A method, apparatus, device and medium for optimizing video images
By performing real-time analysis and partitioning optimization of video frames, combined with model-generated parameters and smoothing processing, the problems of color distortion and insufficient contrast in video display systems are solved, achieving automated video image optimization and improving user experience.
Patent Information
- Application Number
- CN202511561981.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2025-10-29
- Publication Date
- 2026-11-17
- Estimated Expiration
- 2045-10-29
AI Technical Summary
Existing video display systems suffer from problems such as color distortion, loss of detail, and insufficient contrast during video playback, resulting in a poor user experience. Existing automated adjustment methods have limited effectiveness and require manual operation by the user.
By analyzing the image quality, scene, and element regions of video frames in real time, the system generates targeted optimization parameters using a trained model, performs partitioned optimization of video frames, and combines spatial and temporal smoothing processing to achieve automated video image optimization.
It improves the accuracy and adaptability of video image optimization, enhances the user experience, and ensures the consistency and stability of image effects.
Smart Images

Figure CN121644889B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of data processing technology, and in particular to a method, apparatus, device, and medium for optimizing video images. Background Technology
[0002] With the rapid development of digital video technology and the widespread adoption of display devices, users have increasingly higher demands for video playback quality. During video playback, visual parameters such as hue, brightness, and contrast directly affect the user's viewing experience. Traditional video display systems typically employ fixed hue settings, such as cool tones, warm tones, or standard hue modes. These settings remain constant throughout the video playback process. However, video content is diverse and complex. When playing such content using fixed hue settings, problems such as color distortion, loss of detail, and insufficient contrast often occur, severely impacting the user's viewing experience.
[0003] Several improvements have been attempted in existing technologies, but certain limitations remain. First, there's adjustment based on preset modes. Most display devices offer preset display modes, such as "standard mode," "movie mode," and "game mode." These modes adjust the display effect through preset color parameters, and users can manually select the appropriate mode based on the content being viewed. However, this method offers limited optimization and requires manual adjustment by the user. Second, there's automatic brightness adjustment based on ambient light sensors. Some high-end display devices integrate ambient light sensors that can detect the intensity of light in the surrounding environment and automatically adjust the screen brightness. However, this method primarily focuses on brightness adjustment, offering limited optimization and requiring manual adjustment by the user. Third, there's manual adjustment based on user preferences. Users can manually adjust parameters such as hue, saturation, and contrast through the display device's settings menu. While flexible, this method still requires frequent manual operation, resulting in a poor user experience.
[0004] Therefore, how to automatically optimize video footage and improve the accuracy of the optimization to enhance the user experience has become an urgent problem to be solved. Summary of the Invention
[0005] This invention provides a video image optimization method, apparatus, device, and medium to address the problem of how to automatically optimize video images, improve the accuracy of optimization, and enhance user experience.
[0006] A video image optimization method includes: The system acquires the target video frame to be displayed from the video source in real time, analyzes the image of the target video frame, and determines the image quality, scene, and element area of the image elements contained in the target video frame. Obtain the user's screen configuration data, input the screen configuration data, the screen quality, and the screen scene into a pre-trained model that has undergone screen optimization, and output the parameters for screen optimization of the target video frame and the confidence level of the corresponding parameters through the pre-trained model. The target video frame is divided into regions to obtain screen partitions. For any screen partition, the screen partition is optimized based on the element region, parameters and confidence of the corresponding parameters of all screen elements to obtain optimized screen partitions. Based on all optimized screen partitions, the optimized target video frame is obtained, and the optimized target video frame is played and displayed.
[0007] A video image optimization device, comprising: The image analysis module is used to acquire the target video frame to be displayed from the video source in real time, analyze the image of the target video frame, and determine the image quality, scene, and element area of the image elements contained in the target video frame. The model prediction module is used to acquire the user's screen configuration data, input the screen configuration data, the screen quality, and the screen scene into a pre-trained model that has undergone screen optimization, and output the parameters for screen optimization of the target video frame and the confidence level of the corresponding parameters through the pre-trained model. The image optimization module is used to divide the image of the target video frame into regions to obtain image partitions. For any image partition, the image partition is optimized based on the element region, parameters and confidence of the corresponding parameters of all image elements to obtain optimized image partitions. The screen display module is used to obtain the optimized target video frame based on all optimized screen partitions, and to play and display the optimized target video frame.
