The region of interest is divided into regions according to a preset second image block size, and the second image block corresponding to the region of interest is determined.
[0146] The brightness information in the second image block is statistically analyzed to determine the brightness information of the second image.
[0147] The sharpness information in the second image block is statistically analyzed to determine the sharpness information of the second image.
[0148] In an exemplary embodiment, the local image feature determination module 920 can be used to:
[0149] Get the preset second jump block interval;
[0150] Based on the second jump block interval, the brightness information in the second image block is statistically analyzed to determine the brightness information of the second image;
[0151] Based on the second jump block interval, the sharpness information in the second image block is statistically analyzed by jump block analysis to determine the second image sharpness information.
[0152] In an exemplary embodiment, the image evaluation data may include image brightness data and image sharpness data, and the image evaluation data determination module 930 may be used to:
[0153] Obtain the preset first weight data and second weight data;
[0154] Based on the first weight data, the first image brightness information of the candidate image and the second image brightness information of the region of interest are weighted and calculated to determine the image brightness data;
[0155] Based on the second weight data, the first image sharpness information of the candidate image and the second image sharpness information of the region of interest are weighted and calculated to determine the image sharpness data.
[0156] In an exemplary embodiment, the reference image filtering device 900 may further include a candidate image rejection module, which may be used to:
[0157] Determine the brightness difference between the image brightness data of any target candidate image and the image brightness data of other candidate images;
[0158] If the brightness difference is determined to be greater than or equal to the brightness difference threshold, the target candidate image is removed from the candidate images.
[0159] In one exemplary embodiment, the reference image filtering module 940 can be used to:
[0160] The candidate images are sorted according to their image sharpness data to determine the sharpest candidate image corresponding to the maximum image sharpness data and the second sharpest candidate image corresponding to the second maximum image sharpness data.
[0161] If the second image sharpness information corresponding to the first region of interest in the sharpest candidate image is the maximum value among the second image sharpness information corresponding to all regions of interest, then the sharpest candidate image is used as the reference image; or
[0162] If the second image sharpness information corresponding to the second region of interest in the sub-sharp candidate image is the maximum value among the second image sharpness information corresponding to all regions of interest, then the sub-sharp candidate image is used as the reference image.
[0163] In one exemplary embodiment, the reference image filtering device 900 can also be used for:
[0164] If it is determined that there is no region of interest in the candidate image, then the image evaluation data corresponding to each candidate image is determined based on the global image features.
[0165] In one exemplary embodiment, the reference image filtering module 940 can also be used for:
[0166] The candidate images are sorted according to their image sharpness data, and the sharpest candidate image corresponding to the maximum image sharpness data is determined.
[0167] The sharpest candidate image is used as the reference image.
[0168] The specific details of each module in the above-mentioned device have been described in detail in the method section of the implementation. For any undisclosed details, please refer to the implementation content of the method section, and therefore will not be repeated here.
[0169] Those skilled in the art will understand that various aspects of this disclosure can be implemented as a system, method, or program product. Therefore, various aspects of this disclosure can be specifically implemented in the following forms: a completely hardware implementation, a completely software implementation (including firmware, microcode, etc.), or a combination of hardware and software aspects, collectively referred to herein as a "circuit," "module," or "system."
[0170] Exemplary embodiments of this disclosure provide an electronic device for implementing a benchmark image selection method, which may be... Figure 1 The terminal devices 101, 102, 103, or server 105 are included. The electronic device includes at least a processor and a memory, the memory being used to store executable instructions of the processor, and the processor being configured to perform a reference image screening method by executing the executable instructions.
[0171] The following is based on Figure 10 Taking the electronic device 1000 as an example, the construction of the electronic device in this disclosure will be described by way of example. Figure 10The electronic device 1000 shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments disclosed herein.
[0172] like Figure 10 As shown, the electronic device 1000 is presented in the form of a general-purpose computing device. The components of the electronic device 1000 may include, but are not limited to: at least one processing unit 1010, at least one storage unit 1020, a bus 1030 connecting different system components (including storage unit 1020 and processing unit 1010), and a display unit 1040.
