No-reference image quality evaluation method, device and electronic equipment

Abstract

Embodiments of the present application provide a no-reference image quality evaluation method and device and electronic equipment, comprising: obtaining a target image to be evaluated; selecting an image block satisfying a preset condition from the target image as a first image block; constructing a second image block corresponding to the first image block; and obtaining a target quality score of the target image according to the first image block and the second image block. The method can quickly and accurately obtain the target quality score of the target image without the original image of the target image that has not been distorted.

Description

Technical Field

[0001] This disclosure relates to the field of image processing technology, and more specifically, to a method, apparatus, and electronic device for evaluating the quality of a non-reference image. Background Technology

[0002] With the continuous development of image processing technology, precious film and video resources can be restored and reconstructed by using image reconstruction technology, allowing these video resources to be given a new look and return to the public's entertainment life.

[0003] For example, high-fidelity noise reduction, sharpness enhancement, resolution enhancement, and color enhancement image reconstruction models based on machine learning technology can be used to reconstruct images of precious film and video resources so that they can meet the viewing requirements of users.

[0004] However, since there are no undistorted original images as references for film video resources, how to evaluate the quality of the reconstructed images after reconstruction of such video resources, and how to optimize the image reconstruction model based on the quality score obtained from the evaluation, has become an urgent problem to be solved. Summary of the Invention

[0005] One objective of this disclosure is to provide a new technical solution for non-reference image quality evaluation to address the aforementioned problems.

[0006] According to a first aspect of this disclosure, an embodiment of a referenceless image quality assessment method is provided, comprising:

[0007] Obtain the target image to be evaluated;

[0008] From the target image, select an image block that meets the preset conditions as the first image block;

[0009] Construct a second image block corresponding to the first image block;

[0010] The target quality score of the target image is obtained based on the first image block and the second image block.

[0011] Optionally, selecting an image patch that meets preset conditions from the target image as the first image patch includes:

[0012] A sliding window is set up, wherein the sliding window is used to select the first image block in the target image, and the side length of the sliding window is not less than a first preset threshold.

[0013] Using the sliding window, an image block that meets the preset conditions is selected from the target image as the first image block.

[0014] Optionally, using the sliding window to select an image block in the target image that meets the preset conditions as the first image block includes:

[0015] In the target image, the sliding window is slid according to a preset step size to obtain at least one candidate image block, wherein the candidate image block matches the size of the sliding window;

[0016] From the at least one candidate image block, select the image block whose corresponding image features satisfy the preset condition as the first image block.

[0017] Optionally, constructing the second image block corresponding to the first image block includes:

[0018] Obtain the first pixel value of all pixels in the first image block;

[0019] According to a preset mapping function, obtain the second pixel value corresponding to the first pixel value;

[0020] The second image block is constructed by using the second pixel value as the fill value for the pixels in the second image block.

[0021] Optionally, obtaining the second pixel value corresponding to the first pixel value according to a preset mapping function includes:

[0022] The average value of the first pixel is calculated as the second pixel value.

[0023] Optionally, the number of the first image blocks is at least one, and the second image block corresponds to each of the at least one first image block;

[0024] The step of obtaining the target quality score of the target image based on the first image block and the second image block includes:

[0025] Using the second image block as a reference image block for the corresponding first image block, the quality score of the at least one first image block is obtained as the quality score to be determined;

[0026] The target quality score of the target image is obtained based on the quality score to be determined.

[0027] Optionally, obtaining the target quality score of the target image based on the quality score to be determined includes:

[0028] The average value of the quality scores to be determined is obtained as the target quality score.

[0029] Optionally, the target image is an image obtained by performing image reconstruction processing on the original image using an image reconstruction model. After obtaining the target quality score, the method further includes:

[0030] If the target quality score is not greater than the second preset threshold, the reconstruction model is optimized by adjusting the parameters of the reconstruction model.

[0031] According to a second aspect of this disclosure, an embodiment of a referenceless image quality assessment apparatus is provided, comprising:

[0032] The image acquisition module is used to acquire the target image to be evaluated;

[0033] The first image block selection module is used to select an image block that meets preset conditions from the target image as the first image block;

[0034] The second image block construction module is used to generate a second image block based on the first image block;

[0035] The target quality score acquisition module is used to obtain the target quality score of the target image based on the first image block and the second image block.

