LED display screen debugging method and system

Abstract

The invention relates to the technical field of display, in particular to an LED display screen debugging method and system, and the method comprises the steps: obtaining first calibration data representing the physical display characteristics of each pixel region of an LED display screen, the first calibration data describing the chromaticity and brightness attenuation conditions caused by long-time operation; extracting a color characteristic abstract of the input source, wherein the color characteristic abstract comprises a dominant hue range, a color space parameter and a brightness distribution characteristic; according to the color characteristic abstract, corresponding second calibration data are obtained from a preset calibration strategy library, and the calibration strategy library records a calibration scheme which preferentially guarantees the key color display accuracy for different color characteristics; and applying the first calibration data and the second calibration data to an input signal of the input source to complete automatic color debugging of the display effect. The method has the advantages that the color of the display effect of the LED display screen is automatically debugged, and the debugging efficiency and accuracy are improved.

Description

Technical Field

[0001] The present application relates to the field of display technology, and in particular to a method and system for debugging an LED display screen. Background Art

[0002] To create a unique brand atmosphere and visual impact, many stores have installed curved LED screens in their core areas. These screens' primary function is to periodically display carefully crafted seasonal advertising videos, live footage from international fashion weeks, and artistically variably designed visuals custom-made for the flagship store. The sources of this regularly displayed visual information vary. For example, advertising videos are often produced by professional film and television production companies, utilizing film-grade recording equipment and post-production color grading processes, resulting in images with a specific artistic style and color palette. Fashion show footage, on the other hand, may be recorded live from multiple cameras. Due to variations in lighting and lens parameters, even after post-processing, the output images may exhibit subtle variations in basic tones and brightness ranges. Furthermore, the content displayed on LED screens is not static; new video clips or images are frequently added on an ad hoc basis. These ad hoc images often require a short production cycle, and the color profiles, encoding formats, and color management processes of the original footage may differ significantly from those of standard source material. Some sources may even be provided without rigorous color calibration.

[0003] LED displays are typically operated at high intensity for extended periods of time. LED modules in different areas of the screen may have different heat dissipation conditions during installation. Furthermore, prolonged continuous operation can cause LED light-emitting units in some locations to experience faster luminous efficiency decay or color coordinate drift than in others. This uneven aging gradually creates areas of color variation on the screen that are not immediately noticeable to the naked eye but can be measured using instruments. When displaying large areas of solid color or gradient backgrounds, these variations can disrupt the overall integrity and subtlety of the image. Existing calibration solutions may not be able to effectively address the issue of uneven localized screen aging, nor can they be dynamically optimized based on the specific color characteristics of the input content.

[0004] In view of the above problems, the existing technology is in urgent need of improvement. Summary of the Invention

[0005] The purpose of this application is to provide an LED display screen debugging method and system, which has the advantages of automated color debugging of the LED display screen display effect, improving debugging efficiency and accuracy.

[0006] This application provides a method for debugging an LED display screen. The technical solution is as follows: Methods include: Acquire first calibration data representing physical display characteristics of each pixel area of the LED display screen, wherein the first calibration data describes chromaticity and brightness attenuation caused by long-term operation; Extracting a color characteristic summary of the input source, the color characteristic summary includes the main hue range, color space parameters, and brightness distribution characteristics; According to the color characteristic summary, corresponding second calibration data is obtained from a preset calibration strategy library, which records calibration schemes that prioritize key color display accuracy for different color characteristics; The first calibration data and the second calibration data are applied to the input signal of the input source to complete the automatic color adjustment of the display effect.

[0007] Through the above solution, automatic color debugging of the display effect of the LED display can be achieved, and it can be adapted according to the screen aging condition and the color characteristics of the input source, thereby improving the debugging efficiency and accuracy.

[0008] Furthermore, the present application also proposes obtaining first calibration data characterizing the physical display characteristics of each pixel area of the LED display screen, including: Display a standardized test pattern on each pixel area of the LED display; Obtaining brightness and chromaticity measurements of each pixel region displaying a standardized test pattern; The brightness and chromaticity measurement values are compared with preset standard values to obtain the physical attenuation characteristics and brightness attenuation coefficient of each pixel area, and are stored as first calibration data.

[0009] Through the above solution, a specific method for obtaining the physical attenuation characteristics of the screen is provided, making the first calibration data more accurate and reliable.

[0010] Furthermore, the present application also proposes extracting a color characteristic summary of an input source, including: Perform pixel distribution analysis on the key frame image of the input source, and extract the color space parameters of the key frame image in combination with the metadata information of the input source; The main hue range and brightness distribution characteristics are calculated based on the color space parameters; The dominant hue range and brightness distribution characteristics are stored in a structured manner as a color feature summary.

[0011] Through the above scheme, a specific method for extracting the color characteristic summary of the input source is provided, which provides a basis for the subsequent calibration strategy selection.

[0012] Furthermore, the present application also proposes to complete automatic color adjustment of the display effect based on applying the first calibration data and the second calibration data to the input signal of the input source, including: Applying preset basic calibration data to the input signal of the input source to generate a basic picture; Identifying a target pixel corresponding to a preset key color from the base image, and acquiring corresponding second calibration data from a calibration strategy library based on a color characteristic summary of the target pixel; Calculating calibration information of the target pixel based on the first calibration data and the second calibration data; The pixels corresponding to the target pixel positions in the base image are replaced with pixels corresponding to the calibration information, completing the automatic color debugging of the display effect.

[0013] Through the above scheme, detailed steps for automatic color debugging based on two calibration data are provided, especially for the processing of key colors.

[0014] Furthermore, the present application also proposes that the target pixel corresponding to the preset key color is identified from the basic image, including: Based on the basic image, candidate color pixels whose color values are within the preset key color range are screened; Generate context features of the candidate color pixels according to the spatial aggregation or temporal persistence of the candidate color pixels in the continuous frames of the input source; Comparing the context feature with a preset key color determination rule to obtain a comparison result, wherein the key color determination rule defines a spatial or temporal threshold for distinguishing the preset key color from adjacent colors; The candidate color pixels whose comparison results meet the key color determination rule are determined as target pixels.

[0015] Through the above scheme, a robust method for identifying target pixels corresponding to key colors is provided, which combines contextual features such as spatial aggregation and temporal persistence.

[0016] Furthermore, the present application also proposes comparing the context features with the preset key color determination rules to obtain a comparison result, including: Obtaining dynamic characteristic indicators that characterize the degree of change between consecutive frames; Adjust the spatial or temporal threshold in the key color determination rule according to the dynamic characteristic index; The context features are compared with the adjusted key color determination rules to obtain a comparison result.

[0017] Through the above solution, the key color determination rule can be adaptively adjusted according to the dynamic characteristics of the picture, thereby improving the accuracy of key color recognition.

[0018] Furthermore, the present application also proposes obtaining a dynamic characteristic index that characterizes the degree of change between consecutive frames in the input source, including: Divide each frame of the input source into multiple picture blocks; Calculate the content change of each picture block between consecutive frames and generate the corresponding block change index; According to the change index of each block, the screen blocks with similar block change indexes are classified into the same dynamic characteristic partition; A partition dynamic index is determined for each dynamic characteristic partition, and the partition dynamic index is combined as the dynamic characteristic index.

