Image quality parameter determination method and apparatus, electronic device, and storage medium

Abstract

The application provides a picture quality parameter determination method and device, electronic equipment and a storage medium. The method comprises the following steps: in response to the need to determine the parameters of any picture quality IP algorithm of a display panel to be adjusted, determining the parameters to be adjusted of the any picture quality IP algorithm; obtaining a preset image set; using a trained image generation model to generate a display effect image of any image in the image set at any value within the value range of the parameters to be adjusted, thereby forming a display effect image set; determining an evaluation index, and evaluating the images in the display effect image set according to the evaluation index; selecting a target image from the display effect image set according to the evaluation result, and determining the value of the target image as a target value.

Description

Technical Field

[0001] This application relates to the field of data analysis technology, and in particular to a method, apparatus, electronic device and storage medium for determining image quality parameters. Background Technology

[0002] With technological advancements, various types of display panels have emerged in the current display field, such as LCD (Liquid Crystal Display) panels and OLED (Organic Light-Emitting Diode) panels. Due to the different display characteristics of these panels, displaying the same image will result in certain differences. This necessitates adjusting the parameters of different display panels to varying degrees.

[0003] However, for a single display panel or a specific type of display panel, the current parameter tuning process usually requires manual experience, which is very time-consuming and laborious. Its cost and efficiency are no longer suitable for the requirements of the field. Summary of the Invention

[0004] In view of this, this application proposes a method, apparatus, electronic device and storage medium for determining image quality parameters, in order to solve or partially solve the above problems.

[0005] In view of the above objectives, firstly, this application provides a method for determining image quality parameters, comprising:

[0006] In response to the need to determine the parameters of any image quality IP algorithm of the display panel to be tuned, the parameters to be tuned for the any image quality IP algorithm are determined.

[0007] Get the preset image set;

[0008] Using the trained image generation model, within the range of values ​​of the parameters to be adjusted, generate a display effect image of any image in the image set at any value, thereby forming a display effect image set;

[0009] Determine the evaluation indicators, and evaluate the images in the display effect image set according to the evaluation indicators;

[0010] Based on the evaluation results, a target image is selected from the set of display effect images, and the value of the parameter to be adjusted corresponding to the target image is determined as the target value.

[0011] In some exemplary embodiments, the training process of the image generation model includes:

[0012] Obtain the initial images from the preset image set;

[0013] The initial image is displayed on the display panel to be adjusted, and the display screen of the display panel to be adjusted is captured by an external shooting device to generate a captured image;

[0014] The initial image and the captured image are combined to form a training set;

[0015] The image generation model is trained using the training set.

[0016] In some exemplary embodiments, the training termination condition of the image generation model includes:

[0017] After the initial image is input into the image generation model, the generated output image achieves a similarity to the captured image that meets a preset requirement.

[0018] or

[0019] The training cycle has reached the preset number of iterations.

[0020] In some exemplary embodiments, generating the captured image includes:

[0021] The captured image is cropped so that its size is consistent with the initial image.

[0022] In some exemplary embodiments, the image generation model is a U-Net deep learning model.

[0023] In some exemplary embodiments, determining the evaluation index includes:

[0024] Determine the set of metrics for image quality evaluation, and at least one reference image;

[0025] Display any reference image for different values ​​of the parameter to be adjusted, determine the value of at least one indicator in the indicator set, and thereby determine the variation pattern between the parameter to be adjusted and the at least one indicator;

[0026] The evaluation index is selected based on the fact that the change pattern is irregular.

[0027] In some exemplary embodiments, determining the value of the parameter to be adjusted corresponding to the target image as the target value includes:

[0028] The values ​​of the parameters to be adjusted corresponding to the target image are statistically summarized using the average or mode method, and the statistical summary result is used as the target value.

[0029] In some exemplary embodiments, the preset image set contains no more than 100 images, and the preset image set is classified according to photographic subject matter, with each type of image containing more than a set threshold number of elements.

[0030] Based on the same concept, in a second aspect, this application also provides an image quality parameter determining device, comprising:

[0031] The first module is used to determine the parameters of any image quality IP algorithm of the display panel to be adjusted in response to the need to determine the parameters of the image quality IP algorithm to be adjusted.