[0036] According to a third aspect of this disclosure, an embodiment of an electronic device is provided, including the means as described in the second aspect of this specification; or,

[0037] The electronic device includes:

[0038] Memory is used to store executable instructions;

[0039] A processor, configured to perform the referenceless image quality assessment method as described in the first aspect of this specification, under the control of the executable computer program.

[0040] One beneficial effect of the embodiments of this disclosure is that, according to the embodiments of this disclosure, for a target image to be evaluated, by selecting an image block that meets preset conditions from the target image as a first image block, and by constructing a second image block corresponding to the first image block, the target quality score of the target image can be obtained based on the first image block and the second image block. The method provided by the embodiments of this disclosure does not require obtaining the original, undistorted image of the target image to be evaluated as a reference. Instead, by selecting a first image block from the target image and automatically constructing a second image block as a reference for the first image block, the target quality score of the target image can be accurately obtained based on the first image block and the second image block. Thus, the quality of the target image can be judged based on the target quality score to determine whether continuous optimization of the image reconstruction model corresponding to the target image is necessary.

[0041] Other features and advantages of this specification will become clear from the following detailed description of exemplary embodiments with reference to the accompanying drawings. Attached Figure Description

[0042] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments of this specification and, together with their description, serve to explain the principles of this specification.

[0043] Figure 1 This is a schematic diagram of a scenario for the no-reference image quality evaluation method provided in this embodiment.

[0044] Figure 2 This is a schematic diagram of the structure of an electronic device that can be used to implement the embodiments of this disclosure.

[0045] Figure 3 This is a flowchart illustrating a no-reference image quality assessment method according to one embodiment.

[0046] Figure 4 This is a block diagram of a no-reference image quality evaluation device according to one embodiment.

[0047] Figure 5 This is a schematic diagram of the hardware structure of an electronic device according to one embodiment. Detailed Implementation

[0048] Various exemplary embodiments of the present disclosure will now be described in detail with reference to the accompanying drawings. It should be noted that, unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps set forth in these embodiments do not limit the scope of the invention.

[0049] The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit the invention or its application or use.

[0050] Techniques, methods, and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and equipment should be considered part of the specification.

[0051] In all the examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values.

[0052] It should be understood that although the terms first, second, etc., may be used to describe various information in one or more embodiments of this specification, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, first may also be referred to as second without departing from the scope of one or more embodiments of this specification, and similarly, second may also be referred to as first.

[0053] It should be noted that similar labels and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be discussed further in subsequent figures.

[0054] This disclosure relates to application scenarios for quality evaluation of image and / or video resources.

[0055] Image quality assessment (IQA) can generally be divided into subjective image quality assessment and objective image quality assessment, depending on whether human involvement is involved. Subjective image quality assessment refers to an evaluation method where humans, as observers, subjectively evaluate images, striving to accurately reflect human visual perception. Objective image quality assessment refers to an evaluation method based on manually designed algorithmic models that reflect human subjective perception, providing numerical calculation results for the image being evaluated.

[0056] Subjective image quality assessment relies on manual labor, which results in tedious tasks, time consumption, and non-reproducible evaluation results, making it unsuitable for large-scale application.

[0057] Objective image quality assessment is generally divided into full-reference (FR) image quality assessment, reduced-reference (RR) image quality assessment, and no-reference (NR) image quality assessment. Since no-reference image quality assessment does not require any information from a reference image, it is currently the most widely researched and used image quality assessment method.

[0058] Specifically, in this application, when restoring and reconstructing precious film and video resources—which lack undistorted original images—and then needing to evaluate the quality of the reconstructed target image to continuously fine-tune the image reconstruction model based on the quality score, the commonly used quality evaluation methods with reference images are inapplicable because these video resources lack undistorted original images for reference. This results in inaccurate quality evaluation of the target image. Furthermore, traditional no-reference quality evaluation methods suffer from high hardware requirements and processing complexity due to their complex model structure design.

