Equipment control method and system for Mini LED display

Abstract

The invention relates to the technical field of display screen control, in particular to an equipment control method and system for Mini LED display, and the method comprises the steps: playing a training video set through to-be-controlled equipment, obtaining a playing feature vector set, calculating a video similarity set of the training video set, obtaining a fitness key value set according to the playing feature vector set, and obtaining a fitness key value set according to the fitness key value set; extracting a training fitness key value group, iterating the particle swarm based on training feature vectors in the training fitness key value group and the video similarity set to obtain an optimal brightness vector set, classifying the optimal brightness vector set based on the training video set to obtain a same-video brightness vector set, and obtaining a same-video brightness vector set according to the same-video brightness vector set. The fitness key value set is obtained, the steps are repeated until each fitness value is larger than the fitness threshold value, and the optimal fitness key value set is obtained. According to the invention, the brightness regulation and control accuracy of Mini LED display can be improved, and the watching experience of a user is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of display screen control, and particularly to a device control method and system for Mini LED display. Background Art

[0002] With the development of technology, Mini LED display devices, with their advantages such as high brightness, high contrast, and excellent color performance, are widely used in high-end display devices, such as high-end TVs, professional monitors, laptops, and tablets. In order to fully utilize the potential of Mini LED display devices and provide the best viewing experience, precise brightness control of the devices is particularly important.

[0003] Traditional brightness control methods for Mini LED display devices often adjust based on preset brightness values or simple sensor feedback. Although this method can achieve brightness control to a certain extent, it is difficult to respond to the dynamic changes of video content in real time and accurately, and cannot achieve the best brightness adjustment in different scenarios, thus affecting the user's viewing experience. Summary of the Invention

[0004] The present invention provides a device control method and system for Mini LED display, whose main purpose is to improve the accuracy of brightness regulation of Mini LED display and enhance the user's viewing experience.

[0005] To achieve the above object, a device control method for Mini LED display provided by the present invention includes: Receiving a display control instruction, and determining a device to be controlled based on the display control instruction, where the device to be controlled includes a plurality of areas to be controlled; Using the device to be controlled to play a pre-acquired training video set, obtaining a set of playback feature vectors, and calculating a set of video similarity degrees of the training video set, where each playback feature vector in the set of playback feature vectors includes: a playback brightness vector and a playback color vector; Using a preset questionnaire to obtain a set of fitness values of the training video set, and performing key-value pairing on each playback feature vector in the set of playback feature vectors according to the set of fitness values to obtain a set of fitness key-value groups; Successively extracting training fitness key-value groups in the set of fitness key-value groups, and determining training feature vectors in the training fitness key-value groups, where the training feature vectors include: a brightness training vector and a color training vector; Initializing a pre-constructed particle swarm based on the training feature vectors and the set of video similarity degrees to obtain an initial particle swarm, where the particle swarm includes a plurality of particles; Iterate the initial particle swarm to obtain the optimal position, and determine the optimal brightness vector based on the optimal position. Aggregate the optimal brightness vectors to obtain the optimal brightness vector set; Based on the training video set, classify the optimal brightness vector set to obtain the same-video brightness vector group set. According to the same-video brightness vector group set, play the training video set to obtain the adjusted fitness key value group set; Take the adjusted fitness key value group set as the fitness key value group set, and return to the step of obtaining the fitness value set of the training video set using the preset questionnaire until each fitness value in the fitness value set is greater than the preset fitness threshold, and obtain the optimal fitness key value group set. Based on the optimal fitness key value group set, complete the control of the device to be controlled.

[0006] Optionally, playing the pre-obtained training video set using the device to be controlled to obtain the playback feature vector set includes: Perform the following operations on each training video in the training video set: Set the start playback time, and at the start playback time, play the training video using multiple areas to be controlled in the device to be controlled; According to the start playback time, sample each area to be controlled in the playback step to obtain multiple playback brightness values and multiple playback color values, where the playback brightness values and playback color values correspond one-to-one to the areas to be controlled; According to the multiple playback brightness values and multiple playback color values, obtain the playback feature vector, where the playback feature vector includes: the playback brightness vector and the playback color vector; Calculate the intermediate sampling time according to the preset sampling interval and the start playback time; Take the intermediate sampling time as the start playback time, and return to the step of sampling each area to be controlled in the playback step according to the start playback time until the intermediate sampling time is not less than the preset end playback time; Aggregate the playback feature vectors to obtain the playback feature vector group, and merge the playback feature vector groups corresponding to each training video to obtain the playback feature vector set.

[0007] Optionally, calculating the video similarity set of the training video set includes: Extract the first training video in the training video set, and construct the video feature matrix of the first training video, where the video feature matrix includes color features; Remove the first training video from the training video set to obtain the removed training video set, and sequentially extract the comparison training videos in the removed training video set; Construct the comparison feature matrix of the comparison training video, and calculate the video similarity based on the video feature matrix and the comparison feature matrix; Use the filtered training video set as the training video set, and return the step of extracting the first training video from the training video set until the filtered training video set is an empty set; Summarize the video similarities to obtain a video similarity set.

[0008] Optionally, constructing the video feature matrix of the first training video includes: Extract a set of single-frame pictures from the first training video according to a preset number of pictures; Extract features from each single-frame image in the set of single-frame pictures to obtain a set of picture feature vectors, where the feature extraction includes: color feature extraction, and the picture feature vectors in the set of picture feature vectors correspond one by one to the single-frame pictures in the set of single-frame pictures; Construct a video feature matrix according to the set of picture feature vectors.

[0009] Optionally, using a preset questionnaire to obtain a set of fitness values for the training video set includes: Sequentially extract the completed playback videos from the training video set; Based on the completed playback videos, identify multiple video viewers, and send the questionnaire to the multiple video viewers to obtain multiple viewing experience data; Perform mean calculation based on the multiple viewing experience data to obtain comprehensive experience data, and calculate the fitness value according to the comprehensive experience data; Summarize the fitness values corresponding to each completed playback video to obtain a set of fitness values.

[0010] Optionally, initializing a pre-constructed particle swarm based on the training feature vector and the video similarity set to obtain an initial particle swarm includes: Remove the training fitness key-value groups from the set of fitness key-value groups to obtain a set of comparison fitness key-value groups; Sequentially extract comparison fitness key-value groups from the set of comparison fitness key-value groups, and determine the comparison parameter groups in the comparison fitness key-value groups, where the comparison parameter groups include: brightness comparison vectors, color comparison vectors, and comparison fitness values; Sequentially extract particles in the particle swarm to generate an initial velocity of the particles; Generate an initial query vector based on the brightness training vector in the training feature vector, and record the initial query vector as the initial position; According to the initial query vector, determine the fitness value weight and comparison similarity of the comparison fitness key-value group; Summarize the fitness value weights, comparison similarities, and comparison fitness values of each comparison fitness key-value group in the set of comparison fitness key-value groups to obtain a set of fitness value weights, a set of comparison similarities, and a set of comparison fitness values; Calculate the initial fitness value according to the fitness value weight set, the comparison similarity set, and the comparison fitness value set; Generate initial particles based on the initial fitness value, the initial position, and the initial velocity, and aggregate the initial particles to obtain an initial particle swarm.