[0032] The second module is used to acquire a preset image set;

[0033] The third module is used to generate, within the range of values ​​of the parameters to be adjusted, a display effect image of any image in the image set at any value, using the trained image generation model, thereby forming a display effect image set;

[0034] The fourth module is used to determine evaluation indicators and evaluate the images in the display effect image set according to the evaluation indicators;

[0035] The fifth module is used to select from the set of display effect images based on the evaluation results, determine the target image, and determine the value of the parameter to be adjusted corresponding to the target image as the target value.

[0036] Based on the same concept, in a third aspect, this application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the method as described in any of the preceding claims.

[0037] Based on the same concept, in a fourth aspect, this application also provides a non-transitory computer-readable storage medium storing computer instructions for causing a computer to perform the method as described in any of the preceding claims.

[0038] As described above, this application provides a method, apparatus, electronic device, and storage medium for determining image quality parameters. When adjusting the parameters of any image quality IP algorithm for a display panel to be adjusted, this application first uses an image generation model to simulate and convert images in an image set into images displayed on the screen under different parameter values, forming a display effect image set. Then, the display effect image set is evaluated according to determined evaluation indicators to identify target images that meet the requirements. Finally, the parameter values ​​corresponding to these target images are used as target values ​​for adjusting the parameters of the image quality IP algorithm for the display panel to be adjusted. This method allows for image generation and evaluation with minimal operator intervention, automatically determining the required parameters, greatly improving the efficiency of image quality parameter determination, reducing costs, and significantly enhancing the user experience. Attached Figure Description

[0039] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the drawings described below are only embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0040] Figure 1 A flowchart illustrating an exemplary method provided in an embodiment of this application.

[0041] Figure 2 This is a schematic diagram illustrating the image effects presented by different gamma values ​​provided in the embodiments of this application.

[0042] Figure 3 This is a schematic diagram illustrating the effect of different indicators changing with the parameter to be adjusted, as provided in the embodiments of this application.

[0043] Figure 4 A schematic diagram of the structure of an exemplary device provided in an embodiment of this application.

[0044] Figure 5 This is a schematic diagram of the electronic device structure provided in an embodiment of this application. Detailed Implementation

[0045] To make the objectives, technical solutions, and advantages of this specification clearer, the following detailed description is provided in conjunction with specific embodiments and the accompanying drawings.

[0046] It should be noted that, unless otherwise defined, the technical or scientific terms used in the embodiments of this application should have the ordinary meaning understood by one of ordinary skill in the art to which this application pertains. The terms "first," "second," and similar terms used in the embodiments of this application do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element, object, or method step preceding the term covers the element, object, or method step listed after the term and its equivalents, without excluding other elements, objects, or method steps. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are only used to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0047] As described in the background section, parameter tuning is a time-consuming and labor-intensive task, currently relying mainly on manual experience. Furthermore, the same parameter configuration can produce significantly different display effects on different types and models of display panels. Therefore, how to reduce manual intervention while ensuring performance and improving overall efficiency has become a pressing issue in this field.

[0048] In light of the above-mentioned practical situation, this application provides a scheme for determining image quality parameters. When adjusting the parameters of any image quality IP algorithm for a display panel to be adjusted, this application first uses an image generation model to simulate and convert images in an image set into images displayed on the screen under different parameter values, forming a display effect image set. Then, the display effect image set is evaluated according to determined evaluation indicators to identify target images that meet the requirements. Finally, the parameter values ​​corresponding to these target images are used as target values ​​for adjusting the parameters of the image quality IP algorithm for the display panel to be adjusted. This method allows for image generation and evaluation with minimal operator intervention, automatically determining the required parameters, greatly improving the efficiency of image quality parameter determination, reducing costs, and significantly enhancing the user experience.

[0049] Figure 1 A flowchart illustrating an exemplary method provided in an embodiment of this application is shown.

[0050] like Figure 1 As shown in the embodiment of this application, an image quality parameter determination method is provided by way of example. This method may specifically include the following steps.

[0051] Step 102: In response to the need to determine the parameters of any image quality IP algorithm of the display panel to be adjusted, determine the parameters to be adjusted for the any image quality IP algorithm.