[0059] To enable rapid and accurate quality assessment of video resources such as precious film reels, embodiments of this disclosure provide a reference-free image quality assessment method. Please refer to... Figure 1 This is a schematic diagram of a scenario for the referenceless image quality evaluation method provided in this embodiment. The method provided in this embodiment can be applied to electronic devices, which can be servers or terminal devices; no special limitations are imposed here.

[0060] Taking the target image to be evaluated as an example, which is the image obtained after reconstructing precious film video resources using an image reconstruction model. For instance... Figure 1 As shown, after reconstructing the precious film video resources using the image reconstruction model to obtain the target image to be evaluated, the target image can be uploaded to the electronic device 1000 implementing the method. After obtaining the target image, the electronic device 1000 selects an image block that meets the preset conditions from the target image as the first image block, and constructs a second image block corresponding to the first image block. Based on the first image block and the second image block, the target quality score of the target image can be obtained quickly and accurately. After obtaining the target quality score, the user can make an accurate assessment of the quality of the reconstructed target image based on the target quality score. For example, if the target quality score is not greater than a second preset threshold, the image reconstruction model can be continuously optimized to reconstruct the target image that meets the requirements.

[0061] <Hardware Configuration>

[0062] Figure 2 This is a schematic diagram of the structure of an electronic device that can be used to implement the embodiments of this disclosure.

[0063] The electronic device 1000 can be a server, smartphone, laptop, desktop computer, tablet computer, etc., and is not limited thereto.

[0064] The electronic device 1000 may include, but is not limited to, a processor 1100, a memory 1200, an interface device 1300, a communication device 1400, a display device 1500, an input device 1600, a speaker 1700, a microphone 1800, etc. The processor 1100 may be a central processing unit (CPU), a graphics processing unit (GPU), a microprocessor (MCU), etc., used to execute computer programs, which can be written using instruction sets of architectures such as x86, Arm, RISC, MIPS, SSE, etc. The memory 1200 may include, for example, ROM (Read-Only Memory), RAM (Random Access Memory), and non-volatile memory such as a hard disk. The interface device 1300 may include, for example, a USB interface, a serial interface, a parallel interface, etc. The communication device 1400 may be capable of wired communication using fiber optic cables or cables, or wireless communication, specifically including WiFi communication, Bluetooth communication, 2G / 3G / 4G / 5G communication, etc. The display device 1500 may be, for example, an LCD screen, a touch screen, etc. The input device 1600 may include, for example, a touch screen, a keyboard, motion input, etc. Speaker 1700 is used to output audio signals. Microphone 1800 is used to capture audio signals.

[0065] In the embodiments of this disclosure, the memory 1200 of the electronic device 1000 is used to store a computer program that controls the processor 1100 to operate in order to implement the method according to the embodiments of this disclosure. Those skilled in the art can design this computer program based on the scheme disclosed herein. How the computer program controls the processor to operate is well known in the art and will not be described in detail here. The electronic device 1000 may be equipped with a smart operating system (e.g., Windows, Linux, Android, iOS, etc.) and application software.

[0066] Those skilled in the art should understand that, although in Figure 2 The present invention illustrates multiple devices of an electronic device 1000; however, the electronic device 1000 of the present invention may refer to only some of the devices, for example, only the processor 1100 and the memory 1200.

[0067] Hereinafter, various embodiments and examples according to the present invention will be described with reference to the accompanying drawings.

[0068] <Method Implementation>

[0069] Figure 3 This is a flowchart illustrating a no-reference image quality assessment method according to one embodiment. This embodiment can be implemented by an electronic device, for example, by... Figure 2 The electronic device 1000 shown is implemented.

[0070] like Figure 3As shown, the no-reference image quality evaluation method of this embodiment may include the following steps S33100-S3400, which will be described in detail below.

[0071] Step S3100: Obtain the target image to be evaluated.

[0072] The target image is the image to be evaluated to obtain a quality score that characterizes its quality. The target image can be a direct image format, or it can be a video frame from a video resource. More specifically, the target image can be a video frame from a reconstructed video resource obtained after image reconstruction processing of precious film video resources using an image reconstruction model.