[0008] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-described video image optimization method.
[0009] A computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described video image optimization method.
[0010] The aforementioned video image optimization method, apparatus, computer equipment, and storage medium analyze the target video frames to be displayed in real time to determine the image quality, scene, and element regions of the included image elements. The user's image configuration data, image quality, and scene are input into a pre-trained image optimization model, which outputs parameters for image optimization of the target video frames and the confidence levels of these parameters. The target video frames are then divided into image partitions. For any given image partition, optimization is performed based on the element regions, parameters, and confidence levels of all image elements to obtain optimized image partitions. Based on all optimized image partitions, optimized target video frames are obtained, and these optimized target video frames are then played and displayed.
[0011] Specifically, by analyzing the content of target video frames in real time, identifying their image quality, scene, and element regions, and using a pre-trained image optimization model, targeted optimization parameters and their confidence levels are generated. Based on this, fine-grained optimization is performed on the image partitions of the target video frames, resulting in optimized target video frames for playback and display. This automatically optimizes the video image, improves the accuracy and adaptability of image optimization, enhances the effect of video optimization, and thus improves the user experience. Attached Figure Description
[0012] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0013] Figure 1 This is a schematic diagram of an application environment for a video image optimization method according to an embodiment of the present invention; Figure 2 This is a flowchart of a video image optimization method according to an embodiment of the present invention; Figure 3 This is another flowchart of a video image optimization method in one embodiment of the present invention; Figure 4 This is another flowchart of a video image optimization method in one embodiment of the present invention; Figure 5 This is another flowchart of a video image optimization method in one embodiment of the present invention; Figure 6 This is a schematic diagram of a video image optimization device according to an embodiment of the present invention; Figure 7 This is a schematic diagram of a computer device according to an embodiment of the present invention. Detailed Implementation
[0014] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0015] The video image optimization method provided in this embodiment of the invention can be applied to, for example... Figure 1 The application environment shown. Specifically, this video image optimization method is applied in a video image optimization system, which includes, as shown in the example, [example of the system]. Figure 1 The diagram illustrates a client and server that communicate over a network to address the challenge of automating video image optimization, improving accuracy, and enhancing user experience. The client, also known as the user terminal, is the program that provides local services to the client, corresponding to the server. The client can be installed on, but is not limited to, various personal computers, laptops, smartphones, tablets, and portable wearable devices. The server can be a standalone server or a server cluster.
[0016] In one embodiment, such as Figure 2 As shown, a video image optimization method is provided, which is applied to... Figure 1 Taking the server in the example, the following steps are included: Step S201: Obtain the target video frame to be displayed from the video source in real time, analyze the picture of the target video frame, and determine the picture quality, picture scene and the element area of the picture elements contained in the target video frame.
[0017] Step S202: Obtain the user's screen configuration data, input the screen configuration data, screen quality and screen scene into the model that has been trained for screen optimization, and output the parameters for screen optimization of the target video frame and the confidence of the corresponding parameters through the model that has been trained for screen optimization.
[0018] In this embodiment, the target video frame can refer to a video frame to be played and displayed and requiring image optimization; the video source can refer to the source of the target video frame, such as a real-time video stream and a video file stored on a disk; the image quality can refer to the quality of the image displayed by the target video frame; the image scene can refer to the scene to which the image displayed by the target video frame belongs, such as indoor, outdoor, daytime, nighttime, sunny, and rainy days; the image element can refer to an independently identifiable entity in the image displayed by the target video frame, such as text, faces, and objects; and the element area can refer to the area occupied by the image element in the image of the target video frame.
[0019] Image configuration data can refer to image optimization parameter data combined with user preferences. The image optimization model can refer to a deep learning model that has been trained to predict image optimization parameters. This model, after training, can predict optimization parameters based on image configuration data, image quality, and image scene. The parameters can refer to the parameters used to optimize the target video frame, which may include parameters such as hue, saturation, and contrast.