[0019] The above solution provides a specific method for obtaining the dynamic characteristic index of the picture, which can quantify the dynamic changes of different areas.

[0020] Furthermore, the present application also proposes to classify picture blocks with similar block change indicators into the same dynamic characteristic partition according to each block change indicator, including: Counting block change indices of all picture blocks corresponding to the current frame in the continuous frame pictures, and obtaining the numerical distribution of the block change indices; According to the numerical distribution, determine the clustering center of indicators representing different dynamic levels; According to the proximity between each block change index and the cluster center of each index, the image blocks are classified into corresponding dynamic characteristic partitions.

[0021] Through the above solution, a specific method for performing dynamic characteristic partitioning according to block change indicators is provided, which can identify areas with different dynamic levels in the picture.

[0022] Furthermore, the present application also proposes to determine the index aggregation center representing different dynamic levels based on the numerical distribution, including: Analyze the distribution of each value in a loop, identify multiple local clusters based on the local density characteristics of the value distribution and use them as the aggregation center of the basic indicators. The identification of local density characteristics includes calculating the central tendency of the local neighborhood of each block change indicator, and adjusting the contribution weight of each block change indicator in the local density feature identification based on the degree of deviation from the central tendency calculation; When the value of the basic indicator aggregation center meets the preset stability condition, the basic indicator aggregation center is used as the final indicator aggregation center.

[0023] Through the above scheme, a specific method for determining the aggregation center of dynamic grade indicators is provided, making the dynamic characteristic partitioning more accurate.

[0024] Furthermore, the present application also proposes an LED display debugging system, comprising: An acquisition module is used to acquire first calibration data representing the physical display characteristics of each pixel area of the LED display screen, where the first calibration data describes the chromaticity and brightness attenuation caused by long-term operation; A parameter extraction module is used to extract the color characteristic summary of the input source, the color characteristic summary includes the main color range, color space parameters and brightness distribution characteristics; A calibration strategy module is used to obtain corresponding second calibration data from a preset calibration strategy library based on the color characteristic summary. The calibration strategy library records calibration schemes that prioritize key color display accuracy for different color characteristics; The debugging module is used to apply the first calibration data and the second calibration data to the input signal of the input source to complete the automatic color debugging of the display effect.

[0025] Through the above solution, a system for implementing the above debugging method is provided, which is convenient for practical application.

[0026] From the above, it can be seen that the LED display debugging method and system provided by the present application solves the problem in the prior art that it is difficult to perform content adaptation and key color priority automatic color debugging on special-shaped aging screens within a limited time by obtaining screen aging data, extracting the color characteristics of the input source and combining the calibration strategy library for automatic debugging. It has the advantages of being able to realize automatic color debugging of the display effect of the LED display, and can adapt according to the screen aging conditions and the color characteristics of the input source, thereby improving the debugging efficiency and accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 A flowchart of a method for debugging an LED display screen according to one embodiment of the present application is provided.

[0028] Figure 2 This is one of the flow charts of a LED display debugging method provided in another embodiment of the present application.

[0029] Figure 3 This is a second flow chart of a method for debugging an LED display screen provided in another embodiment of the present application.

[0030] Figure 4 This is a third flow chart of a method for debugging an LED display screen provided in another embodiment of the present application.

[0031] Figure 5 This is a fourth flow chart of a method for debugging an LED display screen provided in another embodiment of the present application.

[0032] Figure 6 This is a fifth flow chart of a method for debugging an LED display screen provided in another embodiment of the present application.

[0033] Figure 7 This is a sixth flow chart of a method for debugging an LED display screen provided in another embodiment of the present application.

[0034] Figure 8 This is a seventh flow chart of a method for debugging an LED display screen provided in another embodiment of the present application.

[0035] Figure 9 This is an eighth flow chart of a method for debugging an LED display screen provided in another embodiment of the present application.

[0036] Figure 10 This is a flowchart of an LED display debugging system provided in another embodiment of the present application.

[0037] In the figure: 1. Acquisition module; 2. Parameter extraction module; 3. Calibration strategy module; 4. Debugging module. DETAILED DESCRIPTION

[0038] The technical solutions in this application will be clearly and completely described below in conjunction with the drawings in this application. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all of the embodiments. The components of the present application generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application for which protection is claimed, but merely represents selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work are within the scope of protection of this application.

[0039] It should be noted that similar reference numerals and letters represent similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined or explained in subsequent drawings. At the same time, in the description of this application, the terms "first", "second", etc. are only used to distinguish the description and should not be understood as indicating or implying relative importance.

[0040] Traditional debugging methods for curved LED displays struggle to accurately compensate for uneven color and brightness degradation caused by long-term operation. Furthermore, existing methods struggle to adapt quickly to frequent switching between input sources with widely varying color characteristics, resulting in subpar display quality. Manual or semi-automated debugging processes are inefficient, especially within limited maintenance timeframes, making it difficult to ensure accurate display of key brand colors. Automated debugging is also challenging to implement to accommodate rapid content updates and reduce operator skill requirements.

[0041] Reference Figure 1 , this application proposes a LED display debugging method, the method comprising: S1000: Acquire first calibration data representing physical display characteristics of each pixel area of the LED display screen, where the first calibration data describes chromaticity and brightness attenuation caused by long-term operation; S2000: Extracting a color characteristic summary of the input source, which includes the main hue range, color space parameters, and brightness distribution characteristics; S3000: According to the color characteristic summary, corresponding second calibration data is obtained from a preset calibration strategy library, where the calibration strategy library records calibration schemes that prioritize key color display accuracy for different color characteristics; S4000: Applying the first calibration data and the second calibration data to the input signal of the input source to complete automatic color adjustment of the display effect.

[0042] In this embodiment, the first calibration data refers to data characterizing the physical display characteristics of each pixel area of the LED display screen. This data can be obtained by obtaining the brightness and chromaticity measurement values of each pixel area through a measuring device and comparing them with preset standard values. It is mainly used to describe the attenuation of chromaticity and brightness caused by long-term operation. The color characteristic summary refers to a summary description of the color characteristics of the input source. This data can be obtained by performing pixel distribution analysis on the key frame images of the input source and extracting color space parameters in combination with metadata information. It is mainly used to identify the color style and key color information of the input source. The calibration strategy library refers to a preset database containing multiple calibration schemes. This data can be implemented by storing lookup tables (LUTs) or parameter sets corresponding to different color characteristics. This data is mainly used to provide targeted calibration schemes based on the color characteristics of the input source, prioritizing the accuracy of key color display. The second calibration data refers to calibration data corresponding to the color characteristic summary of the input source, obtained from the calibration strategy library. This data can be implemented by directly reading the lookup table or parameter set matching the color characteristic summary in the calibration strategy library. This data is mainly used to adjust the display effect, especially the display of key colors, according to the content of the input source.

[0043] The solution of the present application obtains the physical attenuation characteristic data of the display screen itself, and on this basis, analyzes the color characteristics of the current input source, and selects the calibration solution that best suits the current content from the preset strategy library according to the analysis results. Subsequently, the first calibration data representing the physical state of the screen and the second calibration data representing the content adaptation strategy are organically combined and applied to the input signal of the input source. It is precisely because of this calibration method that combines the state of the screen itself and the characteristics of the content that the final display effect can simultaneously compensate for the uneven aging of the screen and optimize the color performance of the currently playing content in a targeted manner, especially ensuring the accurate restoration of the brand's key colors, thereby solving the problem of efficient and accurate color debugging of special-shaped and aging display screens in complex environments. The entire process does not require a lot of manual intervention and realizes automated operation.