[0073] In practice, after the target image is obtained by image reconstruction processing of precious film and video resources, the user can upload the target image to the electronic device implementing the method to obtain the target quality score of the target image; or, the electronic device performing the image reconstruction processing can directly send the target image to the electronic device implementing the method according to the user's preset processing flow after the reconstruction is obtained, so that the electronic device can implement the method to obtain its target quality score.

[0074] The preset processing flow can be as follows: after the target image is reconstructed, a task message to be processed containing the target image is generated; the task message to be processed is added to a task queue; then, an electronic device can retrieve a task message to be processed from the task queue at preset time intervals to perform a no-reference quality evaluation of the target image to be evaluated.

[0075] Step S3200: Select an image block that meets the preset conditions from the target image as the first image block.

[0076] The first image block can be an image block adaptively selected from the target image by the electronic device based on the resolution of the target image, whose corresponding image features meet preset conditions. In this embodiment, the image features can be the texture features of the image block, and the preset condition can be whether the first image block is a flat region.

[0077] Specifically, after obtaining the target image to be evaluated through step S3100 above, in order to quickly and accurately obtain its target quality score, in this embodiment, instead of evaluating the target image as a whole, at least one first image patch that meets preset conditions can be selected from the target image based on its resolution. At least one second image patch corresponding to each of these at least one first image patch is then constructed as a reference. This allows for referenceless image quality evaluation of the target image without the need for a distorted original image as a reference, thereby obtaining its target quality score. The following section will first explain in detail how to select the first image patch from the target image.

[0078] In one embodiment, selecting an image block that meets a preset condition from the target image as a first image block includes: setting a sliding window, wherein the sliding window is used to select the first image block in the target image, and the side length of the sliding window is not less than a first preset threshold; and using the sliding window to select an image block that meets the preset condition from the target image as the first image block.

[0079] Specifically, the first image block can be an image block selected from the target image after traversing the target image, which is square in shape, does not overlap with each other, and has a side length not less than a first preset threshold.

[0080] In this embodiment, the first preset threshold can be determined based on the resolution of the target image, specifically 5% of the number of pixels on the shorter side of the target image. For example, if the resolution of the target image is 1920*1080, the first preset threshold can be 54.

[0081] In addition, in order to achieve the traversal operation of the target image in a specific implementation, a sliding window can be set in this embodiment. The sliding window is square in shape and the side length of the window is not less than the first preset threshold mentioned above. By sliding the sliding window in the target image and using the area marked by the sliding window as a candidate image block, the first image block that meets the preset conditions can be selected.

[0082] In a specific implementation, the step of using the sliding window to select an image block in the target image that meets the preset conditions as the first image block includes: sliding the sliding window in the target image according to a preset step size value to obtain at least one candidate image block, wherein the candidate image block matches the size of the sliding window; and selecting an image block whose corresponding image features meet the preset conditions from the at least one candidate image block as the first image block.

[0083] Specifically, a step size of 1 can be set, starting from any boundary of the target image. By gradually sliding the sliding window, image blocks that match the size of the sliding window are selected as candidate image blocks, so as to select at least one candidate image block from the target image. Then, by extracting the image features of each candidate image block, the image block that meets the preset conditions can be selected as the first image block based on the image features.

[0084] In one embodiment, selecting an image block whose corresponding image features satisfy the preset condition from the at least one candidate image block as the first image block includes: using a preset corner detection algorithm to detect whether the at least one candidate image block is a flat region, and selecting a candidate image block that is a flat region as the first image block.

[0085] The preset corner detection method can be the Moravec corner detection algorithm, the Harris corner detection algorithm, or the Shi-Tomasi corner detection algorithm, etc. The detailed processing process will not be described here.

[0086] It should be noted that, in specific implementation, other methods can also be used to select image patches from the target image. For example, the target image can be input into a pre-trained image patch cutting model, so that a second image patch that meets the preset conditions can be output according to the image patch cutting model.

[0087] After step S3100, step S3300 is executed to construct a second image block corresponding to the first image block.

[0088] After the above steps are performed to select a first image block that meets the preset conditions from the target image, a second image block corresponding to the first image block can be constructed as a reference. By evaluating the quality of the local part of the target image, a target quality score reflecting its overall quality can be obtained.

