Image definition identification method and device, equipment and storage medium

Abstract

The invention provides an image definition recognition method and device, equipment and a storage medium, and relates to the technical field of image definition recognition, and the method comprises the steps: constructing a Laplacian pyramid corresponding to an input image, and obtaining a first edge image and a second edge image corresponding to the input image; fusing the first edge image and the second edge image to obtain edge information after the first edge image and the second edge image are fused; determining a pixel intensity fluctuation ratio corresponding to the input image according to the edge information; under the condition that the pixel intensity fluctuation ratio is greater than a first preset pixel intensity fluctuation ratio threshold value, determining that the definition of the to-be-identified and to-be-analyzed image is a first definition; through multi-scale edge fusion, high-precision, high-efficiency and high-robustness image definition judgment is realized, and the problems that a traditional method is susceptible to noise interference, incomplete in detail capture, single in evaluation index and the like in a complex scene are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of image sharpness recognition, and particularly to a method, device, equipment and storage medium for recognizing image sharpness. Background Art

[0002] Studying image sharpness evaluation methods has profound significance and purpose, and plays a key role especially in multiple fields such as vehicle cockpit vision systems, image processing, video coding and transmission, medical imaging, and autonomous driving. In image and video processing, automatically evaluating image sharpness can be used for quality control to ensure that the image maintains sufficient sharpness during processes such as transmission, compression, and enhancement. This is particularly important for application scenarios such as video streaming services, telemedicine, and security monitoring. Additionally, sharpness evaluation can help improve and optimize image processing algorithms, such as denoising, sharpening, super-resolution reconstruction, etc. By quantifying image sharpness, it can guide algorithm design to produce the best results under different conditions.

[0003] Traditional methods for determining image sharpness mainly include the gradient method: evaluating sharpness by calculating the intensity of image gradients (such as Sobel and Laplacian operators). However, it is sensitive to noise and vulnerable to interference in complex texture scenes, resulting in misjudgments; the frequency domain analysis method: using Fourier transform or discrete cosine transform (DCT) to extract high-frequency energy as the sharpness index. However, the frequency domain method has high computational complexity and is difficult to distinguish noise from real details; single-scale edge detection: based on edge extraction at a fixed scale (such as the Canny operator) combined with statistical features (such as variance, entropy) for determination. However, a single scale cannot comprehensively capture detail features of different sizes in the image and has poor adaptability to blur types (such as motion blur, defocus blur). In view of the above problems, there is an urgent need for an efficient, robust, and multi-scale-detail-adaptive sharpness determination method. Summary of the Invention

[0004] In view of the above technical problems, the present invention provides a method, device, equipment and storage medium for recognizing image sharpness to solve the technical problems that the existing methods for determining image sharpness are inefficient and cannot adapt to the multi-scale details of images, and achieve the effects of high efficiency and adaptability to multi-scale details; the technical solution adopted by the present invention is as follows:

[0005] According to the first aspect of the present application, a method for recognizing image sharpness is provided, and the method includes:

[0006] Obtain an input image corresponding to the image to be recognized;

[0007] Construct a Laplacian pyramid corresponding to the input image to obtain a first edge image and a second edge image corresponding to the input image;

[0008] Fuse the first edge image and the second edge image to obtain the edge information after fusing the first edge image and the second edge image;

[0009] Determine the pixel intensity volatility corresponding to the input image according to the edge information;

[0010] When the pixel intensity volatility is greater than the first preset pixel intensity volatility threshold, determine that the clarity of the image to be recognized is the first clarity.

[0011] Further, constructing the Laplacian pyramid corresponding to the input image to obtain the first edge image and the second edge image corresponding to the input image includes:

[0012] Perform one downsampling on the input image to obtain a first intermediate image;

[0013] Perform one downsampling on the first intermediate image to obtain a second intermediate image;

[0014] Perform one upsampling on the first intermediate image to obtain a third intermediate image;

[0015] Perform two upsamplings on the second intermediate image to obtain a fourth intermediate image;

[0016] Calculate the difference between the input image and the third intermediate image to obtain the first edge image;

[0017] Calculate the difference between the input image and the fourth intermediate image to obtain the second edge image.