[0020] Specifically, the system acquires the target video frame to be displayed from the video source in real time, analyzes the image of the target video frame, determines the image quality, scene, and element regions of the image elements contained in the target video frame, inputs the user's image configuration data, image quality, and scene into the pre-trained image optimization model, and outputs the parameters for image optimization of the target video frame and the confidence level of the corresponding parameters through the pre-trained image optimization model.
[0021] Step S203: Divide the target video frame into regions to obtain screen partitions. For any screen partition, optimize the screen partition based on the element regions, parameters, and confidence levels of all screen elements to obtain optimized screen partitions.
[0022] Step S204: Based on all optimized screen partitions, obtain the optimized target video frame, and play and display the optimized target video frame.
[0023] In this embodiment, screen partitioning can refer to the sub-regions obtained by dividing the screen of the target video frame.
[0024] Specifically, the target video frame is divided into screen partitions. For any screen partition, the screen partition is optimized based on the element area, parameters, and confidence of the corresponding parameters of all screen elements to obtain an optimized screen partition. Based on all optimized screen partitions, an optimized target video frame is obtained, and the optimized target video frame is played and displayed.
[0025] In this embodiment, by analyzing the content of the target video frame in real time, the image quality, scene, and element regions of the image elements are identified. Based on the pre-trained image optimization model, targeted optimization parameters and their confidence levels are generated. On this basis, fine-grained optimization is performed on the image partitions of the target video frame to obtain the optimized target video frame for playback and display. This automatically optimizes the video image, improves the accuracy and adaptability of image optimization, enhances the effect of video optimization, and thus improves the user experience.
[0026] In one embodiment, such as Figure 3 As shown, a video frame optimization method is provided. In step S201 above, the frame of the target video frame is analyzed to determine the scene of the target video frame, including the following steps: Step S301: For any pixel in the target video frame, determine the hue value of the pixel.
[0027] Step S302: Determine the tone distribution characteristics of the target video frame based on the tone values of all pixels.
[0028] Step S303: Based on the color tone distribution characteristics, perform scene recognition on the target video frame to obtain the scene.
[0029] In this embodiment, tone distribution features can refer to feature data that characterizes the tone distribution trend of pixels in the target video frame.
[0030] Specifically, for any pixel in the target video frame, the hue value of that pixel is calculated. Based on the hue values of all pixels, a hue histogram is generated. The hue histogram is analyzed to identify hue distribution characteristics. Based on the hue distribution characteristics, the scene of the target video frame is determined.
[0031] Optionally, the image of the target video frame is analyzed to determine the image quality of the target video frame, including: for any pixel in the image of the target video frame, determining the pixel's luminance value, gradient value, and first chroma value; determining the maximum and minimum luminance values from all luminance values; determining the contrast score of the target video frame based on the maximum and minimum luminance values; determining the sharpness score of the target video frame based on the gradient values of all pixels in the target video frame; obtaining the second chroma value of each pixel in the reference video frame; determining the color accuracy score of the target video frame based on the first chroma values of all pixels in the target video frame and the second chroma values of all pixels in the reference video frame; and obtaining the image quality based on the contrast score, sharpness score, and color accuracy score.
[0032] The first chromaticity value can refer to the chromaticity value of a pixel in the target video frame, and the second chromaticity value can refer to the chromaticity value of a pixel in the reference video frame. The chromaticity value is the LAB (L channel - brightness, A channel - green-red axis, B channel - blue-yellow axis) value. The reference video frame can refer to a video frame with standard chromaticity. The contrast score can refer to the score that measures the overall brightness of the target video frame. The sharpness score can refer to the score that measures the sharpness of the edges of objects in the target video frame. The color accuracy score can refer to the score that measures how close the colors in the target video frame are to the standard colors.
[0033] Specifically, the maximum and minimum brightness values are subtracted to obtain the difference, and then summed to obtain the sum. The difference is then divided by the sum to obtain the contrast score. A contrast score closer to 1 indicates extremely high contrast, with both very bright and very dark areas in the image. A contrast score closer to 0 indicates extremely low contrast, with the image appearing grayish and lacking tonal gradation. The average of the pixel gradient values in all target video frames is calculated to obtain the sharpness score. A higher sharpness score indicates higher overall image sharpness.