[0089] In one embodiment, constructing a second image block corresponding to the first image block includes: obtaining the first pixel value of all pixels in the first image block; obtaining the second pixel value corresponding to the first pixel value according to a preset mapping function; and constructing the second image block by using the second pixel value as the fill value of the pixels in the second image block.

[0090] The preset mapping function is a function that reflects the relationship between the pixel values ​​of pixels in the first image block and the pixel values ​​of corresponding pixels in the second image block.

[0091] In this embodiment, the preset mapping function is used to construct a second image block whose image quality meets preset requirements based on the pixel values ​​of the pixels in the first image block. That is, an image block with a smooth and noise-free texture region is constructed as a reference for the corresponding first image block to evaluate the image quality of the first image block.

[0092] Specifically, after selecting the first image block from the target image, the pixel value of the corresponding pixel in the second image block can be obtained through the preset mapping function based on the pixel values ​​of all pixels in the first image block.

[0093] In one embodiment, obtaining the second pixel value corresponding to the first pixel value according to a preset mapping function includes: calculating the average value of the first pixel value as the second pixel value.

[0094] In this embodiment, in order to reduce computational complexity, the preset mapping function can be directly the average value of all pixels in the first image block.

[0095] Taking a 10*10 resolution image block as an example, the fill value of all pixels in the second image block can be the average of the pixel values ​​of the 100 pixels contained in the first image block.

[0096] Step S3400: Obtain the target quality score of the target image based on the first image block and the second image block.

[0097] After selecting at least one first image block from the target image through the above steps and obtaining a second image block corresponding to the at least one first image block as a reference, the target quality score of the target image can be calculated based on the first image block and the second image block.

[0098] In one embodiment, the number of the first image blocks is at least one, and the second image blocks correspond to the at least one first image block respectively; obtaining the target quality score of the target image based on the first image blocks and the second image blocks includes: using the second image block as a reference image block for the corresponding first image block, obtaining the quality score of the at least one first image block as a quality score to be determined; and obtaining the target quality score of the target image based on the quality score to be determined.

[0099] Specifically, after completing the above steps S3300 and constructing the second image blocks corresponding to the first image blocks, the existing full-reference image quality evaluation method can be used to calculate the quality score of each first image block as a score to be determined reflecting the quality of the target image; then, the target quality score of the target image can be calculated based on the obtained score to be determined.

[0100] In a specific implementation, the quality score of the first image block can be obtained by using the corresponding second image block as its reference image block and by using a full reference image quality evaluation method; wherein, the full reference image quality evaluation method can be, for example, Peak Signal-to-Noise Ratio (PSNR).

[0101] In one embodiment, obtaining the target quality score of the target image based on the quality score to be determined includes: obtaining the average value of the quality scores to be determined as the target quality score.

[0102] Specifically, in this embodiment, in order to improve the data processing speed, after obtaining the quality score of the first image block as the quality score to be determined of the target image, the average value of the at least one quality score to be determined can be used as the target quality score of the target image.

[0103] It should be noted that, in specific implementation, other methods can also be used to obtain the target quality score. For example, after obtaining the quality score of each first image block as the score to be determined, the weight of the score can be calculated based on the position of the corresponding first image block in the target image. Thus, a more accurate target quality score can be obtained by weighted averaging of each score to be determined.

[0104] As can be seen from the above description, when the target image is an image obtained by reconstructing the original image using an image reconstruction model, after obtaining the target quality score of the target image through the above steps, the method may further include: when the target quality score is not greater than a second preset threshold, optimizing the reconstruction model by adjusting the parameters of the reconstruction model.

[0105] In summary, the method provided by the embodiments of this disclosure, for a target image to be evaluated, selects an image patch that meets preset conditions as a first image patch from the target image, and constructs a second image patch corresponding to the first image patch. Based on the first and second image patches, a target quality score of the target image can be obtained. The method provided by the embodiments of this disclosure does not require obtaining an undistorted original image of the target image to be evaluated as a reference. Instead, by selecting a first image patch from the target image and automatically constructing a second image patch as a reference for the first image patch, the target quality score of the target image can be accurately obtained based on the first and second image patches. This target quality score can then be used to evaluate the quality of the target image and determine whether continuous optimization of the image reconstruction model corresponding to the target image is necessary.