[0052] In this step, the display panel to be adjusted can be any type of display panel, such as an LCD panel or an OLED panel. As mentioned earlier, although the image quality parameters corresponding to different types or even different models of display panels may be similar in type, requiring the setting of parameters such as contrast, white balance, and gamma, to present the optimal image quality or the image quality preferred by the user, it is necessary to confirm the parameters for each display panel individually; they cannot be directly applied to each other. Image quality IP algorithms typically refer to a series of image processing algorithms used to improve image quality. These algorithms can be used for image enhancement, restoration, compression, and other aspects. Image quality IP algorithms are generally relatively simple image processing algorithms, and the number of parameters is incomparable to the hundreds of millions of parameters in neural network models. Image quality IP algorithms generally have only a few, at most a dozen, parameters, and the value range of these parameters is not very large. The parameters to be adjusted here are the parameters corresponding to any image quality IP algorithm.

[0053] Specifically, because each type of display panel has different display characteristics, the same algorithm parameters will produce different results on different display panels. Suppose a certain image quality IP algorithm has only one parameter, 'a'. When a=1, it has the best display effect on an LCD screen; when a=2, it has the best display effect on an OLED screen. Therefore, in order to achieve the best display effect for the same algorithm on display modules with different characteristics, it is necessary to perform targeted parameter tuning based on the display characteristics of different display modules.

[0054] In more specific scenarios, the differences between different display modules include, but are not limited to, the following aspects: (1) Display technology: Display modules can be based on different display technologies, such as liquid crystal displays (LCDs), organic light-emitting diodes (OLEDs), micro light-emitting diodes (Micro LEDs), and Mini LEDs. Each technology has its specific working principle and performance characteristics. (2) Pixel structure: Different display modules have different pixel structures. For example, LCDs usually use RGB arrangement, while OLEDs may use Pentile diamond arrangement, which affects pixel density and display effect. (3) Backlight technology: For LCD technology, backlight technology is an important difference. For example, traditional LCDs use larger LED beads as backlight, while Mini LEDs use smaller LED beads to achieve finer backlight control. (4) Contrast and color performance: Due to its self-emissive characteristics, OLED can display pure black and provide a near-infinite contrast ratio, while LCD has a relatively low contrast ratio due to the presence of a backlight layer. (5) Response time: OLED screens have a very short response time, which is suitable for displaying dynamic images, while LCD screens have a relatively long response time, which may cause ghosting in fast-moving scenes. (6) Power consumption: OLED screens typically consume less power when displaying dark or black images because each pixel can be independently controlled to turn on and off. LCD screens, on the other hand, consume relatively more power due to the presence of a backlight layer.

[0055] Step 104: Obtain the preset image set.

[0056] In this step, the preset image set serves as the reference set for parameter determination. It can consist of multiple images belonging to various fields and containing diverse elements. During parameter determination, images with different parameter effects can be generated based on these images. Then, the image with the desired effect is selected, and its corresponding parameter value becomes the required parameter value. Different images, due to variations in composition and included elements, will cause fluctuations in parameter values. Statistical analysis and integration of these values ​​ensure that the final parameter values ​​can handle various scenarios and elements. Therefore, the images in the preset image set should cover more scenes and contain more elements. Since a single parameter may correspond to multiple values ​​during parameter tuning, each image in the preset image set will generate an image corresponding to each value. If the number of images in the preset image set is too large, the image generation process will be very time-consuming. Therefore, ideally, the number of images in the preset image set should not be too large.

[0057] In other words, to ensure the final image quality algorithm is adaptable to most everyday scenarios, the image set used for algorithm validation must be small and high-quality. A large dataset would be extremely time-consuming during validation. Furthermore, the validation images must be carefully selected to cover as many scenarios as possible, and the image data from different scenarios should cover diverse colors, textures, brightness, etc. That is, a category of images should cover more scenes and elements; for example, animal images should ideally include various climates, ecosystems, and different animals.

[0058] In more specific scenarios, the preset image set contains no more than 100 images. The preset image set is categorized according to photographic subject matter, and the number of elements contained in each type of image exceeds a set threshold. Photographic subjects, by definition, can be categorized into portrait photography, architectural photography, landscape photography, animal photography, lifestyle photography, and even document photography, etc. Multiple representative images are selected from each category to collectively form the image set.

[0059] Step 106: Using the trained image generation model, generate a display effect image of any image in the image set at any value within the range of the parameters to be adjusted, thereby forming a display effect image set.