[0044] In some preferred embodiments, first, a measurement unit scans and measures the LED display screen to obtain brightness and chromaticity attenuation data for different areas of the screen and generate first calibration data. Simultaneously, a content analysis unit receives the input source signal, analyzes the key frames of the video stream, extracts its main color range, color space parameters, and brightness distribution characteristics, and generates a color characteristic summary. A strategy selection unit searches for a matching calibration solution in a pre-built calibration strategy library based on the color characteristic summary and obtains the corresponding second calibration data. Finally, a calibration processing unit takes the input source signal, the first calibration data, and the second calibration data as input, and processes the input signal in real time, such as adjusting the color and brightness values of the pixels through a lookup table or matrix operation, to generate a calibrated output signal to drive the LED display screen. This process can be triggered periodically or automatically when the input source is switched, without the need for manual intervention in the measurement and adjustment process.

[0045] Reference Figure 2 In another embodiment of the present application, step S1000 includes: S1100: Displaying a standardized test pattern in each pixel area of the LED display; S1200: Obtaining brightness and chromaticity measurement values of each pixel region displaying a standardized test pattern; S1300: Compare the brightness and chromaticity measurement values with preset standard values to obtain the physical attenuation characteristics and brightness attenuation coefficient of each pixel area, and store them as first calibration data.

[0046] In this embodiment, the standardized test pattern refers to an image or sequence with a known, stable, and uniform color and brightness distribution, such as a solid color, grayscale, checkerboard, etc., with the purpose of providing a controllable and consistent display benchmark for accurate measurement and comparison. The preset standard value refers to the brightness and chromaticity reference values measured under ideal conditions, or provided by the manufacturer, or determined according to industry standards, used to measure the performance of the display screen. The purpose is to provide an objective comparison benchmark to quantify the deviation between the actual measurement value and the ideal state. The physical attenuation characteristics and brightness attenuation coefficient refer to quantitative indicators calculated by comparing the actual measurement value with the preset standard value, which characterize the degree of brightness reduction and chromaticity drift of each pixel area of the LED display screen due to long-term use. The brightness attenuation coefficient is one of the specific quantitative parameters, and its purpose is to accurately describe the aging state of each pixel area and provide a basis for subsequent calibration.

[0047] The solution of the present application ensures the uniformity of the measurement environment and the measurement object by displaying a standardized test pattern in each pixel area of the LED display screen, and eliminates the measurement error caused by the difference in display content. Then, the brightness and chromaticity measurement values of the standardized test pattern displayed in each pixel area are obtained, and objective data reflecting the actual display effect of the pixel area is obtained. Subsequently, the brightness and chromaticity measurement values are compared with the preset standard values, the physical attenuation characteristics and brightness attenuation coefficient of each pixel area are calculated, and these data are stored as the first calibration data. By comparing with the standard values, the influence of individual differences of the LED display screen at the time of leaving the factory and factors such as ambient light on the measurement results can be effectively eliminated, thereby more accurately reflecting the attenuation of the LED display screen due to long-term operation. This accurate first calibration data, combined with the second calibration data obtained in the subsequent steps and applied to the input signal, can more effectively compensate for the aging and unevenness of the display screen, especially in terms of ensuring the accuracy of key color display, providing a reliable data basis, so that the overall color debugging effect is improved.

[0048] In some preferred embodiments, the specific implementation is as follows: First, standardized test patterns of full white, full red, full green, full blue and 16 levels of gray are displayed on the LED display screen in sequence. A high-precision luminance and colorimeter is used to measure the brightness and chromaticity of each pixel area of the display screen, and the corresponding measurement values are recorded. For example, for the full white pattern, the brightness value and CIE color coordinate value of each pixel area are recorded. Then, these measurement values are compared with the standard measurement values of the display screen of this model when it leaves the factory. For example, the ratio of the actual brightness of each pixel area to the standard brightness is calculated as the brightness attenuation coefficient. The color difference between the actual color coordinates and the standard color coordinates of each pixel area is calculated as the chromaticity attenuation characteristic. Finally, these calculated brightness attenuation coefficients and chromaticity attenuation characteristic data are stored in the form of a two-dimensional array as the first calibration data, where the index of the array corresponds to the pixel area position of the display screen.

[0049] Reference Figure 3 In another embodiment of the present application, step S3000 includes: S3100: Analyze pixel distribution of the key frame image of the input source, and extract color space parameters of the key frame image in combination with metadata information of the input source; S3200: Calculate the main color range and brightness distribution characteristics based on color space parameters; S3300: Store the main hue range and brightness distribution characteristics as color characteristic summaries in a structured manner.

[0050] Among them, in this embodiment, the key frame image refers to the image frame that can represent the main content and color characteristics of the input source, which can be selected by analyzing the content change rate of the video stream or a fixed time interval, and its purpose is to efficiently capture the overall color characteristics with a small number of image samples; pixel distribution analysis refers to the statistics and analysis of the color or brightness values of the pixels in the image, which can be specifically through constructing a color histogram, a brightness histogram or performing cluster analysis, and its purpose is to understand the distribution of different colors and brightness in the image; the metadata information of the input source refers to additional information related to the input source file, which can specifically include encoding format, color profile, gamma curve information, etc., and its purpose is to provide prior knowledge about the color characteristics of the input source to assist in more accurate understanding and processing of pixel data; color space parameters refer to the description of the image color space. The main color range refers to the most representative or concentrated color area in the input source image, which can be determined by analyzing the pixel distribution and identifying the high-density area, and its purpose is to summarize the main color tendency of the input source; the brightness distribution feature refers to the overall distribution of pixel brightness in the input source image, which can be described by calculating the statistics of the brightness histogram, and its purpose is to reflect the light and dark levels and contrast of the image; storage in a structured manner means organizing and saving the extracted color characteristic summary information in a predetermined data format, which can be done using JSON, XML or other custom data structures, and its purpose is to facilitate the subsequent system to read, parse and use this information.

[0051] The solution in this application first selects representative keyframe images and, combined with metadata from the input source, performs detailed pixel-level analysis on them to extract accurate color space parameters. Based on these parameters, the system further calculates the dominant color range that summarizes the overall color tendency of the input source, as well as the brightness distribution characteristics that reflect the gradation of light and dark. Finally, this key color information is stored in a structured format that is easily processed and recognized by machines, forming a color characteristic summary. This process ensures that the extracted summary information is based not only on the visual content of the image itself but also takes into account the original color encoding standard, thereby more comprehensively and accurately reflecting the true color characteristics of the input source. Applying this accurately extracted color characteristic summary to the subsequent calibration strategy selection enables the system to intelligently match the most appropriate calibration scheme based on the specific color style of the input source, for example, prioritizing the display accuracy of key colors while also considering the overall color balance of the image. Compared to relying solely on preset calibration or manual judgment, this significantly improves the intelligence and effectiveness of automated debugging, effectively solving the problem of large differences in color characteristics of different input sources and the difficulty of unified adaptation, and provides a solid foundation for accurate color rendering on LED displays.