[0018] Further, fusing the first edge image and the second edge image includes:

[0019] Obtain a first fusion weight corresponding to the first edge image and a second fusion weight corresponding to the second edge image; wherein, the first fusion weight is greater than the second fusion weight;

[0020] Assign the first fusion weight to the first edge image and the second fusion weight to the second edge image for weighted fusion.

[0021] Further, the method further includes:

[0022] When the pixel intensity volatility corresponding to the input image is greater than or equal to the second preset pixel intensity volatility threshold and less than or equal to the first preset pixel intensity volatility threshold, determine that the clarity of the image to be recognized is the second clarity; and / or,

[0023] When the pixel intensity volatility corresponding to the input image is less than the second preset pixel intensity volatility threshold, determine the clarity of the image to be recognized as the third clarity.

[0024] Further, the third clarity is less than the second clarity, and the second clarity is less than the first clarity.

[0025] Further, obtaining the input image corresponding to the image to be recognized includes: performing grayscale processing and image denoising processing based on wavelet transform on the image to be recognized in sequence to obtain the input image.

[0026] Further, the sum of the first fusion weight and the second fusion weight is equal to 1.

[0027] According to another aspect of the present application, there is also provided an apparatus for recognizing image clarity, the apparatus includes:

[0028] An input image acquisition module, configured to acquire an input image corresponding to the image to be recognized;

[0029] A Laplacian pyramid construction module, configured to construct a Laplacian pyramid corresponding to the input image to obtain a first edge image and a second edge image corresponding to the input image;

[0030] An edge information fusion module, configured to fuse the first edge image and the second edge image to obtain edge information after fusing the first edge image and the second edge image;

[0031] A pixel intensity volatility determination module, configured to determine the pixel intensity volatility corresponding to the input image according to the edge information;

[0032] A first clarity determination module, configured to determine the clarity of the image to be recognized as the first clarity when the pixel intensity volatility is greater than a first preset pixel intensity volatility threshold.

[0033] According to another aspect of the present application, there is also provided a non-transitory computer-readable storage medium, in which at least one instruction or at least one program segment is stored, and at least one instruction or at least one program segment is loaded and executed by a processor to implement the above-mentioned method for recognizing image clarity.

[0034] According to another aspect of the present application, there is also provided an electronic device, including a processor and the above-mentioned non-transitory computer-readable storage medium.

[0035] The present invention has at least the following beneficial effects:

[0036] The method for identifying the image sharpness of the present invention constructs a corresponding Laplacian pyramid for the input image to obtain a first edge image and a second edge image, fuses the first edge image and the second edge image to obtain edge information, determines the pixel intensity volatility corresponding to the input image according to the edge information, and further determines the sharpness of the image to be identified; through multi-scale edge fusion, the present invention realizes high-precision, high-efficiency, and strong-robust image sharpness determination, and solves the problems of traditional methods being vulnerable to noise interference, incomplete detail capture, and single evaluation index in complex scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0038] Figure 1 It is a flowchart of the method for identifying the image sharpness provided by the embodiment of the present invention;

[0039] Figure 2 It is a specific implementation flowchart of the method for identifying the image sharpness provided by the embodiment of the present invention;

[0040] Figure 3 It is a schematic structural diagram of the device for identifying the image sharpness provided by the embodiment of the present invention;

[0041] Figure 4 It is a schematic diagram of the application scenario of the method for intelligent execution of tasks in a home scenario provided by the embodiment of the present invention;

[0042] Figure 5 It is a schematic structural diagram of the electronic device provided by the embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0043] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0044] It should be noted that based on this disclosure, those skilled in the art should understand that one aspect described herein can be implemented independently of any other aspect, and two or more of these aspects can be combined in various ways. For example, any number of aspects described herein can be used to implement the device and / or practice the method. Additionally, this device can be implemented and this method can be practiced using other structures and / or functionality in addition to one or more of the aspects described herein.

[0045] Studying image sharpness evaluation methods has profound significance and purposes, and plays a key role especially in multiple fields such as computer vision, image processing, video coding and transmission, medical imaging, and autonomous driving. In image and video processing, automatically evaluating image sharpness can be used for quality control to ensure that the image maintains sufficient sharpness during processes such as transmission, compression, and enhancement. This is particularly important for application scenarios such as video streaming services, telemedicine, and security monitoring. Additionally, sharpness evaluation can help improve and optimize image processing algorithms, such as denoising, sharpening, super-resolution reconstruction, and other techniques. By quantifying image sharpness, it can guide algorithm design to produce optimal results under different conditions.