[0011] Optionally, the determining the fitness value weight and the comparison similarity of the comparison fitness key value group according to the initial query vector includes: Calculate the fitness value weight according to the brightness comparison vector, the initial query vector, the color training vector, and the color comparison vector; Respectively determine the main training video and the comparison training video corresponding to the training fitness key value group and the comparison fitness key value group, and identify the comparison similarity between the main training video and the comparison training video in the video similarity set.

[0012] Optionally, the calculating the fitness value weight according to the brightness comparison vector, the initial query vector, the color training vector, and the color comparison vector includes: Calculate the fitness value weight using the following formula: Wherein, represents the fitness value weight, represents the activation function, represents the brightness comparison vector, represents the initial query vector, represents the vector dimension of the brightness comparison vector, represents the color comparison vector, represents the color training vector, represents the vector dimension of the color comparison vector.

[0013] Optionally, the calculating the initial fitness value includes: Calculate the initial fitness value according to the fitness value weight set, the comparison similarity set, and the comparison fitness value set using the following formula: Wherein, represents the initial fitness value, represents the number of comparison similarities in the comparison similarity set, represents the th comparison similarity in the comparison similarity set, represents the th fitness value weight in the fitness value weight set, represents the th comparison fitness value in the comparison fitness value set.

[0014] To achieve the above object, the present invention further provides a device control system for Mini LED display, including: A playback vector extraction module, configured to receive a display control instruction, determine a device to be controlled based on the display control instruction, where the device to be controlled includes a plurality of areas to be controlled, and use the device to be controlled to play a pre-acquired training video set to obtain a playback feature vector set, and calculate a video similarity set of the training video set, where each playback feature vector in the playback feature vector set includes: a playback brightness vector and a playback color vector; A fitness survey module, configured to obtain a fitness value set of the training video set by using a preset questionnaire, perform key-value pairing on each playback feature vector in the playback feature vector set according to the fitness value set to obtain a fitness key-value group set, sequentially extract training fitness key-value groups in the fitness key-value group set, and determine training feature vectors in the training fitness key-value groups, where the training feature vectors include: a brightness training vector and a color training vector; An optimal brightness determination module, configured to initialize a pre-constructed particle swarm based on the training feature vectors and the video similarity set to obtain an initial particle swarm, where the particle swarm includes a plurality of particles, perform iteration on the initial particle swarm to obtain an optimal position, and determine an optimal brightness vector based on the optimal position, and summarize the optimal brightness vectors to obtain an optimal brightness vector set; A video playback return module, configured to classify the optimal brightness vector set based on the training video set to obtain a same-video brightness vector group set, play the training video set according to the same-video brightness vector group set to obtain an adjusted fitness key-value group set, use the adjusted fitness key-value group set as the fitness key-value group set, and return to the step of obtaining the fitness value set of the training video set by using the preset questionnaire until each fitness value in the fitness value set is greater than a preset fitness threshold, and obtain an optimal fitness key-value group set.

[0015] To solve the above problems, the present invention further provides an electronic device, where the electronic device includes: A memory, storing at least one instruction; and A processor, configured to execute the instruction stored in the memory to implement the above-mentioned device control method for Mini LED display.

[0016] To solve the above problems, the present invention further provides a computer-readable storage medium, where at least one instruction is stored in the computer-readable storage medium, and the at least one instruction is executed by a processor in an electronic device to implement the above-mentioned device control method for Mini LED display.

[0017] To solve the problems described in the background art, the present invention first plays a training video set and obtains a set of playback feature vectors and a set of video similarities, which provides feature data support based on the actual playback scenario for subsequent brightness optimization, helps to more accurately analyze the performance of different video contents under different brightness conditions, and provides a data basis for brightness optimization. Then, it collects the subjective viewing experience of users through a questionnaire, quantifies it into a fitness value, and associates the fitness value with the playback feature vector, which can combine the subjective feelings of users with objective brightness and color characteristics, provide feedback basis at the user perception level for subsequent optimization algorithms, and make the brightness optimization more in line with the actual viewing needs of users. Further, based on the training feature vector and the set of video similarities, the particle swarm is initialized to obtain an initial particle swarm. By combining the training feature vector and the set of video similarities to initialize the particle swarm, the distribution of the initial particle swarm can be made more targeted and reasonable, which helps to improve the efficiency and effect of the particle swarm optimization algorithm. Then, the initial particle swarm is iterated to obtain the optimal position, and the optimal brightness vector is determined based on the optimal position. The optimal brightness vectors are summarized to obtain a set of optimal brightness vectors. Through this iterative optimization process, the brightness control strategy can be continuously adjusted to gradually approach the optimal brightness vector, thereby realizing brightness optimization based on the user's viewing experience. Finally, by using the adjusted fitness key value group set as the fitness key value group set to complete the loop iteration optimization and obtaining the optimal fitness key value group set, through this loop iteration optimization, the brightness control strategy can be continuously adjusted and improved until the fitness threshold is reached, thereby ensuring that the final brightness control scheme can meet the viewing experience requirements of users. Therefore, the present invention can improve the brightness regulation accuracy of Mini LED displays and enhance the user's viewing experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 FIG. is a schematic flowchart of a device control method for Mini LED display provided by an embodiment of the present invention; Figure 2 FIG. is a functional module diagram of a device control system for Mini LED display provided by an embodiment of the present invention; Figure 3 FIG. is a schematic structural diagram of an electronic device for implementing the device control method for Mini LED display provided by an embodiment of the present invention.

[0019] DESCRIPTION OF REFERENCE NUMERALS: 1. Electronic device; 10. Processor; 11. Memory; 12. Bus.

[0020] The realization, functional features and advantages of the object of the present invention will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0021] It should be understood that the specific embodiments described herein are merely used to explain the present invention and are not intended to limit the present invention.

[0022] An embodiment of the present application provides a device control method for Mini LED display. The execution subject of the device control method for Mini LED display includes, but is not limited to, at least one of electronic devices such as a server, a terminal, etc. that can be configured to execute the method provided in the embodiment of the present application. In other words, the device control method for Mini LED display can be executed by software or hardware installed in a terminal device or a server device, and the software can be a blockchain platform. The server includes, but is not limited to: a single server, a server cluster, a cloud server, or a cloud server cluster, etc.