[0052] In some preferred embodiments, a keyframe image is first selected from the input source video stream every five seconds or upon scene change. Pixel value statistics are then performed on the selected keyframe image, for example, by calculating the histogram of each channel in the RGB color space and analyzing the pixel distribution in color spaces such as HSV or Lab. Simultaneously, the metadata of the input source file is read to obtain its color profile information, for example, identifying that the video uses the Rec.709 color standard. Based on the pixel analysis results and Rec.709 color space parameters, a clustering algorithm is used to identify the color clusters with the highest pixel concentration and determine the dominant color range. For example, it is determined that the image primarily contains blue and green, and their dominant color ranges correspond to the color intervals of sky blue and grass green, respectively. Simultaneously, a luminance histogram of the keyframe image is calculated, and its mean, standard deviation, and median are calculated as luminance distribution features. Finally, the extracted dominant color range and luminance distribution features are organized into a JSON object and stored as a color feature summary file.

[0053] Reference Figure 4 In another embodiment of the present application, step S4000 includes: S4100: applying preset basic calibration data to an input signal of an input source to generate a basic image; S4200: Identifying a target pixel corresponding to a preset key color from the basic image, and acquiring corresponding second calibration data from a calibration strategy library based on the color characteristic summary of the target pixel; S4300: Calculating calibration information of a target pixel based on the first calibration data and the second calibration data; S4400: replacing pixels corresponding to the target pixel positions in the basic image with pixels corresponding to the calibration information, thereby completing automatic color adjustment of the display effect.

[0054] Among them, in this embodiment, the preset basic calibration data refers to the preliminary and general color calibration data of the display screen before fine calibration for key colors is performed. It can be implemented by factory calibration data, general calibration data generated during regular maintenance, or calibration data based on a certain standard color space. Its purpose is to provide a relatively consistent and standardized picture basis for subsequent key color identification and fine calibration; the basic picture refers to the picture obtained after applying the preset basic calibration data to the input signal of the input source, which can be represented in the form of a digital image frame or a pixel matrix; the preset key color refers to a specific color or color range whose display accuracy needs to be prioritized in the display content. It can be implemented by a specific chromaticity coordinate (such as Lab, Luv) range, RGB value range or color range defined based on a brand standard color card. Its purpose is to focus calibration resources and ensure that the core visual elements are accurately displayed. Accurate restoration; target pixels refer to a set of pixels in the basic image that are identified as corresponding to a preset key color, which can be represented in the form of a pixel coordinate list, a binary mask image or a pixel index set, with the purpose of accurately locking the area that requires fine calibration; color characteristic summary refers to a general description of the color attributes of the target pixels, which can be implemented using statistical information such as the average chromaticity, brightness, saturation, color distribution histogram of the target pixels or texture features of the key color area, with the purpose of providing a basis for selecting a suitable calibration strategy from the calibration strategy library; calibration information refers to parameters or data calculated based on the first calibration data and the second calibration data for adjusting the color of the target pixels, which can be represented in the form of a new RGB value of the target pixels, a lookup table or a color conversion matrix applied to the target pixels, with the purpose of providing specific instructions for accurate color compensation of the target pixels.

[0055] The solution of this application applies preset base calibration data to the input signal from the source to generate a preliminarily adjusted base image, providing a standardized starting point for subsequent fine-tuning. Next, the base image identifies target pixels corresponding to preset key colors. Based on the color characteristic summary of these target pixels, targeted secondary calibration data is retrieved from a calibration strategy library. This process eliminates the need for blindly applying calibration to the entire image, allowing precise targeting of key color areas requiring optimization. Subsequently, based on the first calibration data representing the physical attenuation characteristics of the screen and the calibration strategy (second calibration data) for the key colors, the calibration information required to compensate for these target pixels is calculated. This combination considers the physical state of the screen and the characteristics of the key colors in the content, resulting in more accurate calculated calibration information. Finally, the pixels corresponding to the target pixels in the base image are replaced with those corresponding to the calculated calibration information, completing automated color tuning for the display. By replacing only the pixels in the key color area, the accuracy of the key color display is ensured while minimizing the impact on the colors in other areas of the image, maintaining visual harmony and consistency with the key color across the entire image. This targeted local calibration method overcomes the problems of color distortion in non-critical color areas or overall picture disharmony that may result from directly applying calibration data to the entire input signal. It can also make full use of the acquired first and second calibration data, refining their application from the overall level to the local critical color area, thereby improving the visual quality of the overall picture while ensuring the accurate display of critical colors.

[0056] In some preferred embodiments, assume that the input source is playing a video containing a brand logo (a specific red color). First, the system applies preset factory calibration data to the video input signal to generate a base image. Within this base image, the system identifies pixels in the red region where the brand logo resides; these pixels are designated as target pixels. The system analyzes the color characteristics of these red target pixels, for example, by calculating their average chromaticity coordinates, and discovers that their current red color exhibits a slight orange tint. Based on this "orange tint" color characteristic summary, the system searches a calibration strategy library for a corresponding calibration strategy, such as obtaining a lookup table (second calibration data) to compensate for the orange tint. Simultaneously, the system obtains physical attenuation data (first calibration data) for the screen pixels corresponding to this red region, e.g., an 8% attenuation of the brightness of the red sub-pixels in this region. Based on this 8% attenuation data and the lookup table used to compensate for the orange tint, the system calculates calibration information for the final adjustment of these red target pixels, such as by calculating new RGB values or generating a local lookup table to apply to these pixels. Finally, the pixels in the brand logo region of the base image are replaced with the new pixel values derived from this calibration information, thereby completing the precise color calibration of the brand logo region, displaying the standard brand red.

[0057] Reference Figure 5 In another embodiment of the present application, it is further proposed that the sub-step of step S4200: identifying the target pixel corresponding to the preset key color from the basic image includes: S4210: Based on the basic image, screening candidate color pixels whose color values are within a preset key color range; S4220: Generate context features of the candidate color pixels according to spatial concentration or temporal persistence of the candidate color pixels in consecutive frames of the input source; S4230: Comparing the context feature with a preset key color determination rule to obtain a comparison result, where the key color determination rule defines a spatial or temporal threshold for distinguishing the preset key color from adjacent colors; S4240: Determine the candidate color pixel whose comparison result satisfies the key color determination rule as the target pixel.

[0058] In this embodiment, a candidate color pixel refers to a pixel whose color value falls within a preset key color range, which can be achieved by defining a color range in a specific color space (such as HSL, Lab or RGB) (for example, by using center color coordinates and tolerance values); a context feature refers to information reflecting the distribution or change characteristics of the candidate color pixel in a surrounding spatial region or a continuous time series, which can be generated by calculating the number and density of pixels of the same or similar color in the neighborhood of the candidate color pixel, or by tracking the stability of the position, color or existence state of the candidate color pixel in multiple consecutive frames; a key color determination rule refers to a set of standards or conditions for determining whether a candidate color pixel truly belongs to a preset key color, which defines a spatial or temporal threshold for distinguishing the preset key color from adjacent colors. For example, a spatial threshold can be set to require that there must be a sufficient number of pixels of the same color within a certain range around the candidate color pixel, or a temporal threshold can be set to require that the candidate color pixel exist stably in multiple consecutive frames.