[0046] Next, with reference to Figure 1 the flowchart of the method for identifying image sharpness shown, a method for identifying image sharpness will be introduced.

[0047] The method for identifying image sharpness may include the following steps:

[0048] Step S100, obtain an input image corresponding to the image to be identified.

[0049] In this embodiment, the image to be identified can be understood as the original image. For example, automotive vision devices are usually installed in intelligent vehicles, and the images captured by the automotive vision devices can be the images to be identified, which can be used to determine the usability of the road images captured by the camera in real time to ensure decision-making safety; after obtaining the image to be identified, preprocessing of the image to be identified is required.

[0050] Further preprocessing of the image to be identified includes: performing grayscale processing on the image to be identified and image denoising processing based on wavelet transform.

[0051] In this embodiment, grayscaling of the image refers to the process of converting a color image into a grayscale image. In a grayscale image, each pixel is represented by a single grayscale value instead of the RGB three-channel values in a color image. Grayscaling is usually used in image processing and computer vision tasks because grayscale images have lower computational costs compared to color images and also simplify the process of extracting image features.

[0052] Wavelet transform can well distinguish signals and noises in different frequency domains. During processing, wavelet coefficients at low resolutions (large scales) are retained, while wavelet coefficients at high resolutions (small scales) are selected or discarded according to the determined threshold. Since the wavelet transform of noise is mainly concentrated in various small-scale levels, after the above processing, the noise is basically filtered out, and edges and details can be better retained. Wavelet image denoising has been regarded as an important denoising algorithm in image processing.

[0053] Step S200: Construct the Laplacian pyramid corresponding to the input image to obtain the first edge image and the second edge image corresponding to the input image.

[0054] Further, the constructing the Laplacian pyramid corresponding to the input image to obtain the first edge image and the second edge image corresponding to the input image includes:

[0055] Perform one downsampling on the input image to obtain the first intermediate image.

[0056] Perform one downsampling on the first intermediate image to obtain the second intermediate image.

[0057] Perform one upsampling on the first intermediate image to obtain the third intermediate image.

[0058] Perform two upsamplings on the second intermediate image to obtain the fourth intermediate image.

[0059] Calculate the difference between the input image and the third intermediate image to obtain the first edge image.

[0060] Calculate the difference between the input image and the fourth intermediate image to obtain the second edge image.

[0061] In this embodiment, the first intermediate image: The input image undergoes one downsampling, and its size is 1 / 4 of the original, the resolution is reduced, some high-frequency details are lost, and more low-frequency information is retained.

[0062] The second intermediate image: The first intermediate image undergoes another downsampling, and its size is reduced to 1 / 4 again. At this time, the resolution of the second intermediate image is lower, and the retained information is more biased towards lower-frequency structures.

[0063] The third intermediate image: The first intermediate image undergoes one upsampling, and its size is restored to the original size. For example: If the first intermediate image is 1 / 4 the size of the input image, then the upsampled third intermediate image may be restored to the original size, but it will be blurred because upsampling usually uses interpolation and cannot fully restore the lost high-frequency information.

[0064] Fourth intermediate image: The second intermediate image is upsampled twice. The first upsampling is to the size of the first intermediate image, and the second upsampling is to the size of the input image. There will also be blurring, which is more blurred than the third intermediate image because the second intermediate image itself is already the result of two downsamplings and has lost more high-frequency information.

[0065] Then, a subtraction operation is performed: First edge image = input image - third intermediate image. At this time, the third intermediate image is the upsampled version of the first intermediate image. The first intermediate image is the image after being downsampled once. After upsampling and subtracting from the input image, the obtained edge image contains the high-frequency information lost in the first downsampling, that is, relatively high-frequency edge details.

[0066] Second edge image = input image - fourth intermediate image. The fourth intermediate image is the image after the second intermediate image is upsampled twice. The second intermediate image is the image after being downsampled twice and has lost more high-frequency information. Therefore, the fourth intermediate image after two upsamplings is more blurred. The edge image obtained by subtracting from the input image contains lower-frequency structural differences, or because of more upsampling times, the accumulated error may be larger, resulting in coarser edge information.