[0023] Refer to Figure 1 As shown, it is a schematic flowchart of a device control method for Mini LED display provided by an embodiment of the present invention. In this embodiment, the device control method for Mini LED display includes: S1. Receive a display control instruction, and determine a device to be controlled based on the display control instruction, where the device to be controlled includes a plurality of areas to be controlled.

[0024] It can be understood that the display control instruction refers to an instruction manually initiated to control a specific display device. The device to be controlled refers to the specific display device pointed out by the display control instruction. This device to be controlled is a Mini LED display device, and the device to be controlled is composed of multiple screens. The area to be controlled refers to the screen in the device to be controlled.

[0025] S2. Use the device to be controlled to play a pre-acquired training video set, obtain a set of playback feature vectors, and calculate a set of video similarity degrees of the training video set, where each playback feature vector in the set of playback feature vectors includes: a playback brightness vector and a playback color vector.

[0026] It can be understood that the training video set refers to a set of videos used to train the device to be controlled. There are multiple training videos in this training video set. And to ensure the comprehensiveness of training, each training video has a different type, for example: natural scenery videos, including scenes such as mountains, forests, lakes, etc., tunnel or night scene videos in low-light environments, fast-moving scene videos, such as sports events or racing scenes, etc.

[0027] Furthermore, when using the device to be controlled to play the training video, change the backlight brightness of the device to be controlled at different playback times, so that the training video has different visual effects, and use the brightness corresponding to the best visual effect as the actual brightness for comparison in actual control, where the visual effect corresponds to the subsequent fitness value.

[0028] It is understandable that the playback feature vector refers to a vector representing the screen performance during the playback of the training video. This vector represents the features of the training video from multiple dimensions, including the brightness dimension and the color dimension. Among them, the brightness dimension and the color dimension correspond to the playback brightness vector and the color vector respectively. The video similarity set includes multiple video similarities, and the video similarity refers to the similarity between two different training videos in the training video set. This similarity is the similarity in the screen performance of the training video, rather than the similarity in content.

[0029] Specifically, using the device to be controlled to play the pre-acquired training video set to obtain a playback feature vector set includes: Perform the following operations on each training video in the training video set: Set the start playback time, and at the start playback time, use multiple areas to be controlled in the device to be controlled to play the training video; According to the start playback time, sample each area to be controlled in the playback step to obtain multiple playback brightness values and multiple playback color values. Among them, the playback brightness values and the playback color values correspond one-to-one to the areas to be controlled; According to the multiple playback brightness values and the multiple playback color values, obtain the playback feature vector, where the playback feature vector includes: a playback brightness vector and a playback color vector; According to the preset sampling interval and the start playback time, calculate the intermediate sampling time; Use the intermediate sampling time as the start playback time, and return to the step of sampling each area to be controlled in the playback step according to the start playback time until the intermediate sampling time is not less than the preset end playback time; Summarize the playback feature vectors to obtain a playback feature vector group, and merge the playback feature vector groups corresponding to each training video to obtain a playback feature vector set.

[0030] It should be explained that the start playback time refers to the time when the training video starts to be played. The sampling refers to using a brightness sensor and a color sensor to detect the brightness and color in each area to be controlled. The playback brightness value is the brightness value in an area to be controlled (unit: nits), and the playback color value refers to the average value of the color values of each lamp bead in an area to be controlled. Among them, the color value is the weighted average value of the respective color channel values of the corresponding lamp bead.

[0031] Further, obtaining the playback feature vector according to the multiple playback brightness values and the multiple playback color values is: The playback feature vector is , where represents the playback brightness value, Indicates the playback color value. The sampling interval refers to a constant set artificially. The intermediate sampling time is expressed as: , where represents the intermediate sampling time, represents the start playback time, represents the sampling interval. Merging the playback feature vector groups corresponding to each training video to obtain a playback feature vector set: The playback feature vector group corresponding to training video A is (vector A1, vector A2), and the playback feature vector group corresponding to training video B is (vector B1, vector B2). Then, after merging the playback feature vector groups corresponding to training video A and training video B, the obtained playback feature vector set is (vector A1, vector A2, vector B1, vector B2).

[0032] Specifically, calculating the video similarity set of the training video set includes: Extracting the first training video from the training video set and constructing a video feature matrix of the first training video, where the video feature matrix includes color features; Removing the first training video from the training video set to obtain a removed training video set, and sequentially extracting comparison training videos from the removed training video set; Constructing a comparison feature matrix of the comparison training video, and calculating the video similarity based on the video feature matrix and the comparison feature matrix; Taking the removed training video set as the training video set, and returning to the step of extracting the first training video from the training video set until the removed training video set is an empty set; Summarizing the video similarities to obtain a video similarity set.

[0033] It should be explained that the first training video refers to the training video ranked first in the training video set. The video feature matrix refers to the matrix representing the video features of the first training video, which includes luminance features and color features. The removed training video set refers to the training video set after removing the first training video. The comparison training video refers to the removed training video in the removed training video set. Since this comparison training video is used for subsequent comparison with the first training video, it is named the comparison training video. The comparison feature matrix refers to the matrix representing the video features of the comparison training video.

[0034] Specifically, constructing the video feature matrix of the first training video includes: Extracting a set of single-frame pictures from the first training video according to the preset number of pictures; Performing feature extraction on each single-frame image in the set of single-frame pictures to obtain a set of picture feature vectors. Among them, the feature extraction includes: color feature extraction, and the picture feature vectors in the set of picture feature vectors correspond one by one to the single-frame pictures in the set of single-frame pictures; Construct a video feature matrix based on the set of image feature vectors.

[0035] It is understandable that the number of images refers to a constant set by humans, which represents the number of single-frame images in the single-frame image set. The single-frame image set refers to the set of images of each frame number in the first training video. Exemplarily, if a first training video is 90 seconds in total and the set number of images is 10, then an image is extracted from the first training video every 9 seconds, thus obtaining the single-frame image set. The image feature vector refers to a vector representing the color features in a single-frame image. Among them, the color features are extracted as follows: convert the single-frame image from the RGB color space to the HSV color space, which contains three channels, namely: the hue channel (H), the saturation channel (S), and the value channel (V). Perform block processing on each channel in each HSV color space, divide the image into several sub-regions, calculate the color histogram of each sub-region, where the hue channel uses 36 bins, and the saturation and value channels each use 8 bins, and splice the histogram features of each sub-region in sequence to form a multi-dimensional color feature vector.

[0036] It should be explained that the video feature matrix is expressed as: Among them, represents the video feature matrix, represents the first image feature vector in the set of image feature vectors, represents the vector dimension of the image feature vector, and respectively represent the first vector element in the first image feature vector and the th vector element, represents the th image feature vector in the set of image feature vectors, represents the number of image feature vectors in the set of image feature vectors, and respectively represent the first vector element in the th image feature vector and the th vector element.