[0059] The solution of this application first screens candidate color pixels whose color values fall within a preset key color range based on the underlying image, thereby preliminarily identifying potential target areas and reducing the amount of data required for subsequent processing. Furthermore, contextual features are generated for the candidate color pixels based on their spatial clustering or temporal persistence across consecutive frames of the input source. The introduction of these contextual features eliminates the reliance on the color value of a single pixel and instead considers the temporal and spatial correlations of the pixels. Spatial clustering reflects whether the candidate color pixels are isolated or part of a larger, homochromatic region; temporal persistence reflects whether the candidate color pixels are stable within the dynamic image, rather than representing transient noise or flicker. By combining these spatial and temporal information, the true attributes of the candidate color pixels can be more accurately determined. The generated contextual features are then compared with preset key color determination rules. These key color determination rules define spatial or temporal thresholds for distinguishing the preset key color from adjacent colors. These thresholds are determined based on an analysis of key color characteristics and common interference patterns. By comparing with these rules, pixels whose color values are close to the key color but are spatially isolated or temporally unstable, such as noise points or rapidly changing background elements, can be effectively filtered out. Ultimately, only candidate color pixels whose comparison results meet the key color judgment rules will be determined as the true target pixels. This method of preliminary screening based on color values and then fine-tuning the judgment in combination with spatial and temporal context can significantly improve the accuracy of target pixel identification and avoid misjudging non-key color pixels as target pixels. This accurate target pixel identification provides a reliable foundation for the subsequent steps of calculating calibration information and replacing pixels based on the first calibration data and the second calibration data, ensuring that the calibration operation is only applied to the key color areas that really need calibration, thereby improving the accuracy and effect of the overall color debugging, especially when processing complex pictures containing interference colors or noise close to the key color. Its advantages are more obvious.

[0060] In some preferred embodiments, specifically, a pixel traversal of the base image can be performed. For each pixel, its color value is checked to see if it falls within a preset key color range (e.g., a specific HSL range for a brand logo color). Pixels that meet the criteria are marked as candidate color pixels. Then, for each candidate color pixel, its spatial concentration can be calculated. For example, this can be done by checking how many pixels within a 3x3 pixel neighborhood have color values that differ from the candidate color pixel by less than a preset threshold, or by calculating the density of pixels of the same color within the neighborhood. Simultaneously, the position and color stability of the candidate color pixel in the last five consecutive frames can be tracked, and its temporal persistence characteristics can be calculated. For example, the number of frames in which the candidate color pixel remains in a similar position with minimal color change can be counted. These spatial concentration and temporal persistence values are combined to form a contextual feature for the candidate color pixel. This contextual feature is then compared with preset key color determination rules. For example, the key color determination rule can be set as follows: the spatial concentration value must be greater than a threshold (for example, there must be at least 5 similar pixels in the neighborhood), and the temporal persistence value must be greater than another threshold (for example, it must appear in at least 4 out of 5 consecutive frames). Only candidate color pixels that meet these spatial and temporal thresholds will be finally determined as target pixels that need to be calibrated.

[0061] Reference Figure 6 In another embodiment of the present application, step S4230 further includes: S4231: Obtaining a dynamic characteristic index representing the degree of change between consecutive frames; S4232: Adjusting the spatial or temporal threshold in the key color determination rule according to the dynamic characteristic index; S4233: Compare the context feature with the adjusted key color determination rule to obtain a comparison result.

[0062] Among them, in this embodiment, the dynamic characteristic index refers to a numerical value or set that quantifies the degree of content change between consecutive frames, which can be achieved by calculating the pixel difference between frames, motion vector analysis or scene switching detection, and its purpose is to provide a quantitative basis for the dynamic degree of the picture for the subsequent adjustment of the key color determination rule; the spatial or temporal threshold refers to the critical value used to measure the spatial aggregation degree or temporal continuity of the candidate color pixels in the key color determination rule, which can be represented by a specific pixel distance, number of pixels or number of frames, and its purpose is to serve as a specific standard for determining the key color; the comparison result refers to the judgment conclusion obtained after comparing the context feature with the key color determination rule, which can be represented by a Boolean value (yes / no) or a confidence value.

[0063] The solution of this application obtains a dynamic characteristic index that characterizes the degree of change between consecutive frames, adjusts the spatial or temporal thresholds in the key color determination rule based on this dynamic characteristic index, and then compares the contextual features with the adjusted key color determination rule to obtain a comparison result. This solves the problem of misjudgment when using fixed key color determination rules due to the dramatic changes in dynamic images. It is precisely because the determination threshold is adaptively adjusted according to the dynamic degree of the image that the key color determination rule can better adapt to image content with different dynamic characteristics. For example, when the image changes rapidly, appropriately relaxing the spatial or temporal threshold can prevent the misjudgment of fast-moving key color objects as non-key color. When the image changes slowly, appropriately tightening the spatial or temporal threshold can improve the recognition accuracy of static key color areas and reduce interference from dynamic backgrounds. This adaptive adjustment mechanism makes the key color recognition process more robust and accurate, especially when processing video content with complex motion or scene changes. Combining this solution with the step of identifying target pixels corresponding to preset key colors in the basic solution, this method improves the accuracy of target pixel recognition through more accurate key color determination, thereby improving the accuracy and effectiveness of subsequent color adjustment.

[0064] In some preferred embodiments, a dynamic characteristic index representing the degree of change between consecutive frames can be obtained by analyzing pixel changes, motion information, or scene switching frequency between consecutive frames to obtain a quantitative value. For example, the average pixel difference, optical flow information, or motion vectors between two or more consecutive frames can be calculated and integrated to generate a dynamic characteristic index. A higher value for this index indicates more dramatic image changes. Based on this dynamic characteristic index, the spatial or temporal thresholds in the key color determination rule can be dynamically adjusted. For example, a threshold adjustment function can be pre-defined, with the dynamic characteristic index as input and the adjustment coefficients for the spatial and temporal thresholds as output. When the dynamic characteristic index is low, the adjustment coefficients are close to 1, and the thresholds do not change much. When the dynamic characteristic index is high, the adjustment coefficients are greater than 1, and the spatial and temporal thresholds are increased accordingly. Specifically, if the dynamic characteristic index indicates significant image motion, the spatial aggregation range required for key color determination can be appropriately expanded, or the temporal continuity requirement can be appropriately reduced. Conversely, if the image is essentially static, the spatial aggregation range can be narrowed and the temporal continuity requirement increased. The adjusted key color determination rule is then used to compare with the contextual features of candidate color pixels. For example, if the adjusted spatial threshold allows for greater spatial dispersion, then even if a key color pixel is surrounded by a small number of non-key color pixels, it may still be identified as a key color as long as its spatial concentration meets the adjusted lower limit. Similarly, if the adjusted temporal threshold allows for a shorter duration, then even if a key color object only appears briefly in the image, it may still be identified as a key color as long as its duration meets the adjusted lower limit. In this way, the key color determination process can be optimized based on the actual dynamic situation of the image, improving recognition accuracy.