[0067] The scale changes of the first edge image and the second edge image are different. The more times of downsampling, the more the retained information tends to be low-frequency. Therefore, the image (the second intermediate image) that has undergone more downsamplings reflects larger-scale structural changes after restoration compared to the original image, while the image (the first intermediate image) that has only been downsampled once reflects smaller-scale detail changes after restoration.

[0068] For example, assume the use of downsampling and upsampling of the Gaussian pyramid:

[0069] Each time of downsampling (pyrDown) will first perform Gaussian blurring, and then remove even rows and columns, and the size becomes 1 / 4 of the original.

[0070] Each time of upsampling (pyrUp) will insert zeros or perform interpolation, and then perform Gaussian blurring, and the size doubles.

[0071] Therefore, the third intermediate image is upsampled from the first intermediate image to the size of the original image. However, due to blurring and loss of high-frequency information during downsampling, the upsampled third intermediate image will be more blurred than the original image, and the difference (the first edge image) will contain the high-frequency details lost in the original image during the first downsampling.

[0072] The fourth intermediate image is the second intermediate image upsampled twice to the size of the original image. The second intermediate image has been downsampled twice, that is, twice Gaussian blurring and downsampling, losing more high-frequency information. Therefore, the fourth intermediate image will be blurrier. As a result, the difference between the input images (the second edge image) contains more low-frequency structural differences, or coarser edge information.

[0073] It should be noted that high-frequency details correspond to small-scale changes, and low-frequency structures correspond to large-scale changes. Therefore, the first edge image captures higher-frequency (small-scale) changes, while the second edge image captures lower-frequency (large-scale) changes. Therefore, the scale change of the second edge image is larger.

[0074] The first edge image = the input image - the third intermediate image: This is equivalent to the first layer of the Laplacian pyramid, capturing the details lost in the first downsampling, that is, higher-frequency information, corresponding to smaller scale changes.

[0075] The second edge image = the input image - the fourth intermediate image: This is equivalent to the difference after two downsamplings and then two upsamplings, similar to the second layer of the Laplacian pyramid, but corresponding to a coarser scale, that is, a larger structural change, corresponding to a larger scale.

[0076] Therefore, the scale change of the second edge image is larger, belonging to a larger structural difference, while the first edge image belongs to a finer detail difference, that is, a smaller scale change.

[0077] Step S300, fuse the first edge image and the second edge image to obtain the edge information after fusing the first edge image and the second edge image.

[0078] Further, step S300 may include the following steps:

[0079] Step S310, obtain the first fusion weight corresponding to the first edge image and the second fusion weight corresponding to the second edge image; wherein, the first fusion weight is greater than the second fusion weight.

[0080] Step S320, assign the first fusion weight to the first edge image and assign the second fusion weight to the second edge image for weighted fusion.

[0081] In this embodiment, when fusing the first edge image and the second edge image, it is first necessary to determine the respective fusion weights of the first edge image and the second edge image; a smaller fusion weight is set for a larger scale change, and a larger fusion weight is set for a smaller scale change, that is, the first fusion weight is greater than the second fusion weight.

[0082] Further, the sum of the first fusion weight and the second fusion weight is equal to 1.

[0083] For example, the first fusion weight is 0.7 and the second fusion weight is 0.3.

[0084] Furthermore, the first fusion weight and the second fusion weight can also be determined by the following method:

[0085] Calculate the energy layer by layer for the Laplacian pyramids corresponding to the first edge image and the second edge image, that is, the sum of the absolute values of the pixels, to obtain the energy Q1 corresponding to the first edge image and the energy Q2 corresponding to the second edge image.

[0086] According to Q1 and Q2, determine the first fusion weight W1 = Q2 / (Q1 + Q2), and determine the second fusion weight W2 = Q1 / (Q1 + Q2); the greater the energy, the greater the corresponding image scale change, and the smaller the corresponding weight; thus, the first edge image with richer image details has a larger fusion weight, and the second edge image with coarser image details has a smaller fusion weight, improving the accuracy of determining the image sharpness.

[0087] It should be noted that after determining the fusion weights corresponding to the first edge image and the second edge image, the first edge image and the second edge image can be fused through the following formula; the edge information Fused = W1 × the first edge image + W2 × the second edge image.

[0088] Those skilled in the art can use existing image fusion methods according to actual needs to fuse the first edge image and the second edge image, which will not be elaborated here.