[0037] S3. Obtain the fitness value set of the training video set using a preset questionnaire, and perform key-value pairing on each playback feature vector in the playback feature vector set according to the fitness value set to obtain the fitness key-value group set.

[0038] It is understandable that the questionnaire refers to a questionnaire containing multiple questions. In order to quantify the impact of different brightness control strategies (the brightness values of each area to be controlled in the device to be controlled) on the viewing experience of viewers, multiple questions with numerical options are set in this questionnaire, which include multiple aspects. For example, since different brightness strategies will present different clarity under different video contents. For instance, under the same brightness performance, when the video picture is a tunnel, the viewing clarity of the viewer is relatively low, while when the video picture is a blue sky, the viewing clarity of the viewer is relatively high. Therefore, questions about viewing clarity need to be set. Similarly, different brightness will also affect the eye comfort of viewers, and questions about eye comfort can also be set.

[0039] Exemplarily, the questionnaire is as follows: Question 1: How do you think the details of the current picture are presented? A. Very clear, B. Relatively clear, C. Average, D. Not clear; Question 2: Do you think the current brightness is dazzling? A. Very soft, B. Relatively soft, C. Slightly dazzling, D. Very dazzling. Corresponding to A, B, C, and D, numerical values 1, 2, 3, and 4 are assigned respectively.

[0040] It is understandable that the fitness value refers to a numerical value that quantifies the visual perception during the playback of the training video. The higher the fitness value, the better the visual perception. Since the visual perception during playback is affected by the screen brightness, and under the same brightness, different screen colors will also produce different visual perceptions, it is necessary to introduce the subsequent particle swarm optimization algorithm to find out the brightness adjustment strategy that can have the best visual perception under the same color conditions. Therefore, in the subsequent particle swarm optimization algorithm, the position of the particle represents the playback brightness of multiple areas to be controlled, and the fitness value of the particle is the fitness value here.

[0041] Furthermore, each playback feature vector corresponds to a training video, and then the fitness value corresponding to the training video is identified, and the playback feature vector and the fitness value are paired by key value to obtain a fitness key value group, which is expressed as (U: P), where U represents the playback feature vector and P represents the fitness value.

[0042] Specifically, the method of obtaining the fitness value set of the training video set by using the preset questionnaire includes: Sequentially extract the completed playback videos in the training video set; Based on the completed playback videos, identify multiple video viewers, and send the questionnaire to multiple video viewers to obtain multiple viewing experience data; Perform mean calculation based on the multiple viewing experience data to obtain comprehensive experience data, and calculate the fitness value according to the comprehensive experience data; Summarize the fitness values corresponding to each completed playback video to obtain the fitness value set.

[0043] It should be explained that the completed playback video refers to the training video that has been completed in the training video set. The video viewer refers to the experimenter who has watched the completed playback video. The viewing experience data refers to the data obtained after filling out the questionnaire. Each viewing experience data corresponds to a video viewer. The acquisition method of the viewing experience data is as follows: After a certain video viewer fills out the questionnaire, the options corresponding to each question in the filled questionnaire are A, A, B, and D respectively. Among them, the values corresponding to these options are 1, 1, 2, and 4 respectively. Then the viewing experience data is expressed as (1, 1, 2, 4). The comprehensive experience data refers to the experience data obtained after mean calculation. Among them, the mean calculation based on multiple viewing experience data is as follows: Multiple viewing experience data are respectively (1, 1, 2, 4), (1, 2, 2, 4), and (2, 2, 3, 3). Then the mean calculation of these viewing experience data is: (1 + 1 + 2) / 3 = 1.3, (1 + 2 + 2) / 3 = 1.7, (2 + 2 + 3) / 3 = 2.3, (4 + 4 + 3) / 3 = 3.7. Then the comprehensive experience data is (1.3, 1.7, 2.3, 3.7). The calculation of the fitness value according to the comprehensive experience data is: Add up each data in the comprehensive experience data to obtain the fitness value.

[0044] S4. Sequentially extract the training fitness key value groups from the fitness key value group set, and determine the training feature vectors in the training fitness key value groups, where the training feature vectors include: brightness training vectors and color training vectors.

[0045] It can be understood that the training fitness key value group refers to the fitness key value group in the fitness key value group set. The training feature vector refers to the playback feature vector in the training fitness key value group. The brightness training vector and the color training vector respectively represent the playback brightness vector and the playback color vector in the training feature vector.

[0046] S5. Initialize the pre - constructed particle swarm based on the training feature vectors and the video similarity set to obtain the initial particle swarm, where the particle swarm includes multiple particles.

[0047] It can be understood that the particle swarm refers to the set of candidate solutions used to search for the optimal solution in the particle swarm optimization algorithm (PSO). Each particle represents a possible brightness control strategy (that is, the brightness combination of multiple regions to be controlled), and its position and speed are dynamically adjusted during the iteration process to approach the optimal solution. The initial particle swarm refers to the particle swarm after initialization.

[0048] Specifically, the initialization of the pre - constructed particle swarm based on the training feature vectors and the video similarity set to obtain the initial particle swarm includes: Remove the training fitness key value groups from the fitness key value group set to obtain the comparison fitness key value group set; Extract the comparison fitness key-value groups from the set of comparison fitness key-value groups in sequence, and determine the comparison parameter groups in the comparison fitness key-value groups. Among them, the comparison parameter groups include: a brightness comparison vector, a color comparison vector, and a comparison fitness value; Extract particles from the particle swarm in sequence and generate the initial velocity of the particles; Generate an initial query vector based on the brightness training vector in the training feature vector, and denote the initial query vector as the initial position; Determine the fitness value weight and the comparison similarity of the comparison fitness key-value groups according to the initial query vector; Summarize the fitness value weights, the comparison similarities, and the comparison fitness values of each comparison fitness key-value group in the set of comparison fitness key-value groups to obtain a fitness value weight set, a comparison similarity set, and a comparison fitness value set; Calculate the initial fitness value according to the fitness value weight set, the comparison similarity set, and the comparison fitness value set; Generate initial particles based on the initial fitness value, the initial position, and the initial velocity, and summarize the initial particles to obtain an initial particle swarm.