[0065] Reference Figure 7 In another embodiment of the present application, step S4231 includes: S42311: Divide each frame of the input source into multiple image blocks; S42312: Calculate the content change of each image block between consecutive frames and generate a corresponding block change index; S42313: Classifying image blocks with similar block change indices into the same dynamic characteristic partition based on the block change indices; S42314: Determine a partition dynamic index for each dynamic characteristic partition, and combine the partition dynamic index as a dynamic characteristic index.

[0066] Among them, in this embodiment, the picture block refers to a plurality of smaller, independent areas into which each frame of the picture is divided, and its purpose is to be able to capture the dynamic changes of different areas in the picture more finely; the content change amount refers to a numerical value that quantifies the degree of content difference of a picture block between consecutive frames, which can be achieved by calculating pixel value differences, motion vector amplitudes, etc.; the block change index refers to a numerical value obtained after quantifying the content change amount of the picture block, and its purpose is to provide a data basis for subsequent dynamic characteristic partitioning; the dynamic characteristic partitioning refers to a collection formed by classifying picture blocks with similar block change indicators, and its purpose is to stratify the dynamic characteristics in the picture; the partition dynamic index refers to a summary numerical value of the overall dynamic degree of a dynamic characteristic partition, which can be achieved by calculating the average value or weighted average value of all block change indicators in the partition, and its purpose is to reflect the overall dynamic characteristics of each partition.

[0067] The solution of this application divides each frame of the input source into multiple image blocks, enabling detailed analysis of dynamic changes in different image regions. Next, the content change between consecutive frames for each block is calculated, generating corresponding block change indices that quantify the degree of change in each region. Based on these block change indices, image blocks with similar dynamic levels are grouped into the same dynamic characteristic partition, achieving a stratified image dynamic feature. Subsequently, a partition dynamic index is determined for each dynamic characteristic partition, summarizing the overall dynamic level of each partition. These partition dynamic indices are combined to form an overall dynamic characteristic index. This partitioned, stratified approach to obtaining dynamic characteristic indices more accurately reflects both local and overall image dynamic characteristics than simple global indices. This more accurate dynamic characteristic index is used to adjust the spatial or temporal thresholds in the key color determination rules, enabling the key color recognition process to more effectively adapt to dynamic image changes. For example, for areas with drastic dynamic changes, the determination threshold can be appropriately adjusted to reduce false positives or missed positives; for relatively static areas, a stricter threshold can be used to improve recognition accuracy. This rule adjustment based on fine-grained dynamic analysis improves the accuracy of key color recognition in complex dynamic scenes, thereby enhancing the overall color debugging effect.

[0068] In some preferred embodiments, each frame of the input source can be divided into a fixed-size grid, for example, by evenly dividing the width and height of the frame into a number of units to form frame blocks. When calculating the content change of each frame block between consecutive frames, the absolute value of the average pixel brightness difference between the current frame block and the corresponding block in the previous frame can be used as a block change index. Based on each block change index, frame blocks with similar block change indexes can be classified into different dynamic characteristic partitions using a clustering algorithm (e.g., a density-based clustering method), for example, into slowly changing partitions and rapidly changing partitions. When determining a partition dynamic index for each dynamic characteristic partition, the average value of the block change indexes for all partitions within that partition can be calculated. Ultimately, the partition dynamic indexes of all partitions can be combined to form a vector that serves as the dynamic characteristic index.

[0069] Reference Figure 8 In another embodiment of the present application, step S42313 includes: A1: Counting block change indices of all image blocks corresponding to the current frame in the continuous frames, and obtaining the numerical distribution of the block change indices; A2: Based on the numerical distribution, determine the clustering center of indicators representing different dynamic levels; A3: Classify the image blocks into corresponding dynamic characteristic partitions based on the proximity between each block change index and the cluster center of each index.

[0070] In this embodiment, the numerical distribution refers to the set of block change index values of all picture blocks corresponding to the current frame in continuous frames and their distribution on the numerical axis. It can be represented by a histogram, a probability density function curve, or a discrete numerical list, etc., with the purpose of comprehensively understanding the overall dynamic changes of various regions in the current frame. The index cluster center refers to the typical numerical point or numerical range representing different dynamic levels in the numerical distribution of the block change index. It can be determined by a clustering algorithm, a peak detection algorithm, or a method based on statistical features, with the purpose of providing a reference benchmark for classifying the dynamic characteristics of picture blocks. The degree of proximity refers to the degree of similarity or difference between the block change index value of a picture block and the value of a certain indicator cluster center. It can be quantified by Euclidean distance, Manhattan distance, cosine similarity, or a similarity calculation method based on a probability model, with the purpose of determining to which dynamic characteristic partition the picture block should be classified.

[0071] The solution of this application calculates the block change index of all image blocks in the current frame and obtains its numerical distribution, thereby comprehensively understanding the dynamic changes of the current image. Based on this numerical distribution, it further determines the cluster centers of indicators representing different dynamic levels. These cluster centers represent typical dynamic change patterns in the video content. Subsequently, by calculating the proximity between the block change index of each image block and these cluster centers, the image blocks are accurately classified into the partitions that best represent their dynamic characteristics. Compared with simple fixed threshold division, this classification method based on overall distribution and cluster centers can more flexibly adapt to the dynamic characteristics of different video content and more accurately identify different areas in the image, such as static, slow motion, and fast motion, thereby providing more refined basic data for subsequent dynamic characteristic index calculation. Precisely because of this more accurate dynamic characteristic partitioning, when subsequently adjusting the key color determination rules based on dynamic characteristic indicators, it can more accurately identify the key color pixels that need to be prioritized in different motion states, thereby improving the accuracy and adaptability of overall color adjustment.

[0072] In some preferred embodiments, the specific implementation is as follows: First, the block change indexes of all screen blocks in the current frame are counted, and a numerical histogram of these indexes is generated to visualize their distribution. Next, a K-means clustering algorithm is used to perform cluster analysis on these block change index values. For example, the number of clusters is set to 3, representing three levels of low dynamics, medium dynamics, and high dynamics, respectively. The clustering algorithm will automatically calculate the center point values of these three clusters, which are the index clustering centers representing different dynamic levels. Finally, for each screen block, the Euclidean distance between its block change index value and the three index clustering center values is calculated, and the screen block is classified into the dynamic characteristic partition represented by the nearest index clustering center.

[0073] Reference Figure 9 In another embodiment of the present application, step A2 further includes: A21: Circularly analyze the distribution of each value, identify multiple local clusters based on the local density characteristics of the value distribution, and use them as the basic indicator cluster center. Local density feature identification includes calculating the central tendency of the local neighborhood of each block change indicator, and adjusting the contribution weight of each block change indicator in the local density feature identification based on the degree of deviation from the central tendency calculation; A22: When the value of the basic indicator aggregation center meets the preset stability condition, the basic indicator aggregation center will be used as the final indicator aggregation center.