[0034] For any pixel in the target video frame, determine the second chromaticity value of the corresponding pixel color in the reference video frame. Calculate the differences between the first and second chromaticity values of that pixel in three dimensions: lightness, red-green hue, and yellow-blue hue. Square each difference. Sum the three squared results and take the square root of the sum to obtain the color accuracy score of that pixel. Calculate the average of the color accuracy scores of all pixels in the target video frame to obtain the color accuracy score of the target video frame. The smaller the color accuracy score, the higher the color accuracy.
[0035] Optionally, the image of the target video frame is analyzed to determine the element regions of the image elements contained in the target video frame, including: analyzing the image of the target video frame using optical character recognition technology to determine the text elements and corresponding text element regions contained in the target video frame; and analyzing the image of the target video frame using a preset target recognition algorithm to determine the face elements, corresponding face element regions, object elements, and corresponding object element regions contained in the target video frame.
[0036] That is, by using optical character recognition technology, the text and the area occupied by the corresponding text in the target video frame are identified, and by using a preset target recognition algorithm (a pre-set algorithm for recognizing objects and faces), the faces, objects, the area occupied by the faces, and the area occupied by the objects in the target video frame are identified.
[0037] In this embodiment, multiple quality indicators such as scene recognition, contrast, sharpness, and color accuracy are obtained by analyzing the tonal distribution features of the image. Key visual elements and corresponding element regions are located based on optical character recognition and target recognition algorithms. The basic condition of the target video frame is objectively evaluated, providing an objective data basis for subsequent image optimization, ensuring the targeting of image optimization, and improving the effect of image optimization.
[0038] In one embodiment, such as Figure 4 As shown, a video image optimization method is provided. In step S203 above, for any image partition, the image partition is optimized based on the element region, parameters, and confidence levels of all image elements to obtain the optimized image partition. The method includes the following steps: Step S401: For any screen partition, determine the element area contained in the screen partition based on the element area of all screen elements, and determine the area ratio of screen elements in the screen partition based on the screen partition and the element area contained in the screen partition.
[0039] Step S402: Determine the weight of the screen partition based on the area ratio, and optimize the screen partition based on the weight, all parameters and the confidence of the corresponding parameters to obtain the optimized screen partition.
[0040] In this embodiment, the region proportion can refer to the proportion of the element area in the corresponding region of the screen partition, and the weight can refer to the value that represents the importance of the screen partition based on the region proportion.
[0041] Specifically, for any screen partition, the element region in the screen partition is determined. Based on the element region in the screen partition and the corresponding region of the screen partition, the region ratio of the screen elements in the screen partition is determined. Based on the region ratio, the weight of the screen partition is determined. Based on the weight, all parameters and the confidence of the corresponding parameters, the screen partition is optimized to obtain the optimized screen partition.
[0042] Optionally, the image partition is optimized based on the weights, all parameters, and the confidence level of the corresponding parameters to obtain the optimized image partition. This includes: for any parameter, determining the initial optimization strength of the parameter based on the confidence level of the parameter; adjusting the initial optimization strength according to the weights to obtain the target optimization strength; and optimizing the image partition based on all parameters and the target optimization strength of the corresponding parameters to obtain the optimized image partition.
[0043] The initial optimization intensity can refer to the parameter optimization intensity determined based on confidence level, while the target optimization intensity can refer to the parameter optimization intensity after weight adjustment.
[0044] That is, for any parameter, the initial optimization strength of the parameter is determined according to the confidence level of the parameter (an aggressive strategy is adopted when the confidence level is high and a conservative strategy is adopted when the confidence level is low). The initial optimization strength of the parameter is adjusted according to the weight of the screen partition to obtain the target optimization strength. Based on all parameters and the target optimization strength of the corresponding parameters, the screen partition is optimized to obtain the optimized screen partition.
[0045] In this embodiment, by calculating the regional proportion of screen elements in each screen partition and dynamically allocating optimization weights accordingly, the initial optimization intensity corresponding to the confidence level of the model output parameters is adjusted based on the weights. This achieves adaptive screen optimization based on the importance of partition content, improves the accuracy and adaptability of screen optimization, and enhances the effect of video optimization.