[0089] Step S400, determine the pixel intensity volatility corresponding to the input image according to the edge information.

[0090] In this embodiment, the variance of the edge information can be calculated as the pixel intensity volatility corresponding to the input image.

[0091] Pixel intensity volatility = (1 / M) ∑ M i＝1 (x i - μ) 2 ; where x i is the pixel value of the fused image, μ is the pixel mean, and M is the total number of pixels.

[0092] Step S500, if the pixel intensity volatility corresponding to the input image is greater than the first preset pixel intensity volatility threshold, determine the sharpness of the image to be recognized as the first sharpness.

[0093] In this embodiment, when the pixel intensity volatility of the input image is greater than the first preset pixel intensity volatility threshold, the sharpness of the image to be recognized is determined to be the first sharpness.

[0094] Further, the method further includes the following steps:

[0095] Step S600, when the pixel intensity volatility of the input image is greater than or equal to the second preset pixel intensity volatility threshold and less than or equal to the first preset pixel intensity volatility threshold, determine that the sharpness of the image to be recognized is the second sharpness.

[0096] In this embodiment, when the pixel intensity volatility of the input image is greater than or equal to the second preset pixel intensity volatility threshold and less than or equal to the first preset pixel intensity volatility threshold, it indicates that the fused image has certain details but is not sharp enough, belonging to medium sharpness.

[0097] Step S700, when the pixel intensity volatility of the input image is less than the second preset pixel intensity volatility threshold, determine that the sharpness of the image to be recognized is the third sharpness.

[0098] In this embodiment, when the pixel intensity volatility of the input image is less than the second preset pixel intensity volatility threshold, it indicates that the fused image is blurred and has little edge information, belonging to low sharpness. Step S600 and step S700 are in an "and / or" relationship.

[0099] The above first preset pixel intensity volatility threshold and second preset pixel intensity volatility threshold can be obtained by analyzing and statistically processing a large number of fused images with different sharpness levels.

[0100] For the specific implementation process of the method in the above embodiment, reference can be made to Figure 2 as shown.

[0101] In this embodiment, by constructing a corresponding Laplacian pyramid for the input image, a first edge image and a second edge image are obtained, and the first edge image and the second edge image are fused to obtain edge information. Based on the edge information, the pixel intensity volatility of the input image is determined, and further the sharpness of the image to be recognized is determined. Through multi-scale edge fusion, the present invention realizes high-precision, high-efficiency, and strong-robustness image sharpness determination, and solves problems such as being vulnerable to noise interference, incomplete detail capture, and single evaluation index in traditional methods in complex scenarios.

[0102] The above method can be applied in the field of intelligent vehicles. For example, it can be applied to the vehicle vision system of intelligent vehicles. Through the above method, first, it is determined whether the sharpness of the captured image meets the requirements. If the sharpness meets the requirements, then the image is processed.

[0103] In addition, although the various steps of the methods in the present disclosure are described in a specific order in the drawings, this does not require or imply that these steps must be performed in that specific order, or that all of the steps shown must be performed to achieve the desired result. Additionally or alternatively, certain steps may be omitted, multiple steps may be combined into one step for execution, and / or one step may be decomposed into multiple steps for execution, etc.

[0104] In an exemplary embodiment, there is provided an apparatus for recognizing image sharpness as Figure 3 shown, characterized in that the apparatus includes:

[0105] An input image acquisition module, configured to acquire an input image corresponding to the image to be recognized.

[0106] A Laplacian pyramid construction module, configured to construct a Laplacian pyramid corresponding to the input image to obtain a first edge image and a second edge image corresponding to the input image.

[0107] An edge information fusion module, configured to fuse the first edge image and the second edge image to obtain edge information after fusing the first edge image and the second edge image.

[0108] A pixel intensity volatility determination module, configured to determine the pixel intensity volatility corresponding to the input image according to the edge information.

[0109] A first sharpness determination module, configured to determine that the sharpness of the image to be recognized is the first sharpness when the pixel intensity volatility is greater than a first preset pixel intensity volatility threshold.

[0110] Further, the Laplacian pyramid construction module includes:

[0111] A first intermediate image processing unit, configured to perform one downsampling on the input image to obtain a first intermediate image.

[0112] A second intermediate image processing unit, configured to perform one downsampling on the first intermediate image to obtain a second intermediate image.