[0074] Among them, in this embodiment, the local density feature refers to the density of other numerical points in the neighborhood around a certain numerical point in the numerical distribution, which can be measured by calculating the number of numerical points in the neighborhood, the weighted sum of distances, or density estimation based on kernel functions. Its purpose is to identify areas with higher density in the data, which may correspond to different dynamic levels; local clustering areas refer to a series of continuous or close numerical ranges with higher local density features in the numerical distribution, which can be identified by density-based clustering algorithms or local peak detection methods, and its purpose is to preliminarily define the indicator range that may represent different dynamic levels; basic indicator clustering centers refer to the areas of the identified multiple local clusters. The numerical value representing the center position of each clustering area preliminarily determined in the clustering area can be determined by calculating the value corresponding to the mean, median or density peak in the local clustering area. Its purpose is to provide an initial clustering center candidate for subsequent stability judgment; the local neighborhood refers to the set of other block change index values within a certain value range with a certain block change index value as the center in the value distribution. It can be defined by a fixed radius neighborhood or a k-nearest neighbor neighborhood. Its purpose is to examine the relationship between a single block change index and its surrounding data; the central tendency calculation refers to the statistical calculation of the block change index values in the local neighborhood to obtain the typical value level of the neighborhood, which can be used to determine the relationship between the block change index and the surrounding data. This is achieved by calculating the mean, median or weighted average within the local neighborhood, with the aim of providing a reference point for measuring the degree of concentration of data within the local neighborhood; the degree of deviation refers to the difference between the value of a single block change index and the result of the calculation of the central tendency of its local neighborhood, which can be measured by calculating the absolute difference, square difference or relative difference, with the aim of quantifying the consistency of a single data point with its local environment; the contribution weight refers to the different importance coefficients assigned to the values of different block change indicators in the process of local density feature recognition, which can be adjusted in inverse proportion to the degree of deviation between the block change index and the central tendency calculation. The greater the deviation, the lower the weight, with the aim of reducing noise or outliers Impact on local density estimation; The preset stability conditions refer to a series of predetermined standards used to determine whether the basic indicator aggregation center is sufficiently representative and reliable, which may include the position change of the basic indicator aggregation center in multiple loop iterations being less than a specific threshold, the number of block change indicators contained around the basic indicator aggregation center reaching a specific proportion, or the local density of the basic indicator aggregation center reaching a specific threshold, etc. The purpose is to ensure that the final indicator aggregation center can stably represent the true dynamic level; the final indicator aggregation center refers to the indicator value that is determined to be able to accurately represent different dynamic levels after stability judgment, which can be used as the basis for subsequent dynamic characteristic partitioning of picture blocks.

[0075] The solution of this application can more comprehensively capture potential clustering patterns in the data by cyclically analyzing the numerical distribution of block change indicators. Local density feature recognition based on numerical distribution can effectively identify high-density areas in the data, which correspond to different levels of dynamics. Importantly, during the local density feature recognition process, by calculating the central tendency of the local neighborhood of each block change indicator and adjusting the contribution weight of each block change indicator in the local density feature recognition based on the degree of deviation from the central tendency calculation, this solution can effectively suppress the impact of noise data or outliers on the local density estimate, making the identification of local clustered areas more accurate. These accurately identified local clustered areas are then used as basic indicator cluster centers. To further ensure the reliability of the determined cluster centers, this solution sets preset stability conditions. Only when the value of the basic indicator cluster center meets these conditions will it be determined as the final indicator cluster center. This method, combining local density feature recognition, contribution weight adjustment, and stability assessment, can overcome the shortcomings of traditional methods that are susceptible to noise interference and accurately identify indicator cluster centers representing different levels of dynamics. Precisely because of the ability to accurately determine the indicator cluster center, the accuracy of the subsequent categorization of image blocks into corresponding dynamic characteristic partitions based on the proximity of each block's change indicator to the respective indicator cluster center is improved. Accurate dynamic characteristic partitioning makes the partition dynamic indicator determined for each dynamic characteristic partition more reliable, thereby making it more reasonable to adjust the spatial or temporal thresholds in the key color determination rules based on the partition dynamic indicator. Ultimately, this improves the accuracy of identifying target pixels corresponding to preset key colors, thereby improving the accuracy of calculating target pixel calibration information based on the first and second calibration data, ultimately achieving more accurate automated color adjustment for display effects.

[0076] In some preferred embodiments, first, a set of numerical values of the block change index obtained by statistics is obtained to form a numerical distribution. Then, the following steps are performed in a loop: based on the current numerical distribution, a density-based clustering method, such as a variant of the DBSCAN algorithm, is used to identify multiple local clustering areas. When calculating the local density, for each block change index value, its local neighborhood on the numerical axis is determined, for example, a fixed radius ε is set, and a central tendency calculation is performed on all block change index values within the local neighborhood, such as calculating their median. Next, the degree of deviation between the block change index value and the calculated median is calculated, such as calculating the absolute difference, and the contribution weight of the block change index value in the local density calculation is adjusted according to the degree of deviation, the greater the degree of deviation, the lower the contribution weight. Based on the weighted local density calculation results, areas with higher density are identified as local clustering areas. From these local clustering areas, the basic indicator aggregation center is determined, such as calculating the average or median of the values in each area. The algorithm determines whether the values of these basic indicator cluster centers meet preset stability conditions. For example, it checks whether the position change of each basic indicator cluster center between the current cycle and the previous cycle is less than a preset small threshold δ, and also checks whether the proportion of block change indicators contained around each basic indicator cluster center (for example, within a range on the numerical axis less than a certain threshold ρ) is greater than a preset minimum proportion τ. If the values of all basic indicator cluster centers meet the preset stability conditions, the current basic indicator cluster center is determined as the final indicator cluster center, and the loop ends. Otherwise, the clustering results are updated based on the current basic indicator cluster center, and the next loop is entered until the stability conditions are met or the maximum number of loops is reached.

[0077] Reference Figure 10 In another embodiment of the present application, a LED display debugging system is further proposed, the system comprising: Acquisition module 1 is used to obtain first calibration data representing the physical display characteristics of each pixel area of the LED display screen, where the first calibration data describes the chromaticity and brightness attenuation caused by long-term operation; Parameter extraction module 2, used to extract the color characteristic summary of the input source, the color characteristic summary includes the main color range, color space parameters and brightness distribution characteristics; Calibration strategy module 3, for obtaining corresponding second calibration data from a preset calibration strategy library based on the color characteristic summary, wherein the calibration strategy library records calibration schemes that prioritize key color display accuracy for different color characteristics; The debugging module 4 is used to apply the first calibration data and the second calibration data to the input signal of the input source to complete the automatic color debugging of the display effect.

[0078] Among them, the acquisition module refers to a unit for performing data acquisition functions, which can be implemented by a sensor, an interface circuit, or a data reading program; the parameter extraction module refers to a unit for analyzing input data and generating summary information, which can be implemented by an image processing algorithm, a data analysis program, or a dedicated processing chip; the calibration strategy module refers to a unit for selecting a preset strategy based on input information, which can be implemented by a database query, a rule matching algorithm, or a decision tree model; the debugging module refers to a unit for adjusting the output signal using calibration data, which can be implemented by a signal processor, a lookup table application logic, or a color conversion circuit; the first calibration data refers to a data set that characterizes the physical display state of the LED display screen, which can be implemented by an attenuation coefficient matrix, a pixel-level lookup table, or aging model parameters; the color characteristic summary refers to a summary description of the color characteristics of the input source image, which can be implemented by a color histogram, a color space parameter set, or key color statistics; the calibration strategy library refers to a data structure that stores multiple preset calibration schemes, which can be implemented by a relational database, a file system, or a set of lookup tables in memory; the second calibration data refers to data selected from the calibration strategy library for a specific calibration task, which can be implemented by a color conversion matrix, a gamma correction curve, or pixel-level adjustment parameters.