[0049] It should be explained that the set of comparison fitness key-value groups refers to the set of fitness key-value groups after removing the training fitness key-value groups. The comparison parameter group refers to a combination including a brightness comparison vector, a color comparison vector, and a comparison fitness value. Among them, the brightness comparison vector and the color comparison vector respectively refer to the playback brightness vector and the playback color vector included in the playback feature vector in the comparison fitness key-value group. The comparison fitness value refers to the fitness value in the comparison fitness key-value group. The initial query vector refers to a vector obtained by performing a random function change on the basis of the brightness training vector. The random function change refers to adding Gaussian noise to each dimension of the brightness training vector. Among them, the Gaussian noise is expressed as: , where represents a preset noise intensity coefficient, which is set manually, represents a normal distribution random number with a mean of 0 and a variance of . The fitness weight refers to the similarity between the keys (brightness comparison vector and color comparison vector) in the comparison fitness key-value group and the initial query vector and the color training vector. The comparison similarity refers to the video similarity between the training video corresponding to the comparison fitness key-value group and the training video corresponding to the training fitness key-value group. The initial velocity refers to the initial moving direction and rate of the particle in the solution space, and the generation method is: randomly generate within a preset velocity range.

[0050] Furthermore, since it is impossible to obtain the fitness value of each particle in each iterative update through a questionnaire in real-time particle swarm iteration, an attention mechanism is introduced here. Among them, all fitness values in the comparison fitness key-value group are used as values, and all brightness comparison vectors and color comparison vectors in the comparison fitness key-value group are used as keys. The attention weight of each value is obtained through the initial query vector, color training vector, brightness comparison vector, and color comparison vector. Considering the influence of different training videos on the fitness value, a comparison similarity is introduced into this attention weight. Through the attention weight corresponding to each comparison fitness key-value group and the corresponding value (comparison fitness), the fitness value of the particle can be obtained (the acquisition method can refer to the calculation method of the initial fitness).

[0051] It is understandable that based on the initial fitness value, initial position, and initial velocity, generating an initial particle means that the particle position, particle velocity, and particle fitness value of the initial particle are the initial position, initial velocity, and initial fitness value respectively.

[0052] Specifically, the determining of the fitness value weight and comparison similarity of the comparison fitness key-value group according to the initial query vector includes:[[]] Calculating the fitness value weight according to the brightness comparison vector, initial query vector, color training vector, and color comparison vector; Respectively determining the main training video and comparison training video corresponding to the training fitness key-value group and the comparison fitness key-value group, and identifying the comparison similarity between the main training video and the comparison training video in the video similarity set.

[0053] It is understandable that the main training video and the comparison training video respectively refer to the training videos corresponding to the training fitness key-value group and the comparison fitness key-value group.

[0054] Specifically, the calculating of the fitness value weight according to the brightness comparison vector, initial query vector, color training vector, and color comparison vector includes:[[]] Calculating the fitness value weight using the following formula:[[]] where,[[]] represents the fitness value weight,[[]] represents the activation function,[[]] represents the brightness comparison vector,[[]] represents the initial query vector,[[]] represents the vector dimension of the brightness comparison vector,[[]] represents the color comparison vector,[[]] represents the color training vector,[[]] represents the vector dimension of the color comparison vector.

[0055] Specifically, the calculation of the initial fitness value includes: According to the fitness value weight set, the comparison similarity set, and the comparison fitness value set, calculate the initial fitness value using the following formula: Wherein, represents the initial fitness value, represents the number of comparison similarities in the comparison similarity set, represents the th comparison similarity in the comparison similarity set, represents the th fitness value weight in the fitness value weight set, represents the th comparison fitness value in the comparison fitness value set.

[0056] S6. Iterate the initial particle swarm to obtain the optimal position, and determine the optimal brightness vector based on the optimal position, and summarize the optimal brightness vectors to obtain the optimal brightness vector set.

[0057] It can be understood that the optimal position refers to the position corresponding to the particle with the highest fitness value during the particle swarm optimization iteration process, and the brightness vector corresponding to this position can maximize the visual experience of the viewer (i.e., the highest fitness value). The optimal brightness vector refers to the vector corresponding to the optimal position.

[0058] Furthermore, the optimal position is represented as the optimal brightness vector, and the optimal brightness vector contains multiple optimal brightnesses. Among them, the multiple optimal brightnesses represent the optimal brightness adjustment strategies of the corresponding training videos at different times.

[0059] S7. Based on the training video set, classify the optimal brightness vector set to obtain the same-video brightness vector group set, and play the training video set according to the same-video brightness vector group set to obtain the adjusted fitness key value group set.

[0060] It can be understood that since the optimal brightness vector set contains the optimal brightness vectors of all training videos (according to the content recorded in S2 of this embodiment, a playback brightness vector corresponds to a start playback moment of a training video, so the optimal brightness vector also corresponds to a certain start playback moment of a certain training video), it is necessary to classify the optimal brightness vectors according to the training video set. Multiple optimal brightness vectors corresponding to the same training video are recorded as the same-video brightness vector group. When playing the training video, the corresponding same-video brightness vector group will be used to control the brightness during the playback process of the training video. The adjusted fitness key value group set refers to the fitness key value group set obtained after playing according to the same-video brightness vector group set.

[0061] Exemplarily, a certain same-video brightness vector group is as follows: At time A, the same-video brightness vector is (600 nits, 650 nits, 700 nits), where 600 nits, 650 nits, and 700 nits respectively represent the brightness of multiple areas to be controlled in the device to be controlled at time A. At time B, the same-video brightness vector is (600 nits, 600 nits, 700 nits). Then, when playing the training video corresponding to this same-video brightness vector group, the brightness will be adjusted according to the brightness corresponding to the time.

[0062] S8. Use the adjusted fitness key value group set as the fitness key value group set, and return to the step of obtaining the fitness value set of the training video set using the preset questionnaire, until each fitness value in the fitness value set is greater than the preset fitness threshold, and obtain the optimal fitness key value group set. Based on the optimal fitness key value group set, complete the control of the device to be controlled.

[0063] It is understandable that the fitness threshold refers to a constant set by humans. When the fitness value is greater than this fitness threshold, it means that the current brightness regulation strategy can bring a better viewing experience.

[0064] Furthermore, the optimal fitness key value group set refers to the fitness key value group set obtained in the last return step. In the actual playback process, the played video will not be the training video set. Denote the played video as the actual playback video. To ensure the visual perception of the actual playback video at different times, the following operations are performed using the optimal fitness key value group set: At a certain sampling time, obtain the actual color vector of the actual playback video, and obtain the optimal brightness vector for the next time through particle swarm optimization. Among them, the optimal fitness key value group set is used to generate the fitness values of each particle in the particle swarm. The way of generating this fitness value is the same as the above step of calculating the initial fitness value according to the fitness value weight set, the comparison similarity set, and the comparison fitness value set, where the optimal fitness key value group set corresponds to the comparison fitness key value group set.