[0106] <Device Embodiment>

[0107] Figure 4 This is a block diagram of a no-reference image quality evaluation device according to one embodiment. Figure 4 As shown, the referenceless image quality device 4000 may include an image acquisition module 4100, a first image block selection module 4200, a second image block construction module 4300, and a target quality score acquisition module 4400.

[0108] The image acquisition module 4100 is used to acquire the target image to be evaluated.

[0109] The first image block selection module 4200 is used to select an image block that meets preset conditions from the target image as the first image block.

[0110] In one embodiment, when the first image block selection module 4200 selects an image block that meets a preset condition from the target image as the first image block, it can be used to: set a sliding window, wherein the sliding window is used to select the first image block in the target image, and the side length of the sliding window is not less than a first preset threshold; and use the sliding window to select an image block that meets the preset condition in the target image as the first image block.

[0111] In one embodiment, when the first image block selection module 4200 uses the sliding window to select an image block in the target image that meets the preset conditions as the first image block, it can be used to: slide the sliding window in the target image according to a preset step size value to obtain at least one candidate image block, wherein the candidate image block matches the size of the sliding window; and select an image block whose corresponding image features meet the preset conditions from the at least one candidate image block as the first image block.

[0112] The second image block construction module 4300 is used to generate a second image block based on the first image block.

[0113] In one embodiment, when constructing a second image block corresponding to the first image block, the second image block construction module 4300 can be used to: obtain the first pixel value of all pixels in the first image block; obtain the second pixel value corresponding to the first pixel value according to a preset mapping function; and construct the second image block by using the second pixel value as the fill value of the pixels in the second image block.

[0114] In one embodiment, when the second image block construction module 4300 obtains the second pixel value corresponding to the first pixel value according to the preset mapping function, it can be used to: calculate the average value of the first pixel value as the second pixel value.

[0115] The target quality score obtaining module 4400 is used to obtain the target quality score of the target image based on the first image block and the second image block.

[0116] In one embodiment, the number of the first image blocks is at least one, and the second image blocks correspond to the at least one first image block respectively; when the target quality score obtaining module 4400 obtains the target quality score of the target image based on the first image blocks and the second image blocks, it can be used to: use the second image block as a reference image block for the corresponding first image block, obtain the quality score of the at least one first image block as a quality score to be determined; and obtain the target quality score of the target image based on the quality score to be determined.

[0117] In one embodiment, when the target quality score obtaining module 4400 obtains the target quality score of the target image based on the quality score to be determined, it can be used to: obtain the average value of the quality score to be determined as the target quality score.

[0118] <Equipment Example>

[0119] Figure 5 A schematic diagram of the hardware structure of an electronic device according to one embodiment.

[0120] like Figure 5 As shown, the electronic device 500 includes a processor 520 and a memory 510, the memory 510 for storing an executable computer program, and the processor 520 for executing methods as described in any of the above method embodiments under the control of the computer program.

[0121] One or more embodiments of this specification may be a system, method, and / or computer program product. A computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for causing a processor to implement various aspects of this specification.

[0122] Computer-readable storage media can be tangible devices capable of holding and storing instructions for use by an instruction execution device. Computer-readable storage media can be, for example—but not limited to—electrical storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any suitable combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable compact disc read-only memory (CD-ROM), digital multifunction disc (DVD), memory sticks, floppy disks, mechanical encoding devices, such as punch cards or recessed protrusions storing instructions thereon, and any suitable combination thereof. The computer-readable storage media used herein are not to be construed as transient signals themselves, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., light pulses through fiber optic cables), or electrical signals transmitted through wires.

[0123] The computer-readable program instructions described herein can be downloaded from computer-readable storage media to various computing / processing devices, or downloaded via a network, such as the Internet, local area network, wide area network, and / or wireless network, to an external computer or external storage device. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards them to the computer-readable storage media in the respective computing / processing device.

[0124] Computer program instructions used to perform the operations of the embodiments described herein may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, etc., and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The computer-readable program instructions may execute entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, electronic circuitry, such as programmable logic circuitry, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs), is personalized by utilizing state information from the computer-readable program instructions. This electronic circuitry can execute the computer-readable program instructions to implement various aspects of this specification.