[0060] In this step, as mentioned earlier, the number of parameters to be adjusted is usually small, generally one or two. Each parameter typically has a corresponding value range, and the settable values ​​are relatively limited. Therefore, the parameter values ​​can be determined through exhaustive search, and the conclusions obtained through exhaustive search are generally the most reliable. Specifically, this involves iterating through these parameters, processing the image using the image quality IP algorithm with different parameters, and then selecting the optimal parameters. In more specific scenarios, the number of image quality IP algorithm parameters is often not large and has a fixed value range. Therefore, by iterating through the image quality IP algorithm parameters, the image can be processed to obtain the processing effects of the same image under different parameters. For example, for the gamma transformation algorithm, by setting different gamma values, the following can be obtained: Figure 2 The different effects shown.

[0061] Next, the main purpose of the image generation model is to convert the images in the image set into display images that meet the requirements. Meeting the requirements means generating display images corresponding to each value of the parameter to be adjusted, for example, display images corresponding to different gamma values ​​for the same image. The display image is the image displayed on the display panel to be adjusted. In a typical approach, image capture devices such as cameras can be used to capture images and thus create display images. However, this step directly uses the image generation model to automatically convert between the two. Ultimately, each image in the image set generates multiple corresponding display images (with different values ​​of the parameter to be adjusted), thus forming a display image set.

[0062] Furthermore, the image generation model can be any deep learning model used for "image-to-image" generation, meaning a model whose input and output are both images. In this embodiment, since both the input and output are images of the same size, a relatively mature deep learning model for "image-to-image" generation can be used, requiring only adjustment or control of the output image size. Considering factors such as data efficiency, training speed, and convergence speed, in some embodiments, the U-Net deep learning model can be used as the image generation model.

[0063] Training an image generation model first involves generating a training set. In some embodiments, the aforementioned preset image set can be used. Taking one image as the initial image, the initial image can be displayed on the display panel to be adjusted, and then an external camera can capture the displayed image to generate a captured image. The initial image and the captured image then form a training pair. Repeatedly selecting images from the image set generates the training set. During training, the initial image from a training pair is used as input, and the model's output image is compared with the captured image. The loss function can be set based on the similarity between the output image and the captured image. Of course, in some scenarios, to prevent endless loop training, the number of iterations can be limited. That is, in some embodiments, the training process of the image generation model includes: acquiring the initial image from the preset image set; displaying the initial image on the display panel to be adjusted, capturing the displayed image on the display panel using an external camera to generate a captured image; forming a training pair with the initial image and the captured image to form a training set; and training the image generation model using the training set. The training termination conditions for the image generation model include: after inputting the initial image into the image generation model, the similarity between the generated output image and the captured image reaches a preset requirement; or the number of training iterations reaches a preset number. The preset requirement and the preset number of iterations can be specifically set according to the specific scenario, for example, the preset requirement is a similarity of 99%, and the preset number of iterations is 10,000.

[0064] Furthermore, the captured image may be of different sizes due to various external conditions of the external shooting device, thus requiring cropping. The final image size output by the image generation model will be the same as the initial image size, so the captured image can be cropped to make the captured image size consistent with the initial image.

[0065] Step 108: Determine the evaluation index and evaluate the images in the display effect image set according to the evaluation index.

[0066] In this step, the first step is to determine the evaluation metrics. These metrics can be predetermined or automatically determined during execution. Specifically, the evaluation metrics are used to measure the image processing effect. PSNR (Peak Signal-to-Noise Ratio) is a commonly used evaluation metric that measures the signal-to-noise ratio of an image, reflecting its quality. However, PSNR does not always align with human visual perception, so it may be necessary to combine it with other metrics, such as: SSIM (Structural Similarity Index), which measures the visual impact and structural information of an image; VMAF (Video Multi-Method Assessment Fusion), a comprehensive evaluation metric for video content; power consumption, which is also an important consideration in some applications, such as mobile devices or embedded systems; and other factors such as processing time and memory usage, which may also need to be considered depending on the specific application scenario. Furthermore, the same evaluation metric may not be suitable for all image quality IP algorithm evaluations; it needs to be selected specifically for the particular image quality IP algorithm.