[0079] In some preferred embodiments, the present application is implemented as follows. The acquisition module can be configured as an interface unit that communicates with an external color measurement device. This interface unit periodically receives the measurement results of the brightness and chromaticity of each area of the LED display screen from the measurement device and processes and stores these results as first calibration data. The parameter extraction module can be configured as a software program running on a main control processor. This program analyzes key frames of the input video stream, extracts the main tonal range of the image through algorithms such as image segmentation and color clustering, calculates color space parameters (such as gamma values and white point coordinates), analyzes the luminance histogram, and packages this information to generate a color characteristic summary. The calibration strategy module can be configured as a database stored in system memory. This database pre-stores a set of second calibration data optimized for different color characteristics (such as high-saturation animations, low-contrast documentaries, and specific brand color advertisements). Based on the color characteristic summary output by the parameter extraction module, the calibration strategy module executes a database query or matching algorithm to obtain the corresponding second calibration data. The debugging module can be configured as a dedicated image processing chip. This chip receives the input signal and, based on the first calibration data provided by the acquisition module (for pixel-level physical compensation) and the second calibration data provided by the calibration strategy module (for content optimization and key color assurance), performs real-time color space conversion, gamma correction, brightness adjustment, and other processing on the input signal to generate the final signal that drives the LED display. The entire system can be integrated into a standalone hardware device or implemented as part of an LED display control system.

[0080] The above description is merely an embodiment of the present application and is not intended to limit the scope of protection of the present application. For those skilled in the art, various modifications and variations of the present application are possible. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

Claims

1. A LED display debugging method, characterized in that: The method comprises: Acquire first calibration data characterizing physical display characteristics of each pixel area of the LED display screen, wherein the first calibration data describes chromaticity and brightness attenuation caused by long-term operation; Extracting a color characteristic summary of an input source, wherein the color characteristic summary includes a main hue range, color space parameters, and brightness distribution characteristics; According to the color characteristic summary, corresponding second calibration data is obtained from a preset calibration strategy library, wherein the calibration strategy library records calibration schemes that prioritize key color display accuracy for different color characteristics; The first calibration data and the second calibration data are applied to the input signal of the input source to complete automatic color adjustment of the display effect.

2. The LED display debugging method according to claim 1, characterized in that: The obtaining of first calibration data characterizing physical display characteristics of each pixel area of the LED display screen includes: Displaying a standardized test pattern in each pixel area of the LED display screen; Obtaining brightness and chromaticity measurement values of each pixel region displaying the standardized test pattern; The brightness and chromaticity measurement values are compared with preset standard values to obtain the physical attenuation characteristics and brightness attenuation coefficient of each pixel area, and are stored as the first calibration data.

3. The LED display debugging method according to claim 1, characterized in that: The color characteristic summary of the extracted input source includes: Performing pixel distribution analysis on the key frame image of the input source, and extracting color space parameters of the key frame image in combination with metadata information of the input source; Calculating the main hue range and brightness distribution characteristics according to the color space parameters; The main hue range and brightness distribution characteristics are stored in a structured manner as the color characteristic summary.

4. The LED display debugging method according to claim 1, characterized in that: The step of applying the first calibration data and the second calibration data to the input signal of the input source to complete automatic color adjustment of the display effect includes: Applying preset basic calibration data to the input signal of the input source to generate a basic picture; Identifying a target pixel corresponding to a preset key color from the basic image, and acquiring corresponding second calibration data from the calibration strategy library according to the color characteristic summary of the target pixel; Calculating calibration information of the target pixel based on the first calibration data and the second calibration data; The pixels corresponding to the target pixel positions in the basic image are replaced with pixels corresponding to the calibration information, thereby completing automatic color debugging of the display effect.

5. The LED display debugging method according to claim 4, characterized in that: The identifying, from the basic picture, a target pixel corresponding to a preset key color includes: Based on the basic image, screening candidate color pixels whose color values are within the preset key color range; generating context features of the candidate color pixels according to spatial concentration or temporal persistence of the candidate color pixels in consecutive frames of the input source; Comparing the context feature with a preset key color determination rule to obtain a comparison result, wherein the key color determination rule defines a spatial or temporal threshold for distinguishing the preset key color from adjacent colors; The candidate color pixel whose comparison result satisfies the key color determination rule is determined as the target pixel.

6. The LED display debugging method according to claim 5, characterized in that: Comparing the context feature with the preset key color determination rule to obtain a comparison result includes: Obtaining a dynamic characteristic index representing a degree of change between the consecutive frames; Adjusting the spatial or temporal threshold in the key color determination rule according to the dynamic characteristic index; The context feature is compared with the adjusted key color determination rule to obtain the comparison result.

7. The LED display debugging method according to claim 6, characterized in that: The obtaining of a dynamic characteristic index representing a degree of change between the consecutive frames includes: Dividing each frame of the input source into a plurality of picture blocks; Calculating the content change of each of the picture blocks between the consecutive frames and generating a corresponding block change index; Classifying the image blocks having similar block change indicators into the same dynamic characteristic partition according to each block change indicator; A partition dynamic index is determined for each of the dynamic characteristic partitions, and the partition dynamic index is combined as the dynamic characteristic index.

8. The LED display debugging method according to claim 7, characterized in that: The classifying the picture blocks having similar block change indicators into the same dynamic characteristic partition according to the block change indicators includes: Counting the block change indexes of all the picture blocks corresponding to the current frame in the continuous frame pictures, and obtaining the value distribution of the block change indexes; Determining, based on the numerical distribution, the center of concentration of indicators representing different dynamic levels; The image blocks are classified into corresponding dynamic characteristic partitions according to the proximity between each block change indicator and the cluster center of each indicator.

9. The LED display debugging method according to claim 8, characterized in that: Determining the index aggregation center representing different dynamic levels based on the numerical distribution includes: cyclically analyzing each of the numerical distributions, identifying a plurality of local clustering areas based on local density features of the numerical distributions and using them as basic indicator clustering centers, wherein the local density feature identification includes performing a central tendency calculation on a local neighborhood of each of the block change indicators, and adjusting the contribution weight of each of the block change indicators in the local density feature identification based on the degree of deviation of the block change indicator from the central tendency calculation; When the value of the basic indicator aggregation center meets the preset stability condition, the basic indicator aggregation center is used as the final indicator aggregation center.

10. An LED display debugging system, characterized in that: The system comprises: An acquisition module is used to acquire first calibration data representing the physical display characteristics of each pixel area of the LED display screen, wherein the first calibration data describes the chromaticity and brightness attenuation caused by long-term operation; A parameter extraction module is used to extract a color characteristic summary of an input source, wherein the color characteristic summary includes a main hue range, color space parameters, and brightness distribution characteristics; a calibration strategy module, configured to obtain corresponding second calibration data from a preset calibration strategy library based on the color characteristic summary, wherein the calibration strategy library records calibration schemes that prioritize key color display accuracy for different color characteristics; The debugging module is used to apply the first calibration data and the second calibration data to the input signal of the input source to complete automatic color debugging of the display effect.

Images