[0065] To solve the problems described in the background art, the present invention first plays a training video set and obtains a set of playback feature vectors and a set of video similarities, which provides feature data support based on the actual playback scenario for subsequent brightness optimization, helps to more accurately analyze the performance of different video contents under different brightness conditions, and provides a data basis for brightness optimization. Then, it collects the subjective viewing experiences of users through a questionnaire, quantifies them into fitness values, and associates the fitness values with the playback feature vectors, which can combine the subjective feelings of users with the objective brightness and color characteristics, provide feedback basis at the user perception level for subsequent optimization algorithms, and make the brightness optimization more in line with the actual viewing needs of users. Further, based on the training feature vectors and the set of video similarities, the particle swarm is initialized to obtain an initial particle swarm. By combining the training feature vectors and the set of video similarities to initialize the particle swarm, the distribution of the initial particle swarm can be made more targeted and reasonable, which helps to improve the efficiency and effect of the particle swarm optimization algorithm. Then, the initial particle swarm is iterated to obtain the optimal position, and the optimal brightness vector is determined based on the optimal position. The optimal brightness vectors are summarized to obtain a set of optimal brightness vectors. Through this iterative optimization process, the brightness control strategy can be continuously adjusted to gradually approach the optimal brightness vector, thereby realizing brightness optimization based on the user's viewing experience. Finally, by using the adjusted fitness key value group set as the fitness key value group set to complete the cyclic iterative optimization and obtaining the optimal fitness key value group set, through this cyclic iterative optimization, the brightness control strategy can be continuously adjusted and improved until the fitness threshold is reached, thereby ensuring that the final brightness control scheme can meet the requirements of the user's viewing experience. Therefore, the present invention can improve the accuracy of brightness regulation of Mini LED displays and enhance the user's viewing experience.

[0066] As Figure 2 shown, it is a functional module diagram of a device control system for Mini LED displays provided by an embodiment of the present invention.

[0067] The device control system 100 for Mini LED displays described in the present invention can be installed in an electronic device. According to the functions achieved, the device control system 100 for Mini LED displays can include a playback vector extraction module 101, a fitness survey module 102, an optimal brightness determination module 103, and a video playback return module 104. The modules described in the present invention can also be referred to as units, which refer to a series of computer program segments that can be executed by a processor of an electronic device and can complete fixed functions, and are stored in the memory of the electronic device.

[0068] The playback vector extraction module 101 is configured to receive a display control instruction, determine a device to be controlled based on the display control instruction, where the device to be controlled includes a plurality of areas to be controlled, use the device to be controlled to play a pre-acquired training video set, obtain a playback feature vector set, and calculate a video similarity set of the training video set. The playback feature vectors in the playback feature vector set all include: a playback brightness vector and a playback color vector; The fitness survey module 102 is configured to obtain a fitness value set of the training video set by using a preset questionnaire, perform key-value pairing on each playback feature vector in the playback feature vector set according to the fitness value set to obtain a fitness key-value group set, sequentially extract training fitness key-value groups in the fitness key-value group set, and determine the training feature vectors in the training fitness key-value groups. The training feature vectors include: a brightness training vector and a color training vector; The optimal brightness determination module 103 is configured to initialize a pre-constructed particle swarm based on the training feature vectors and the video similarity set to obtain an initial particle swarm, where the particle swarm includes a plurality of particles, iterate the initial particle swarm to obtain an optimal position, and determine an optimal brightness vector based on the optimal position, and summarize the optimal brightness vectors to obtain an optimal brightness vector set; The video playback return module 104 is configured to classify the optimal brightness vector set based on the training video set to obtain a same-video brightness vector group set, play the training video set according to the same-video brightness vector group set to obtain an adjusted fitness key-value group set, use the adjusted fitness key-value group set as the fitness key-value group set, and return to the step of obtaining the fitness value set of the training video set by using the preset questionnaire until each fitness value in the fitness value set is greater than a preset fitness threshold, and obtain an optimal fitness key-value group set.

[0069] Specifically, each module in the device control system 100 for Mini LED display in the embodiments of the present invention uses the same technical means as those Figure 1 described in the device control method for Mini LED display described above, and can produce the same technical effects, which will not be elaborated here.

[0070] As Figure 3 shown, it is a schematic structural diagram of an electronic device for implementing the device control method for Mini LED display provided by an embodiment of the present invention.

[0071] The electronic device 1 may include a processor 10, a memory 11, and a bus 12, and may further include a computer program stored in the memory 11 and executable on the processor 10, such as a device control method program for Mini LED display.

[0072] Among them, the memory 11 includes at least one type of readable storage medium, and the readable storage medium includes flash memory, mobile hard disk, multimedia card, card-type memory (such as SD or DX memory, etc.), magnetic memory, magnetic disk, optical disc, etc. In some embodiments, the memory 11 may be an internal storage unit of the electronic device 1, such as the mobile hard disk of the electronic device 1. In some other embodiments, the memory 11 may also be an external storage device of the electronic device 1, such as a plug-in mobile hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, etc. equipped on the electronic device 1. Further, the memory 11 also includes the internal storage unit of the electronic device 1 and also includes an external storage device. The memory 11 can not only be used to store application software installed in the electronic device 1 and various types of data, such as the code of the device control method program for Mini LED display, etc., but also can be used to temporarily store data that has been output or will be output.

[0073] In some embodiments, the processor 10 may be composed of integrated circuits. For example, it may be composed of a single packaged integrated circuit, or may be composed of multiple integrated circuits with the same or different functions, including a combination of one or more Central Processing Units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips, etc. The processor 10 is the control core (Control Unit) of the electronic device, connecting various components of the entire electronic device through various interfaces and lines, and by running or executing programs or modules stored in the memory 11 (such as the device control method program for Mini LED display, etc.), and calling data stored in the memory 11, to perform various functions of the electronic device 1 and process data.

[0074] The bus 12 may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This bus 12 can be divided into an address bus, a data bus, a control bus, etc. The bus 12 is set to realize the connection and communication between the memory 11 and at least one processor 10, etc.

[0075] Figure 3 Only the electronic device with components is shown. Those skilled in the art can understand that, Figure 3The structure shown does not constitute a limitation on the electronic device 1, and it may include fewer or more components than shown, or combine certain components, or have different component arrangements.

[0076] For example, although not shown, the electronic device 1 may further include a power source (such as a battery) for powering each component. Preferably, the power source can be logically connected to the at least one processor 10 through a power management system, so as to implement functions such as charging management, discharging management, and power consumption management through the power management system. The power source may also include any components such as one or more DC or AC power sources, a recharge system, a power failure detection circuit, a power converter or inverter, a power status indicator, etc. The electronic device 1 may also include a variety of sensors, a Bluetooth module, a Wi-Fi module, etc., which will not be elaborated here.

[0077] Furthermore, the electronic device 1 may further include a network interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a WI-FI interface, a Bluetooth interface, etc.), which is generally used to establish a communication connection between the electronic device 1 and other electronic devices.