[0173] The storage unit 1020 stores program code, which can be executed by the processing unit 1010, causing the processing unit 1010 to perform the reference image screening method in this specification.
[0174] Storage unit 1020 may include a readable medium in the form of a volatile storage unit, such as a random access memory unit (RAM) 1021 and / or a cache memory unit 1022, and may further include a read-only memory unit (ROM) 1023.
[0175] Storage unit 1020 may also include a program / utility 1024 having a set (at least one) program module 1025, such program module 1025 including but not limited to: operating system, one or more application programs, other program modules and program data, each or some combination of these examples may include an implementation of a network environment.
[0176] Bus 1030 can represent one or more of several types of bus structures, including a memory cell bus or memory cell controller, a peripheral bus, a graphics acceleration port, a processing unit, or a local bus using any of the multiple bus structures.
[0177] Electronic device 1000 can also communicate with one or more external devices 1070 (e.g., sensor devices, Bluetooth devices, etc.), one or more devices that enable users to interact with electronic device 1000, and / or any device that enables electronic device 1000 to communicate with one or more other computing devices (e.g., routers, modems, etc.). This communication can be performed via input / output (I / O) interface 1050. Furthermore, electronic device 1000 can also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via network adapter 1060. As shown, network adapter 1060 communicates with other modules of electronic device 1000 via bus 1030. It should be understood that, although not shown in the figures, other hardware and / or software modules can be used in conjunction with electronic device 1000, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, data backup storage systems, and sensor modules (e.g., gyroscope sensors, magnetometers, accelerometers, distance sensors, proximity sensors, etc.).
[0178] From the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein can be implemented by software or by combining software with necessary hardware. Therefore, the technical solutions according to the embodiments of this disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) or on a network, including several instructions to cause a computing device (such as a personal computer, server, terminal device, or network device, etc.) to execute the methods according to the embodiments of this disclosure.
[0179] Exemplary embodiments of this disclosure also provide a computer-readable storage medium having a program product stored thereon capable of implementing the methods described above in this specification. In some possible embodiments, various aspects of this disclosure may also be implemented as a program product including program code that, when the program product is run on a terminal device, causes the terminal device to perform the steps described in the "Exemplary Methods" section of this specification according to various exemplary embodiments of this disclosure.
[0180] It should be noted that the computer-readable medium disclosed herein may be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium may be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof.
[0181] In this disclosure, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in connection with an instruction execution system, apparatus, or device. In this disclosure, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, which can transmit, propagate, or transfer a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wireless, wireline, optical fiber, RF, etc., or any suitable combination thereof.
[0182] Furthermore, program code for performing the operations of this disclosure can be written in any combination of one or more programming languages, including object-oriented programming languages such as Java and C++, and conventional procedural programming languages such as C or similar languages. The program code can execute entirely on the user's computing device, partially on the user's computing device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).
[0183] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and embodiments are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the claims.
[0184] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.
Claims
1. A method for selecting a reference image, characterized in that, include: At least two candidate images are acquired, and the global image features corresponding to the candidate images are determined; wherein, the global image features include any one or more combinations of first image brightness information, first image sharpness information, and first image clarity information; The region of interest (ROI) in the candidate image and the corresponding local image features are determined; wherein, the local image features include any one or a combination of second image brightness information, second image sharpness information, and second image clarity information; the ROI includes a face image region; Based on the global image features and the local image features, determine the image evaluation data corresponding to each candidate image; Based on the image evaluation data, a benchmark image is selected from the candidate images, including: The image sharpness data of each candidate image is determined based on the first image sharpness information and the second image sharpness information of the candidate images, and the candidate images are sorted based on the image sharpness data to determine the sharpest candidate image corresponding to the maximum image sharpness data and the second sharpest candidate image corresponding to the second maximum image sharpness data. If the second image sharpness information corresponding to the first region of interest in the sharpest candidate image is the maximum value among the second image sharpness information corresponding to all regions of interest, then the sharpest candidate image is used as the reference image; or If the second image sharpness information corresponding to the second region of interest in the sub-sharp candidate image is the maximum value among the second image sharpness information corresponding to all regions of interest, then the sub-sharp candidate image is used as the reference image.