[0046] In one embodiment, such as Figure 5 As shown, a video frame optimization method is provided. In step S204 above, the optimized target video frame is obtained based on all optimized frame partitions, and the optimized target video frame is played and displayed. The method includes the following steps: Step S501: Perform spatial smoothing on all optimized image partitions to obtain optimized target video frames.
[0047] Step S502: Obtain the optimized video frame corresponding to the previous video frame of the target video frame, perform temporal smoothing on the optimized video frame corresponding to the previous video frame and the optimized target video frame, and play and display the optimized target video frame.
[0048] Specifically, spatial smoothing is performed on all optimized screen segments to achieve a smooth transition between different optimized screen segments, blurring the boundaries between segments, determining the previous video frame of the target video frame, obtaining the optimized video frame corresponding to the previous video frame, and performing temporal smoothing on the optimized target video frame and the optimized video frame corresponding to the previous video frame to avoid flickering caused by sudden changes in brightness, color, or details between frames.
[0049] In this embodiment, by introducing a dual smoothing mechanism of space and time, the block effect and boundary traces that may be generated by partition optimization are eliminated in the spatial dimension, ensuring the uniformity and naturalness of the overall picture. In the temporal dimension, the jitter and flicker of the inter-frame picture are effectively suppressed, ensuring the continuity and stability of video playback, thereby improving the user experience.
[0050] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.
[0051] In one embodiment, a video image optimization device is provided, which corresponds one-to-one with the video image optimization method described in the above embodiments. For example... Figure 6 As shown, the video image optimization device includes an image analysis module 61, a model prediction module 62, an image optimization module 63, and an image display module 64. Detailed descriptions of each functional module are as follows: The image analysis module 61 is used to acquire the target video frame to be displayed from the video source in real time, analyze the image of the target video frame, and determine the image quality, image scene, and element area of the image elements contained in the target video frame. The model prediction module 62 is used to acquire the user's screen configuration data, input the screen configuration data, the screen quality and the screen scene into the model that has been trained for screen optimization, and output the parameters for screen optimization of the target video frame and the confidence level of the corresponding parameters through the model that has been trained for screen optimization. The image optimization module 63 is used to divide the image of the target video frame into regions to obtain image partitions. For any image partition, the image partition is optimized according to the element region, parameters and confidence of the corresponding parameters of all image elements to obtain optimized image partitions. The screen display module 64 is used to obtain the optimized target video frame based on all optimized screen partitions, and to play and display the optimized target video frame.
[0052] Optionally, the aforementioned image analysis module 61 includes: A tone determination unit is used to determine the tone value of any pixel in the frame of the target video frame. The feature determination unit is used to determine the tone distribution features of the target video frame based on the tone values of all pixels; The scene recognition unit is used to perform scene recognition on the target video frame based on the tone distribution features to obtain the scene.
[0053] Optionally, the aforementioned image analysis module 61 includes: A pixel determination unit is used to determine the brightness value, gradient value, and first chromaticity value of any pixel in the frame of the target video frame. A contrast determination unit is used to determine the maximum brightness value and the minimum brightness value from all brightness values, and to determine the contrast score of the target video frame based on the maximum brightness value and the minimum brightness value. The sharpness determination unit is used to determine the sharpness score of the target video frame based on the gradient values of all pixels in the target video frame. A color determination unit is used to obtain the second chromaticity value of each pixel in the reference video frame, and determine the color accuracy score of the target video frame based on the first chromaticity values of all pixels in the target video frame and the second chromaticity values of all pixels in the reference video. A quality determination unit is used to determine the image quality based on the contrast score, the sharpness score, and the color accuracy score.
[0054] Optionally, the aforementioned image analysis module 61 includes: The text recognition unit is used to analyze the image of the target video frame using optical character recognition technology to determine the text elements contained in the target video frame and the element regions of the corresponding text elements. The target recognition unit is used to analyze the image of the target video frame through a preset target recognition algorithm to determine the face elements, the element regions corresponding to the face elements, the object elements, and the element regions corresponding to the object elements contained in the target video frame.
[0055] Optionally, the aforementioned image optimization module 63 includes: The proportion determination unit is used to determine the element area contained in the screen partition based on the element area of all screen elements for any screen partition, and to determine the area proportion of the screen elements in the screen partition based on the screen partition and the element area contained in the screen partition. The partition optimization unit is used to determine the weight of the screen partition based on the area ratio, and optimize the screen partition based on the weight, all parameters and the confidence level of the corresponding parameters to obtain the optimized screen partition.