[0113] A third intermediate image processing unit, configured to perform one upsampling on the first intermediate image to obtain a third intermediate image.

[0114] A fourth intermediate image processing unit, configured to perform two upsamplings on the second intermediate image to obtain a fourth intermediate image.

[0115] A first edge image generation unit, configured to calculate the difference between the input image and the third intermediate image to obtain the first edge image.

[0116] A second edge image generation unit for calculating the difference between the input image and the fourth intermediate image to obtain the second edge image.

[0117] Further, the edge information fusion module includes:

[0118] A fusion weight acquisition unit for acquiring a first fusion weight corresponding to the first edge image and a second fusion weight corresponding to the second edge image; wherein, the first fusion weight is greater than the second fusion weight.

[0119] A weighted fusion unit for performing weighted fusion by assigning the first fusion weight to the first edge image and the second fusion weight to the second edge image.

[0120] Further, the apparatus further includes:

[0121] A second sharpness determination module for determining that the sharpness of the image to be recognized is the second sharpness if the pixel intensity volatility corresponding to the input image is greater than or equal to a second preset pixel intensity volatility threshold and less than or equal to a first preset pixel intensity volatility threshold.

[0122] A third sharpness determination module for determining that the sharpness of the image to be recognized is the third sharpness if the pixel intensity volatility corresponding to the input image is less than the second preset pixel intensity volatility threshold.

[0123] An embodiment of the present invention further provides a non-transitory computer-readable storage medium, which can be disposed in an electronic device to store at least one instruction or at least one segment of a program related to a method in a method embodiment. The at least one instruction or the at least one segment of the program is loaded and executed by the processor to implement the method provided in the above embodiment.

[0124] The program product can adopt any combination of one or more readable media. The readable media can be a readable signal medium or a readable storage medium. The readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples (non-exhaustive list) of the readable storage medium include: an electrical connection having one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0125] A computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries readable program code. Such propagated data signal may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the foregoing. The readable signal medium may also be any readable medium other than a readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device.

[0126] The program code contained on the readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wired, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0127] The program code for performing the operations of this application may be written in any combination of one or more programming languages, including object-oriented programming languages such as Java, C++, etc., and also including conventional procedural programming languages such as the "C" language or similar programming languages. The program code may be executed entirely on the user's computing device, partially on the user's device, executed as a stand-alone 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. As Figure 4 shown, in the case of a remote computing device, the remote computing device may be connected to the user's computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., by connecting through the Internet using an Internet service provider).

[0128] Embodiments of the present invention also provide an electronic device as Figure 5 shown, including a processor and the aforementioned non-transitory computer-readable storage medium.

[0129] The electronic device is merely an example and should not impose any limitation on the functions and scope of use of the embodiments of this application.

[0130] The electronic device is presented in the form of a general-purpose computing device. The components of the electronic device may include but are not limited to: at least one of the aforementioned processors, at least one of the aforementioned memories, and a bus connecting different system components (including the memory and the processor).

[0131] Among them, the memory stores program code, and the program code can be executed by the processor, so that the processor executes the steps in various embodiments described in this specification.

[0132] The memory may include a readable medium in the form of volatile memory, such as random access memory (RAM) and / or cache memory, and may further include read-only memory (ROM).

[0133] The memory may also include a program / utility having a set (at least one) of program modules, such program modules including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of these examples or some combination thereof may include an implementation of a network environment.

[0134] The bus may represent one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, a graphics acceleration port, a processor, or a local bus using any of a variety of bus structures.

[0135] The electronic device may also communicate with one or more external devices (such as a keyboard, a pointing device, a Bluetooth device, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device, and / or communicate with any device that enables the electronic device to communicate with one or more other computing devices (such as a router, a modem, etc.). Such communication may be through an input / output (I / O) interface. Also, the electronic device may communicate with one or more networks (such as a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) through a network adapter. The network adapter communicates with other modules of the electronic device through the bus. It should be understood that, although not shown in the figures, other hardware and / or software modules may be used in conjunction with the electronic device, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, etc.

[0136] Through the description of the above embodiments, those skilled in the art can easily understand that the exemplary embodiments described herein can be implemented by software, or can be implemented by a combination of software and necessary hardware. Therefore, the technical solutions according to the embodiments of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, including several instructions to cause a computing device (which can be a personal computer, a server, a terminal device, or a network device, etc.) to execute the method according to the embodiments of the present disclosure.