[0125] Various aspects of this specification are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this specification. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer-readable program instructions.

[0126] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that, when executed by the processor of the computer or other programmable data processing apparatus, they create means for implementing the functions / actions specified in one or more blocks of the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium that causes a computer, programmable data processing apparatus, and / or other device to operate in a particular manner; thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing aspects of the functions / actions specified in one or more blocks of the flowchart and / or block diagram.

[0127] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to perform the functions / actions specified in one or more boxes of a flowchart and / or block diagram.

[0128] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this specification. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of an instruction, which contains one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions. It is well known to those skilled in the art that implementation in hardware, implementation in software, and implementation using a combination of software and hardware are equivalent.

[0129] The various embodiments of this specification have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles, practical application, or technical improvements to the embodiments in the market, or to enable others skilled in the art to understand the embodiments disclosed herein. The scope of this application is defined by the appended claims.

Claims

1. A method for evaluating the quality of a referenceless image, characterized in that, include: Obtain the target image to be evaluated; From the target image, an image block that meets a preset condition is selected as the first image block; wherein, the first preset condition is: the first image block is a flat area; Construct a second image block corresponding to the first image block; The target quality score of the target image is obtained based on the first image block and the second image block; The construction of the second image block corresponding to the first image block includes: obtaining the first pixel value of all pixels in the first image block; calculating the average value of the first pixel value as the second pixel value; and using the second pixel value as the fill value of the pixels in the second image block to construct the second image block.

2. The method according to claim 1, characterized in that, Selecting an image block that meets preset conditions from the target image as the first image block includes: A sliding window is set up, wherein the sliding window is used to select the first image block in the target image, and the side length of the sliding window is not less than a first preset threshold. Using the sliding window, an image block that meets the preset conditions is selected from the target image as the first image block.

3. The method according to claim 2, characterized in that, The step of using the sliding window to select an image block in the target image that meets the preset conditions as the first image block includes: In the target image, the sliding window is slid according to a preset step size to obtain at least one candidate image block, wherein the candidate image block matches the size of the sliding window; From the at least one candidate image block, select the image block whose corresponding image features satisfy the preset condition as the first image block; Selecting an image block whose corresponding image features satisfy the preset condition from the at least one candidate image block includes: The preset corner detection algorithm is used to detect whether the at least one candidate image block is a flat region, and the candidate image block that is a flat region is selected as the first image block.

4. The method according to claim 1, characterized in that, The number of the first image blocks is at least one, and the second image blocks correspond to the at least one first image block respectively; The step of obtaining the target quality score of the target image based on the first image block and the second image block includes: Using the second image block as a reference image block for the corresponding first image block, the quality score of the at least one first image block is obtained as the quality score to be determined; The target quality score of the target image is obtained based on the quality score to be determined.

5. The method according to claim 4, characterized in that, The step of obtaining the target quality score of the target image based on the quality score to be determined includes: The average value of the quality scores to be determined is obtained as the target quality score.

6. The method according to claim 1, characterized in that, The target image is an image obtained by performing image reconstruction processing on the original image using an image reconstruction model. After obtaining the target quality score, the method further includes: If the target quality score is not greater than the second preset threshold, the reconstruction model is optimized by adjusting the parameters of the reconstruction model.

7. A referenceless image quality assessment device, comprising: The image acquisition module is used to acquire the target image to be evaluated; The first image block selection module is used to select an image block that meets a preset condition from the target image as a first image block; wherein, the first preset condition is: the first image block is a flat area; The second image block construction module is used to construct a second image block corresponding to the first image block; wherein, constructing the second image block corresponding to the first image block includes: obtaining the first pixel value of all pixels in the first image block; calculating the average value of the first pixel value as the second pixel value; and using the second pixel value as the fill value of the pixels in the second image block to construct the second image block; The target quality score acquisition module is used to obtain the target quality score of the target image based on the first image block and the second image block.

8. An electronic device comprising the apparatus of claim 7; or, The electronic device includes: Memory is used to store executable instructions; A processor, configured to execute, under the control of the executable computer program, the referenceless image quality assessment method according to any one of claims 1-6.

Images