2. The method according to claim 1, characterized in that, The global image features include first image brightness information and first image sharpness information. Determining the global image features corresponding to the candidate image includes: The candidate image is divided into regions according to a preset first image block size to determine the first image block corresponding to the candidate image; The brightness information in the first image block is statistically analyzed to determine the brightness information of the first image. The sharpness information in the first image block is statistically analyzed to determine the sharpness information of the first image.
3. The method according to claim 2, characterized in that, Determining the global image features corresponding to the candidate image includes: Get the preset first jump block interval; Based on the first jump block interval, the brightness information in the first image block is statistically analyzed to determine the brightness information of the first image; Based on the first jump block interval, the sharpness information in the first image block is statistically analyzed by jump block analysis to determine the first image sharpness information.
4. The method according to claim 1, characterized in that, The local image features include second image brightness information and second image sharpness information. Determining the local image features corresponding to the region of interest includes: The region of interest is divided into regions according to a preset second image block size, and the second image block corresponding to the region of interest is determined. The brightness information in the second image block is statistically analyzed to determine the brightness information of the second image. The sharpness information in the second image block is statistically analyzed to determine the sharpness information of the second image.
5. The method according to claim 4, characterized in that, Determining the local image features corresponding to the region of interest includes: Get the preset second jump block interval; Based on the second jump block interval, the brightness information in the second image block is statistically analyzed to determine the brightness information of the second image; Based on the second jump block interval, the sharpness information in the second image block is statistically analyzed by jump block analysis to determine the second image sharpness information.
6. The method according to claim 5, characterized in that, The image evaluation data includes image brightness data and image sharpness data. Determining the image evaluation data corresponding to each candidate image based on the global image features and the local image features includes: Obtain the preset first weight data and second weight data; Based on the first weight data, the first image brightness information of the candidate image and the second image brightness information of the region of interest are weighted and calculated to determine the image brightness data; Based on the second weight data, the first image sharpness information of the candidate image and the second image sharpness information of the region of interest are weighted and calculated to determine the image sharpness data.
7. The method according to claim 6, characterized in that, The method further includes: Determine the brightness difference between the image brightness data of any target candidate image and the image brightness data of other candidate images; If the brightness difference is determined to be greater than or equal to the brightness difference threshold, the target candidate image is removed from the candidate images.
8. A reference image screening device, characterized in that, include: A global image feature determination module is used to acquire at least two candidate images and determine the global image features corresponding to the candidate images; wherein, the global image features include any one or more combinations of first image brightness information, first image sharpness information and first image clarity information; A local image feature determination module is used to determine the region of interest (ROI) in the candidate image and the local image features corresponding to the ROI; wherein, the local image features include any one or a combination of two or more of the following: second image brightness information, second image sharpness information, and second image clarity information; the ROI includes a face image region; The image evaluation data determination module is used to determine the image evaluation data corresponding to each candidate image based on the global image features and the local image features. A benchmark image filtering module is used to filter benchmark images from the candidate images based on the image evaluation data, including: The image sharpness data of each candidate image is determined based on the first image sharpness information and the second image sharpness information of the candidate images, and the candidate images are sorted based on the image sharpness data to determine the sharpest candidate image corresponding to the maximum image sharpness data and the second sharpest candidate image corresponding to the second maximum image sharpness data. If the second image sharpness information corresponding to the first region of interest in the sharpest candidate image is the maximum value among the second image sharpness information corresponding to all regions of interest, then the sharpest candidate image is used as the reference image; or If the second image sharpness information corresponding to the second region of interest in the sub-sharp candidate image is the maximum value among the second image sharpness information corresponding to all regions of interest, then the sub-sharp candidate image is used as the reference image.
9. A computer-readable medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1 to 7.
10. An electronic device, characterized in that, include: processor; as well as Memory for storing the executable instructions of the processor; The processor is configured to execute the method of any one of claims 1 to 7 by executing the executable instructions.