[0056] Optionally, the above-mentioned partition optimization unit includes: An initial determination subunit is used to determine the initial optimization strength of any parameter based on the confidence level of the parameter. An adjustment subunit is used to adjust the initial optimization intensity according to the weights to obtain the target optimization intensity; The parameter optimization subunit is used to optimize the image partition based on all parameters and the target optimization intensity of the corresponding parameters, so as to obtain the optimized image partition.
[0057] Optionally, the above-mentioned screen display module 64 includes: The spatial smoothing unit is used to smooth all optimized image partitions in the spatial dimension to obtain the optimized target video frame. The time smoothing unit is used to obtain the optimized video frame corresponding to the previous video frame of the target video frame, perform time-dimension smoothing on the optimized video frame corresponding to the previous video frame and the optimized target video frame, and play and display the optimized target video frame.
[0058] Specific limitations regarding the video image optimization device can be found in the limitations of the video image optimization method described above, and will not be repeated here. Each module in the aforementioned video image optimization device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the corresponding operations of each module.
[0059] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 7 As shown, the computer device includes a processor, memory, network interface, and database connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and the database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database stores target video frames. The network interface communicates with external terminals via a network connection. When the computer program is executed by the processor, it implements a video image optimization method.
[0060] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the video image optimization method described in the above embodiment, for example... Figure 2 As shown in S201-S204, or Figures 3 to 5 As shown, to avoid repetition, it will not be described again here. Alternatively, when the processor executes the computer program, it implements the functions of each module / unit in this embodiment of the video image optimization device, for example, Figure 6 The functions of the image analysis module 61, model prediction module 62, image optimization module 63, and image display module 64 shown are not described again here to avoid repetition.
[0061] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When executed by a processor, the computer program implements the video image optimization method described in the above embodiment, for example... Figure 2 As shown in S201-S204, or Figures 3 to 5As shown, to avoid repetition, it will not be described again here. Alternatively, when the processor executes the computer program, it implements the functions of each module / unit in this embodiment of the video image optimization device, for example, Figure 6 The functions of the image analysis module 61, model prediction module 62, image optimization module 63, and image display module 64 shown are not described again here to avoid repetition.
[0062] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.
[0063] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.
[0064] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.
Claims
1. A method for optimizing video footage, characterized in that, include: The system acquires the target video frame to be displayed from the video source in real time, analyzes the image of the target video frame, and determines the image quality, scene, and element area of the image elements contained in the target video frame. The system obtains the user's screen configuration data, inputs the screen configuration data, the screen quality, and the screen scene into a pre-trained model that has undergone screen optimization, and outputs the parameters for screen optimization of the target video frame and the confidence level of the corresponding parameters through the pre-trained model. The screen configuration data refers to the screen optimization parameter data combined with the user's preferences. The target video frame is divided into regions to obtain screen partitions. For any screen partition, the screen partition is optimized based on the element region, parameters and confidence of the corresponding parameters of all screen elements to obtain optimized screen partitions. Based on all optimized screen partitions, the optimized target video frame is obtained, and the optimized target video frame is played and displayed. For any given screen partition, the optimized screen partition is obtained by optimizing the screen partition based on the element region, parameters, and confidence levels of all screen elements, including: For any screen partition, determine the element area contained in the screen partition based on the element area of all screen elements, and determine the area ratio of the screen elements in the screen partition based on the screen partition and the element area contained in the screen partition. Based on the area proportion, the weight of the screen partition is determined. Based on the weight, all parameters and the confidence level of the corresponding parameters, the screen partition is optimized to obtain the optimized screen partition. The step of optimizing the screen partition based on the weights, all parameters, and the confidence levels of the corresponding parameters to obtain the optimized screen partition includes: For any parameter, determine the initial optimization strength of the parameter based on the confidence level of the parameter; Based on the weights, the initial optimization intensity is adjusted to obtain the target optimization intensity; Based on all parameters and the target optimization intensity of the corresponding parameters, the image partition is optimized to obtain the optimized image partition.