[0067] In a specific scenario, when evaluating the image display effect after processing by a certain image quality IP algorithm, evaluation metrics such as image power consumption, PSNR, SSIM, STD (Standard Deviation), NIQE (Natural Image Quality Evaluator), average gradient, and comentropy are selected. The image quality evaluation values ​​under different parameters of this image quality IP algorithm are visualized, as shown below. Figure 3 The diagram illustrates the evaluation of four test images using various metrics. In each graph, the horizontal axis represents the changes in the parameters to be adjusted in the current image quality IP algorithm, gradually increasing from left to right; the vertical axis represents the score of the evaluation metric, gradually increasing from bottom to top. The blue line represents power consumption, the green line represents PSNR, the red line represents SSIM, the yellow line represents STD, the cyan line represents NIQE, the purple line represents average gradient, and the black line represents information entropy. Solid lines represent the evaluation of the initial image, while dashed lines represent the evaluation of the captured or displayed image. In the diagram, the results of the blue, yellow, and purple lines are similar, resulting in a near-overlapping state between the lines.

[0068] from Figure 3As can be seen, with the monotonic change of the parameters to be tuned in the tested image quality IP algorithm, in this scenario, all evaluation metrics except for NIQE are monotonically increasing. That is, as the algorithm parameters increase, all other evaluation metrics also increase, making such evaluation metrics unsuitable for this algorithm. Because if these evaluation metrics are used to evaluate image quality, simply setting the algorithm parameters to their maximum values ​​would yield the optimal image quality evaluation metrics, eliminating the need for parameter tuning. Therefore, for this image quality IP algorithm, NIQE can be selected as the image quality evaluation metric. In other embodiments, the changes in different evaluation metrics vary depending on the image quality IP algorithm. Thus, for one image quality IP algorithm, NIQE might be a suitable evaluation metric, while for another, SSIM might be more appropriate. However, in different scenarios, after determining the image quality IP algorithm to be tuned, the evaluation metrics can be determined in the above manner. Of course, the specific time for determining the evaluation metrics can be before the entire method execution (where the result is directly obtained at this step) or during the execution of this step. Of course, in other embodiments, an indicator can be manually specified as the evaluation indicator, or the evaluation indicator can be determined using other methods, without specific limitations. That is, in some embodiments, determining the evaluation indicator includes: determining a set of indicators for image quality evaluation, and at least one reference image; displaying images of any reference image at different values ​​of the parameter to be adjusted; determining the value of at least one indicator in the indicator set, thereby determining the change pattern between the parameter to be adjusted and the at least one indicator; and selecting an indicator whose change pattern is irregular as the evaluation indicator. Here, the indicator set is a collection of the aforementioned indicators that can be used for image quality evaluation; and irregular change refers to... Figure 3 The variation of the NIQE indicator shown is a type of variation for which it is difficult to summarize the rules of change.

[0069] Once the specific evaluation indicators are determined, they can be used to evaluate the images in the display effect image set. The specific evaluation method is determined according to the specific evaluation indicators, and the general evaluation result is the score of the evaluation indicator.

[0070] Step 110: Select from the set of display effect images according to the evaluation results, determine the target image, and determine the value of the parameter to be adjusted corresponding to the target image as the target value.

[0071] In this step, the evaluation results may be presented in different forms in different embodiments, but generally they can be presented as scores. For the display effect image set, each group of images is an image simulating the display panel to be adjusted under different values ​​of the parameters to be adjusted. Then, the required evaluation results can be defined according to pre-set parameters, such as the specific score required. Then, for each group of images in the display effect image set, at least one image that meets the requirements can be selected (since the evaluation indicators may change irregularly, there may be more than one image that meets the requirements). These images are the target images. After collecting all the selected target images, since each target image corresponds to a value of a parameter to be adjusted, the value that meets certain requirements can be selected from these values ​​according to statistical methods. This value is the target value. In some specific scenarios, the target value can be the optimal value of the parameter to be adjusted, such as the optimal value of contrast or gamma.

[0072] Finally, the target value can be used as the final value of the parameter to be adjusted for the image quality IP algorithm, and then the same or similar adjustments can be made to the display panel to be adjusted and other display panels of the same type or model as the display panel to be adjusted.

[0073] In some embodiments, the aforementioned statistical method may be to use the average or mode method, and other statistical methods may be used depending on the specific scenario, without specific limitation. That is, in some embodiments, determining the value of the parameter to be adjusted corresponding to the target image as the target value includes: statistically summarizing the values ​​of the parameter to be adjusted corresponding to the target image using the average or mode method, and using the statistical summary result as the target value.