[0078] Optionally, the electronic device 1 may further include a user interface. The user interface may be a display (Display), an input unit (such as a keyboard (Keyboard)). Optionally, the user interface may also be a standard wired interface or a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch liquid crystal display, and an OLED (Organic Light-Emitting Diode) toucher, etc. Among them, the display may also be appropriately referred to as a display screen or a display unit, which is used to display the information processed in the electronic device 1 and to display a visual user interface.

[0079] The device control method program for Mini LED display stored in the memory 11 of the electronic device 1 is a combination of multiple instructions. When running in the processor 10, it can implement: Receiving a display control instruction, and determining a device to be controlled based on the display control instruction, where the device to be controlled includes multiple areas to be controlled; Using the device to be controlled to play a pre-acquired training video set, obtaining a set of playback feature vectors, and calculating a set of video similarity of the training video set, where each playback feature vector in the set of playback feature vectors includes: a playback brightness vector and a playback color vector; Using a preset questionnaire to obtain a set of fitness values of the training video set, and according to the set of fitness values, performing key-value pairing on each playback feature vector in the set of playback feature vectors to obtain a set of fitness key-value groups; Successively extract the training fitness key-value group set from the fitness key-value group set, and determine the training feature vectors in the training fitness key-value group, where the training feature vectors include: brightness training vectors and color training vectors; Initialize the pre-constructed particle swarm based on the training feature vectors and the video similarity set to obtain an initial particle swarm, where the particle swarm includes multiple particles; Iterate the initial particle swarm to obtain the optimal position, and determine the optimal brightness vector based on the optimal position, and summarize the optimal brightness vectors to obtain the optimal brightness vector set; Classify the optimal brightness vector set based on the training video set to obtain the same-video brightness vector group set, and play the training video set according to the same-video brightness vector group set to obtain the adjusted fitness key-value group set; Use the adjusted fitness key-value group set as the fitness key-value group set, and return to the step of obtaining the fitness value set of the training video set by using the preset questionnaire, until each fitness value in the fitness value set is greater than the preset fitness threshold, and obtain the optimal fitness key-value group set, and complete the control of the device to be controlled based on the optimal fitness key-value group set.

[0080] Specifically, the specific implementation method of the above instructions by the processor 10 can refer to Figures 1 to 3 the description of the relevant steps in the corresponding embodiment, which will not be elaborated here.

[0081] Furthermore, if the modules / units integrated in the electronic device 1 are implemented in the form of software function units and sold or used as independent products, they can be stored in a computer-readable storage medium. The computer-readable storage medium can be volatile or non-volatile. For example, the computer-readable medium can include: any entity or system capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disc, computer memory, read-only memory (ROM, Read-Only Memory).

[0082] The present invention also provides a computer-readable storage medium, where the readable storage medium stores a computer program, and when the computer program is executed by a processor of an electronic device, it can implement: Receive a display control instruction, and determine the device to be controlled based on the display control instruction, where the device to be controlled includes multiple areas to be controlled; Use the device to be controlled to play the pre-obtained training video set to obtain a set of playback feature vectors, and calculate the video similarity set of the training video set, where the playback feature vectors in the set of playback feature vectors all include: playback brightness vectors and playback color vectors; Obtain the fitness value set of the training video set using a preset questionnaire. According to the fitness value set, perform key-value pairing on each playback feature vector in the playback feature vector set to obtain a fitness key-value group set; Extract training fitness key-value groups from the fitness key-value group set in sequence, and determine the training feature vectors in the training fitness key-value groups. Among them, the training feature vectors include: brightness training vectors and color training vectors; Initialize a pre-constructed particle swarm based on the training feature vectors and the video similarity set to obtain an initial particle swarm, where the particle swarm includes multiple particles; Iterate the initial particle swarm to obtain the optimal position, and determine the optimal brightness vector based on the optimal position. Aggregate the optimal brightness vectors to obtain an optimal brightness vector set; Classify the optimal brightness vector set based on the training video set to obtain a same-video brightness vector group set. According to the same-video brightness vector group set, play the training video set to obtain an adjusted fitness key-value group set; Take the adjusted fitness key-value group set as the fitness key-value group set, and return to the step of obtaining the fitness value set of the training video set using the preset questionnaire until each fitness value in the fitness value set is greater than a preset fitness threshold, and obtain an optimal fitness key-value group set. Complete the control of the device to be controlled based on the optimal fitness key-value group set.

[0083] In several embodiments provided by the present invention, it should be understood that the disclosed devices, systems, and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative, and there can be other partitioning methods in actual implementation.

[0084] The modules described as separate components may or may not be physically separated. The components shown as modules may or may not be physical units, that is, they can be located in one place, or they can be distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0085] In addition, the functional modules in each embodiment of the present invention can be integrated into a processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above integrated unit can be implemented in the form of hardware, or in the form of hardware plus software functional modules.

[0086] For those skilled in the art, it is obvious that the present invention is not limited to the details of the above exemplary embodiments, and can be implemented in other specific forms without departing from the spirit or basic characteristics of the present invention.

[0087] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A device control method for Mini LED display, characterized in that, The method includes: Receiving a display control instruction, and determining a device to be controlled based on the display control instruction, where the device to be controlled includes multiple areas to be controlled; Using the device to be controlled to play a pre-acquired training video set to obtain a set of playback feature vectors, and calculating a set of video similarity degrees of the training video set, where each playback feature vector in the set of playback feature vectors includes: a playback brightness vector and a playback color vector; Obtaining a set of fitness values of the training video set by using a preset questionnaire, and performing key-value pairing on each playback feature vector in the set of playback feature vectors according to the set of fitness values to obtain a set of fitness key-value groups; Sequentially extracting training fitness key-value groups from the set of fitness key-value groups, and determining training feature vectors in the training fitness key-value groups, where the training feature vectors include: a brightness training vector and a color training vector; Initializing a pre-constructed particle swarm based on the training feature vectors and the set of video similarity degrees to obtain an initial particle swarm, where the particle swarm includes multiple particles; Iterating the initial particle swarm to obtain an optimal position, and determining an optimal brightness vector based on the optimal position, and summarizing the optimal brightness vectors to obtain a set of optimal brightness vectors; Classifying the set of optimal brightness vectors based on the training video set to obtain a set of brightness vector groups of the same video, and playing the training video set according to the set of brightness vector groups of the same video to obtain a set of adjusted fitness key-value groups; Taking the set of adjusted fitness key-value groups as the set of fitness key-value groups, and returning to the step of obtaining the set of fitness values of the training video set by using the preset questionnaire until each fitness value in the set of fitness values is greater than a preset fitness threshold, and obtaining a set of optimal fitness key-value groups, and completing the control of the device to be controlled based on the set of optimal fitness key-value groups.