[0137] An embodiment of the present invention also provides a computer program product, which includes program code, and when the program product runs on an electronic device, the program code is used to cause the electronic device to execute the steps in the methods according to various exemplary embodiments of the present invention described above in this specification.

[0138] Although some specific embodiments of the present invention have been described in detail by way of examples, those skilled in the art should understand that the above examples are for illustrative purposes only and not for limiting the scope of the present invention. Those skilled in the art should also understand that various modifications can be made to the embodiments without departing from the scope and spirit of the present invention.

Claims

1. A method for identifying image clarity, characterized in that: The method comprises: Obtain an input image corresponding to the image to be recognized; Constructing a Laplacian pyramid corresponding to the input image to obtain a first edge image and a second edge image corresponding to the input image; fusing the first edge image and the second edge image to obtain edge information after the first edge image and the second edge image are fused; Determining a pixel intensity fluctuation rate corresponding to the input image according to the edge information; When the pixel intensity fluctuation rate is greater than a first preset pixel intensity fluctuation rate threshold, the clarity of the image to be identified is determined to be a first clarity.

2. The method for identifying image clarity according to claim 1, characterized in that: Constructing a Laplacian pyramid corresponding to the input image to obtain a first edge image and a second edge image corresponding to the input image, including: Downsampling the input image once to obtain a first intermediate image; Downsampling the first intermediate image once to obtain a second intermediate image; Upsampling the first intermediate image once to obtain a third intermediate image; Upsampling the second intermediate image twice to obtain a fourth intermediate image; Calculating a difference between the input image and the third intermediate image to obtain the first edge image; A difference between the input image and the fourth intermediate image is calculated to obtain the second edge image.

3. The method for identifying image clarity according to claim 1, characterized in that: The step of fusing the first edge image and the second edge image includes: Obtaining a first fusion weight corresponding to the first edge image and a second fusion weight corresponding to the second edge image; wherein the first fusion weight is greater than the second fusion weight; A first fusion weight is assigned to the first edge image and a second fusion weight is assigned to the second edge image for weighted fusion.

4. The method for identifying image clarity according to claim 1, characterized in that: The method further comprises: When the pixel intensity fluctuation rate corresponding to the input image is greater than or equal to the second preset pixel intensity fluctuation rate threshold and less than or equal to the first preset pixel intensity fluctuation rate threshold, determining the definition of the image to be identified to be the second definition; and / or, When the pixel intensity fluctuation rate corresponding to the input image is less than the second preset pixel intensity fluctuation rate threshold, the definition of the image to be identified is determined to be the third definition.

5. The method for recognizing image clarity according to claim 4, characterized in that: The third definition is smaller than the second definition, and the second definition is smaller than the first definition.

6. The method for identifying image clarity according to claim 1, characterized in that: Acquiring an input image corresponding to the image to be identified includes: performing grayscale processing and image denoising processing based on wavelet transform on the image to be identified in sequence to obtain the input image.

7. The method for recognizing image clarity according to claim 3, characterized in that: The sum of the first fusion weight and the second fusion weight is equal to 1.

8. An image definition recognition device, characterized in that: The device comprises: An input image acquisition module is used to acquire an input image corresponding to the image to be recognized; A Laplacian pyramid construction module, used to construct a Laplacian pyramid corresponding to the input image, and obtain a first edge image and a second edge image corresponding to the input image; An edge information fusion module, used for fusing the first edge image and the second edge image to obtain edge information after the first edge image and the second edge image are fused; A pixel intensity fluctuation rate determination module, used to determine the pixel intensity fluctuation rate corresponding to the input image according to the edge information; The first clarity judgment module is used to determine that the clarity of the image to be identified is a first clarity when the pixel intensity fluctuation rate is greater than a first preset pixel intensity fluctuation rate threshold.

9. A non-transitory computer-readable storage medium, wherein at least one instruction or at least one program is stored in the storage medium, characterized in that: The at least one instruction or the at least one program is loaded and executed by the processor to implement the image clarity recognition method as described in any one of claims 1-7.

10. An electronic device, characterized in that: The method comprises a processor and the non-transitory computer-readable storage medium of claim 9.

Images