[0074] In more specific scenarios, for each image in the display effect image set, optimal parameters are determined based on evaluation metrics. This may involve analyzing each image individually to find the best parameter combination under specific evaluation metrics. Optimal parameter data for all images is collected, and statistical learning methods are used to analyze this data. This may include: descriptive statistical analysis to understand the distribution of parameters; correlation analysis to determine which parameters have a greater impact on the evaluation metrics; and regression analysis to predict optimal parameters under different conditions. Through statistical learning, patterns and trends between parameters can be discovered, thereby deriving universal parameters applicable to most images. These universal parameters can serve as the default settings for the algorithm to ensure good image quality performance on various images. In this way, the optimal parameters for image processing algorithms can be systematically found, and the universality of the parameters can be ensured through statistical learning. This method combines automated testing, multi-dimensional evaluation, and data analysis, which can improve the performance and applicability of the algorithm.

[0075] Finally, in a specific application scenario, the overall framework of this solution can be mainly divided into two major steps: the first step is the training of the image generation model, and the second step is the search for the target values ​​of the parameters to be tuned in the image quality IP algorithm.

[0076] In the training process of the image generation model, an initial image is first displayed on the screen of the display module. Then, an external imaging device, such as an industrial camera, is used to capture images of the module. By capturing all images in the image set in the same way, a training set for the image generation model, consisting of the initial images and the captured images, can be obtained. Using this training set to train the image generation model yields the desired image generation model.

[0077] In the process of finding the target values ​​of the parameters to be adjusted in the image quality IP algorithm, the input image is first processed using different values ​​of the parameters to be adjusted in the image quality IP algorithm to obtain a batch of images with different effects of the parameter values. Then, these images are used to generate a set of display effect images through an image generation model. Finally, the optimal image is selected from the set of display effect images using the selected evaluation index. The value of the parameters to be adjusted in the image quality IP algorithm corresponding to the optimal image is the optimal image quality IP algorithm parameter for the display panel to be adjusted.

[0078] As can be seen from the above embodiments, this application provides a method for determining image quality parameters. When adjusting the parameters of any image quality IP algorithm for a display panel to be adjusted, this application first uses an image generation model to simulate and convert images in an image set into images displayed on the screen under different parameter values, forming a display effect image set. Then, the display effect image set is evaluated according to the determined evaluation indicators to determine the target images that meet the requirements. Finally, the parameter values ​​corresponding to these target images are used as target values ​​to adjust the parameters of the image quality IP algorithm for the display panel to be adjusted. This method enables image generation and evaluation with minimal operator intervention, automatically determining the required parameters, greatly improving the efficiency of image quality parameter determination, reducing costs, and significantly enhancing the user experience.

[0079] It should be noted that the method in this application embodiment can be executed by a single device, such as a computer or server. The method in this application embodiment can also be applied in a distributed scenario, where multiple devices cooperate to complete the process. In such a distributed scenario, one of these devices may execute only one or more steps of the method in this application embodiment, and the multiple devices will interact with each other to complete the method described.

[0080] It should be noted that the above description describes specific embodiments of this application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps described in the claims can be performed in a different order than that shown in the above embodiments and still achieve the desired results. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0081] Based on the same concept, corresponding to the methods of any of the above embodiments, this application also provides an image quality parameter determination device.

[0082] refer to Figure 4 The image quality parameter determining device includes:

[0083] The first module 410 is used to determine the parameters of any image quality IP algorithm of the display panel to be adjusted in response to the need to determine the parameters of the image quality IP algorithm to be adjusted.

[0084] The second module 420 is used to acquire a preset image set.

[0085] The third module 430 is used to generate, within the range of values ​​of the parameters to be adjusted, a display effect image of any image in the image set at any given value, using the trained image generation model, thereby forming a display effect image set.

[0086] The fourth module 440 is used to determine evaluation indicators and evaluate the images in the display effect image set according to the evaluation indicators.

[0087] The fifth module 450 is used to select from the set of display effect images based on the evaluation results, determine the target image, and determine the value of the parameter to be adjusted corresponding to the target image as the target value.

[0088] In some exemplary embodiments, the training process of the image generation model includes:

[0089] Obtain the initial images from the preset image set;

[0090] The initial image is displayed on the display panel to be adjusted, and the display screen of the display panel to be adjusted is captured by an external shooting device to generate a captured image;

[0091] The initial image and the captured image are combined to form a training set;

[0092] The image generation model is trained using the training set.

[0093] In some exemplary embodiments, the training termination condition of the image generation model includes:

[0094] After the initial image is input into the image generation model, the generated output image achieves a similarity to the captured image that meets a preset requirement.