2. The video image optimization method according to claim 1, characterized in that, The step of analyzing the scene of the target video frame to determine the scene of the target video frame includes: For any pixel in the target video frame, determine the hue value of the pixel; The tonal distribution characteristics of the target video frame are determined based on the tonal values of all pixels. Based on the color tone distribution characteristics, scene recognition is performed on the target video frame to obtain the scene.
3. The video image optimization method according to claim 1, characterized in that, The step of analyzing the image of the target video frame to determine the image quality of the target video frame includes: For any pixel in the target video frame, determine the pixel's luminance value, gradient value, and first chromaticity value; Determine the maximum and minimum brightness values from all brightness values, and determine the contrast score of the target video frame based on the maximum and minimum brightness values; The sharpness score of the target video frame is determined based on the gradient values of all pixels in the target video frame. Obtain the second chromaticity value of each pixel in the reference video frame, and determine the color accuracy score of the target video frame based on the first chromaticity values of all pixels in the target video frame and the second chromaticity values of all pixels in the reference video. The image quality is obtained based on the contrast score, the sharpness score, and the color accuracy score.
4. The video image optimization method according to claim 1, characterized in that, The step of analyzing the target video frame to determine the element regions of the image elements contained in the target video frame includes: By using optical character recognition technology, the image of the target video frame is analyzed to determine the text elements contained in the target video frame and the element regions of the corresponding text elements. By using a preset target recognition algorithm, the image of the target video frame is analyzed to determine the face elements, the element regions corresponding to the face elements, the object elements, and the element regions corresponding to the object elements contained in the target video frame.
5. The video image optimization method according to claim 1, characterized in that, The step of obtaining optimized target video frames based on all optimized screen partitions and then playing and displaying the optimized target video frames includes: Smoothing is performed on all optimized image partitions in the spatial dimension to obtain the optimized target video frame; Obtain the optimized video frame corresponding to the previous video frame of the target video frame, perform temporal smoothing on the optimized video frame corresponding to the previous video frame and the optimized target video frame, and then play and display the optimized target video frame.
6. A video image optimization device, characterized in that, include: The image analysis module is used to acquire the target video frame to be displayed from the video source in real time, analyze the image of the target video frame, and determine the image quality, scene, and element area of the image elements contained in the target video frame. The model prediction module is used to acquire the user's screen configuration data, input the screen configuration data, the screen quality, and the screen scene into a pre-trained model that has undergone screen optimization, and output the parameters for screen optimization of the target video frame and the confidence level of the corresponding parameters through the pre-trained model that has undergone screen optimization. The screen configuration data refers to the screen optimization parameter data combined with the user's preferences. The image optimization module is used to divide the image of the target video frame into regions to obtain image partitions. For any image partition, the image partition is optimized based on the element region, parameters and confidence of the corresponding parameters of all image elements to obtain optimized image partitions. The screen display module is used to obtain the optimized target video frame based on all optimized screen partitions, and to play and display the optimized target video frame. The image optimization module includes: The proportion determination unit is used to determine the element area contained in the screen partition based on the element area of all screen elements for any screen partition, and to determine the area proportion of the screen elements in the screen partition based on the screen partition and the element area contained in the screen partition. The partition optimization unit is used to determine the weight of the screen partition based on the area ratio, and optimize the screen partition based on the weight, all parameters and the confidence level of the corresponding parameters to obtain the optimized screen partition. The partition optimization unit includes: An initial determination subunit is used to determine the initial optimization strength of any parameter based on the confidence level of the parameter. An adjustment subunit is used to adjust the initial optimization intensity according to the weights to obtain the target optimization intensity; The parameter optimization subunit is used to optimize the image partition based on all parameters and the target optimization intensity of the corresponding parameters, so as to obtain the optimized image partition.
7. A computer device, characterized in that, The computer device includes a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the video image optimization method as described in any one of claims 1 to 5.
8. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the video image optimization method as described in any one of claims 1 to 5.
Citation Information
Patent Citations
Image processing method and device, equipment and medium
CN113297937A
Image quality improving method and device
CN113992859A
Image quality enhancement method and device based on FPGA chip
CN120013800A
Automatic exposure method and device and storage medium
CN120034742A
Region-based picture enhancement method and device
CN120263999A