2. The device control method for Mini LED display according to claim 1, wherein The step of using the device to be controlled to play a pre-acquired training video set to obtain a set of playback feature vectors includes: Performing the following operations on each training video in the training video set: Setting a start playback time, and at the start playback time, using multiple areas to be controlled in the device to be controlled to play the training video; Sampling each area to be controlled in the playback step according to the start playback time to obtain multiple playback brightness values and multiple playback color values, where the playback brightness values and the playback color values correspond to the areas to be controlled one by one; Obtaining a playback feature vector according to the multiple playback brightness values and the multiple playback color values, where the playback feature vector includes: a playback brightness vector and a playback color vector; Calculating an intermediate sampling time according to a preset sampling interval and the start playback time; Taking the intermediate sampling time as the start playback time, and returning to the step of sampling each area to be controlled in the playback step according to the start playback time until the intermediate sampling time is not less than a preset end playback time; Summarizing the playback feature vectors to obtain a group of playback feature vectors, and combining the groups of playback feature vectors corresponding to each training video to obtain a set of playback feature vectors.

3. The device control method for Mini LED display according to claim 2, characterized in that, The step of calculating the set of video similarity degrees of the training video set includes: Extract the first training video from the training video set and construct a video feature matrix for the first training video, where the video feature matrix includes color features; Remove the first training video from the training video set to obtain a removed training video set, and sequentially extract comparison training videos from the removed training video set; Construct a comparison feature matrix for the comparison training video, and calculate the video similarity based on the video feature matrix and the comparison feature matrix; Use the removed training video set as the training video set, and return to the step of extracting the first training video from the training video set until the removed training video set is an empty set; Summarize the video similarities to obtain a video similarity set.

4. The device control method for Mini LED display according to claim 3, wherein, The construction of the video feature matrix for the first training video includes: Extract a set of single-frame pictures from the first training video according to a preset number of pictures; Perform feature extraction on each single-frame image in the set of single-frame pictures to obtain a set of picture feature vectors, where the feature extraction includes: color feature extraction, and the picture feature vectors in the set of picture feature vectors correspond one-to-one with the single-frame pictures in the set of single-frame pictures; Construct a video feature matrix according to the set of picture feature vectors.

5. The device control method for Mini LED display according to claim 4, wherein, The obtaining of the fitness value set of the training video set by using a preset questionnaire includes: Sequentially extract the completed playback videos from the training video set; Based on the completed playback videos, identify multiple video viewers, and send the questionnaire to the multiple video viewers to obtain multiple viewing experience data; Perform mean calculation based on the multiple viewing experience data to obtain comprehensive experience data, and calculate the fitness value according to the comprehensive experience data; Summarize the fitness values corresponding to each completed playback video to obtain a fitness value set.

6. The device control method for Mini LED display according to claim 5, wherein The initialization of the pre-constructed particle swarm based on the training feature vector and the video similarity set to obtain an initial particle swarm includes: Remove the training fitness key-value groups from the set of fitness key-value groups to obtain a set of comparison fitness key-value groups; Sequentially extract comparison fitness key-value groups from the set of comparison fitness key-value groups, and determine the comparison parameter groups in the comparison fitness key-value groups, where the comparison parameter groups include: brightness comparison vectors, color comparison vectors, and comparison fitness values; Sequentially extract particles in the particle swarm to generate initial velocities for the particles; Generate an initial query vector based on the brightness training vector in the training feature vector, and record the initial query vector as the initial position; Determine the fitness value weight and the comparison similarity of the comparison fitness key-value group according to the initial query vector; Summarize the fitness value weights, comparison similarities, and comparison fitness values of each comparison fitness key-value group in the set of comparison fitness key-value groups to obtain a fitness value weight set, a comparison similarity set, and a comparison fitness value set; Calculate the initial fitness value according to the fitness value weight set, the comparison similarity set, and the comparison fitness value set; Generate initial particles based on the initial fitness value, the initial position, and the initial velocity, and summarize the initial particles to obtain an initial particle swarm.

7. The device control method for Mini LED display according to claim 6, wherein, The determination of the fitness value weight and the comparison similarity of the comparison fitness key-value group according to the initial query vector includes: Calculate the fitness value weight according to the brightness comparison vector, the initial query vector, the color training vector, and the color comparison vector; Respectively determine the main training video and the comparison training video corresponding to the training fitness key value group and the comparison fitness key value group, and identify the comparison similarity between the main training video and the comparison training video in the video similarity set.

8. The device control method for Mini LED display according to claim 7, wherein, The calculating of the fitness value weight according to the brightness comparison vector, the initial query vector, the color training vector and the color comparison vector includes: Calculate the fitness value weight using the following formula: Among them, represents the fitness value weight, represents the activation function, represents the brightness contrast vector, represents the initial query vector, represents the vector dimension of the brightness contrast vector, represents the color contrast vector, represents the color training vector, represents the vector dimension of the color contrast vector.

9. The device control method for Mini LED display according to claim 8, wherein, The calculating of the initial fitness value includes: According to the fitness value weight set, the comparison similarity set and the comparison fitness value set, calculate the initial fitness value using the following formula: Among them, represents the initial fitness value, represents the number of comparison similarities in the comparison similarity set, represents the th comparison similarity in the comparison similarity set, represents the th fitness value weight in the fitness value weight set, represents the th comparison fitness value in the comparison fitness value set.

10. A device control system for Mini LED display, characterized in that, The system includes: A playback vector extraction module, configured to receive a display control instruction, determine a device to be controlled based on the display control instruction, where the device to be controlled includes a plurality of areas to be controlled, and use the device to be controlled to play a pre-acquired training video set to obtain a playback feature vector set, and calculate a video similarity set of the training video set, where each playback feature vector in the playback feature vector set includes: a playback brightness vector and a playback color vector; A fitness survey module, configured to obtain a fitness value set of the training video set by using a preset questionnaire, perform key value pairing on each playback feature vector in the playback feature vector set according to the fitness value set to obtain a fitness key value group set, sequentially extract a training fitness key value group from the fitness key value group set, and determine a training feature vector in the training fitness key value group, where the training feature vector includes: a brightness training vector and a color training vector; An optimal brightness determination module, configured to initialize a pre-constructed particle swarm based on the training feature vector and the video similarity set to obtain an initial particle swarm, where the particle swarm includes a plurality of particles, iterate the initial particle swarm to obtain an optimal position, and determine an optimal brightness vector based on the optimal position, and summarize the optimal brightness vectors to obtain an optimal brightness vector set; A video playback return module, configured to classify the optimal brightness vector set based on the training video set to obtain a same-video brightness vector group set, play the training video set according to the same-video brightness vector group set to obtain an adjusted fitness key value group set, use the adjusted fitness key value group set as the fitness key value group set, and return to the step of obtaining the fitness value set of the training video set by using the preset questionnaire until each fitness value in the fitness value set is greater than a preset fitness threshold, and obtain an optimal fitness key value group set.

Images