[0095] or

[0096] The training cycle has reached the preset number of iterations.

[0097] In some exemplary embodiments, generating the captured image includes:

[0098] The captured image is cropped so that its size is consistent with the initial image.

[0099] In some exemplary embodiments, the image generation model is a U-Net deep learning model.

[0100] In some exemplary embodiments, the fourth module 440 is further configured to:

[0101] Determine the set of metrics for image quality evaluation, and at least one reference image;

[0102] Display any reference image for different values ​​of the parameter to be adjusted, determine the value of at least one indicator in the indicator set, and thereby determine the variation pattern between the parameter to be adjusted and the at least one indicator;

[0103] The evaluation index is selected based on the fact that the change pattern is irregular.

[0104] In some exemplary embodiments, the fifth module 450 is further configured to:

[0105] The values ​​of the parameters to be adjusted corresponding to the target image are statistically summarized using the average or mode method, and the statistical summary result is used as the target value.

[0106] In some exemplary embodiments, the preset image set contains no more than 100 images, and the preset image set is classified according to photographic subject matter, with each type of image containing more than a set threshold number of elements.

[0107] For ease of description, the above devices are described in terms of function, divided into various modules. Of course, in implementing the embodiments of this application, the functions of each module can be implemented in one or more software and / or hardware.

[0108] The apparatus described above is used to implement the corresponding image quality parameter determination method in the foregoing embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0109] Based on the same concept, corresponding to the methods of any of the above embodiments, this application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the image quality parameter determination method as described in any of the above embodiments.

[0110] Figure 5 This embodiment illustrates a more specific hardware structure of an electronic device, which may include a processor 1010, a memory 1020, an input / output interface 1030, a communication interface 1040, and a bus 1050. The processor 1010, memory 1020, input / output interface 1030, and communication interface 1040 are interconnected internally via the bus 1050.

[0111] The processor 1010 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this specification.

[0112] The memory 1020 can be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 1020 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented by software or firmware, the relevant program code is stored in the memory 1020 and is called and executed by the processor 1010.

[0113] The input / output interface 1030 is used to connect input / output modules to realize information input and output. Input / output modules can be configured as components within the device (not shown in the figure) or externally connected to the device to provide corresponding functions. Input devices may include keyboards, mice, touchscreens, microphones, various sensors, etc., while output devices may include displays, speakers, vibrators, indicator lights, etc.

[0114] The communication interface 1040 is used to connect a communication module (not shown in the figure) to enable communication between this device and other devices. The communication module can communicate via wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0115] Bus 1050 includes a pathway for transmitting information between various components of the device, such as processor 1010, memory 1020, input / output interface 1030, and communication interface 1040.

[0116] It should be noted that although the above-described device only shows the processor 1010, memory 1020, input / output interface 1030, communication interface 1040, and bus 1050, in specific implementations, the device may also include other components necessary for normal operation. Furthermore, those skilled in the art will understand that the above-described device may only include the components necessary for implementing the embodiments of this specification, and not necessarily all the components shown in the figures.

[0117] The electronic devices described above are used to implement the corresponding image quality parameter determination methods in any of the foregoing embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0118] Based on the same concept, corresponding to the methods of any of the above embodiments, this application also provides a non-transitory computer-readable storage medium storing computer instructions for causing the computer to execute the image quality parameter determination method as described in any of the above embodiments.

[0119] The computer-readable medium of this embodiment includes permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transfer medium that can be used to store information accessible by a computing device.

[0120] The computer instructions stored in the storage medium of the above embodiments are used to cause the computer to execute the image quality parameter determination method as described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0121] Based on the same concept, corresponding to the methods of any of the above embodiments, this application also provides a computer program product, which includes computer program instructions. In some embodiments, the computer program instructions can be executed by one or more processors of a computer to cause the computer and / or the processor to perform the image quality parameter determination method. Corresponding to the execution entity for each step in each embodiment of the image quality parameter determination method, the processor executing the corresponding step can belong to the corresponding execution entity.

[0122] The computer program product of the above embodiments is used to cause the computer and / or the processor to execute the image quality parameter determination method as described in any of the above embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0123] Those skilled in the art should understand that the discussion of any of the above embodiments is merely exemplary and is not intended to imply that the scope of this application (including the claims) is limited to these examples; within the framework of this application, the technical features of the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations of different aspects of the embodiments of this application as described above, which are not provided in the details for the sake of brevity.

[0124] Additionally, to simplify the description and discussion, and to avoid obscuring the embodiments of this application, the well-known power / ground connections to integrated circuit (IC) chips and other components may or may not be shown in the provided drawings. Furthermore, the apparatus may be shown in block diagram form to avoid obscuring the embodiments of this application, and this also takes into account the fact that the details of the implementation of these block diagram apparatuses are highly dependent on the platform on which the embodiments of this application will be implemented (i.e., these details should be fully understood by those skilled in the art). While specific details (e.g., circuits) have been set forth to describe exemplary embodiments of this application, it will be apparent to those skilled in the art that the embodiments of this application can be implemented without these specific details or with variations thereof. Therefore, these descriptions should be considered illustrative rather than restrictive.

[0125] Although this application has been described in conjunction with specific embodiments thereof, many substitutions, modifications, and variations of these embodiments will be apparent to those skilled in the art from the foregoing description. For example, other memory architectures (e.g., dynamic RAM (DRAM)) may be used with the embodiments discussed.

[0126] The embodiments of this application are intended to cover all such substitutions, modifications, and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the embodiments of this application should be included within the protection scope of this application.

Claims

1. A method for determining image quality parameters, characterized in that, include: In response to the need to determine the parameters of any image quality IP algorithm of the display panel to be tuned, the parameters to be tuned for the any image quality IP algorithm are determined. Get the preset image set; Using the trained image generation model, within the range of values ​​of the parameters to be adjusted, generate a display effect image of any image in the image set at any value, thereby forming a display effect image set; Determine the evaluation indicators, and evaluate the images in the display effect image set according to the evaluation indicators; Based on the evaluation results, a target image is selected from the set of display effect images, and the value of the parameter to be adjusted corresponding to the target image is determined as the target value.

2. The method according to claim 1, characterized in that, The training process of the image generation model includes: Obtain the initial images from the preset image set; The initial image is displayed on the display panel to be adjusted, and the display screen of the display panel to be adjusted is captured by an external shooting device to generate a captured image; The initial image and the captured image are combined to form a training set; The image generation model is trained using the training set.

3. The method according to claim 2, characterized in that, The training termination conditions for the image generation model include: After the initial image is input into the image generation model, the generated output image achieves a similarity to the captured image that meets a preset requirement. or The training cycle has reached the preset number of iterations.

4. The method according to claim 2, characterized in that, The generated captured image includes: The captured image is cropped so that its size is consistent with the initial image.

5. The method according to claim 2, characterized in that, The image generation model is the U-Net deep learning model.

6. The method according to claim 1, characterized in that, The determination of evaluation indicators includes: Determine the set of metrics for image quality evaluation, and at least one reference image; Display any reference image for different values ​​of the parameter to be adjusted, determine the value of at least one indicator in the indicator set, and thereby determine the variation pattern between the parameter to be adjusted and the at least one indicator; The evaluation index is selected based on the fact that the change pattern is irregular.

7. The method according to claim 1, characterized in that, The step of determining the value of the parameter to be adjusted corresponding to the target image as the target value includes: The values ​​of the parameters to be adjusted corresponding to the target image are statistically summarized using the average or mode method, and the statistical summary result is used as the target value.

8. The method according to claim 1, characterized in that, The preset image set contains no more than 100 images. The preset image set is classified according to the subject matter of the photograph, and the number of elements contained in each type of image is higher than a set threshold.

9. A device for determining image quality parameters, characterized in that, include: The first module is used to determine the parameters of any image quality IP algorithm of the display panel to be adjusted in response to the need to determine the parameters of the image quality IP algorithm to be adjusted. The second module is used to acquire a preset image set; The third module is used to generate, within the range of values ​​of the parameters to be adjusted, a display effect image of any image in the image set at any value, using the trained image generation model, thereby forming a display effect image set; The fourth module is used to determine evaluation indicators and evaluate the images in the display effect image set according to the evaluation indicators; The fifth module is used to select from the set of display effect images based on the evaluation results, determine the target image, and determine the value of the parameter to be adjusted corresponding to the target image as the target value.

10. An electronic device, characterized in that, It includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the program, implements the method as claimed in any one of claims 1 to 8.

11. A non-transitory computer-readable storage medium, characterized in that, The non-transitory computer-readable storage medium stores computer instructions for causing a computer to perform the method described in any one of claims 1 to 8.

Images