LED light spot data extraction method and apparatus, and computer device

Abstract

The present application relates to an LED light spot data extraction method and apparatus, and a computer device, a storage medium and a computer program product. The method comprises: acquiring image information of an LED display device; performing grayscale processing on the image information, and performing analysis to obtain bright spot information of the image information (S204); screening the image information on the basis of each piece of bright spot information, determining a region boundary of the LED display device, and performing screening to obtain effective bright spot information within the region boundary (S206); and on the basis of the effective bright spot information, performing analysis to obtain each piece of LED light spot data within the region boundary, wherein the LED display device comprises LED light spots (S208).

Description

LED lamp point data extraction method, device and computer equipment

[0001] Related applications

[0002] This application claims priority to Chinese patent application number 2023110235534, filed on August 14, 2023, entitled “LED light point data extraction method, device and computer equipment,” the entire text of which is hereby incorporated by reference. Technical Field

[0003] The present application relates to the field of LED display technology, and in particular to a method, device, computer equipment, computer-readable storage medium, and computer program product for extracting LED light point data. Background Art

[0004] With the advancement of industrial integration, the manufacturing process of LED (light-emitting diode) display devices has become increasingly sophisticated, and various methods for calibrating LED display devices have emerged. Calibration technology is a core technology in the LED display field, primarily used to adjust the brightness of LED light points on LED display devices, which directly affects the display quality. Calibration technologies include LED light point recognition, LED light point brightness extraction, and brightness and chromaticity correction. These technologies complement each other to achieve calibration of LED display devices.

[0005] Among them, the common LED light point recognition technology adopts the big law method, which selects the threshold through the big law and then binarizes the light point image, performs light point detection calculation on the binarized image, and finally finds the light point. However, when using this method, the LED display device needs to be parallel to the imaging plane of the camera, and the LED display device must be adjusted to the correct position in the camera image (that is, the luminous edge of the LED display device is parallel to the edge of the camera imaging plane). Otherwise, calculation errors are likely to occur, and reshooting and calculation are required. The above method has harsh usage conditions in actual use and is not convenient enough.

[0006] Summary of the Invention

[0007] In a first aspect, the present application provides a method for extracting LED light point data, the method comprising:

[0008] Acquire image information of an LED display device, where the LED display device includes LED light points;

[0009] Perform grayscale processing on the image information and analyze the bright spot information of the image information;

[0010] Filtering the image information according to each bright spot information, determining the area boundary of the LED display device, and filtering to obtain valid bright spot information within the area boundary; and

[0011] Based on the effective bright spot information, the data of each LED light point within the area boundary is analyzed and obtained.

[0012] In one embodiment, the image information is filtered according to each bright spot information, the region boundary of the LED display device is determined, and the effective bright spot information within the region boundary is filtered, including:

[0013] The maximum light point area is screened based on each bright spot information and image information to obtain denoised image information;

[0014] Performing boundary detection on the denoised image information to determine the region boundary of the LED display device; and

[0015] The highlight information is filtered based on the region boundary to obtain the valid highlight information within the region boundary.

[0016] In one embodiment, the maximum light point area is screened based on each bright spot information and image information to obtain denoised image information, including:

[0017] Obtaining a maximum light point area according to basic image information in the image information; and

[0018] Based on the maximum light point area, the bright spot information is filtered to obtain the denoised image information.

[0019] In one embodiment, performing boundary detection on the denoised image information to determine the region boundary of the LED display device includes:

[0020] Based on the denoised image information, obtaining horizontal bright spot distribution statistics and vertical bright spot distribution statistics; and

[0021] Convolution calculations are performed on the horizontal and vertical bright spot distribution statistical results respectively, and the obtained left and right boundaries and upper and lower boundaries are integrated into the region boundaries.

[0022] In one embodiment, after filtering the maximum light spot area based on each bright spot information and image information to obtain denoised image information, performing boundary detection on the denoised image information before determining the area boundary of the LED display device, the method further includes:

[0023] Smoothing the bright spot information in the denoised image information to obtain smoothed denoised bright spot information;

[0024] The corrosion threshold is obtained based on the smoothed denoised bright spot information;

[0025] The smoothed denoised bright spot information is corroded according to the corrosion threshold to obtain the corroded denoised bright spot information;

[0026] Performing dilation processing based on the denoised bright spot information after corrosion to obtain dilated denoised bright spot information; and

[0027] The denoised image information is updated based on the denoised bright spot information after dilation.

[0028] In one embodiment, based on the effective bright spot information, the data of each LED light point within the area boundary is analyzed and obtained, including:

[0029] Based on the valid bright spot information, determine the empty points within the region boundary;

[0030] Sort the LED light points according to the empty point and effective bright point information within the area boundary to obtain the sequence number of each LED light point; and

[0031] Combined with the effective bright spot information and the serial number of each LED light point, the data of each LED light point within the area boundary is analyzed and obtained.

[0032] In one embodiment, determining empty points within a region boundary based on valid bright spot information includes:

[0033] Calculate the light point spacing based on the effective bright spot information; and

[0034] The effective bright spot information is analyzed according to the light point spacing and the preset deflection angle to determine the empty points within the area boundary.

[0035] In one embodiment, analyzing effective bright spot information based on the light spot spacing and the preset deflection angle to determine empty points within the region boundary includes:

[0036] According to the distance between the light points, the theoretical bright spot position is obtained; and

[0037] The theoretical bright spot position is matched with the effective bright spot information, and the theoretical bright spot position that does not match the effective bright spot information is determined as a null point.

[0038] In one embodiment, determining a theoretical bright spot position that does not match valid bright spot information as a null point includes:

[0039] Determine the search range based on the distance between the light points; and

[0040] The search is performed based on the initial point and the search range, and the theoretical bright spot position where there is no valid bright spot information is determined as a null point.

[0041] In one embodiment, grayscale processing is performed on the image information to analyze and obtain highlight information of the image information, including:

[0042] Perform grayscale processing on the image information and obtain a histogram through statistics;

[0043] Smoothing the histogram to obtain a smoothed histogram; and

[0044] The smoothed histogram is analyzed to obtain a light point threshold, and bright spots are identified based on the light point threshold to obtain bright spot information of the image information.

[0045] In one embodiment, the LED light point data includes position information of the LED light point and brightness information of the LED light point.

[0046] In a second aspect, the present application further provides a device for extracting LED light point data, the device comprising:

[0047] An image acquisition module is used to acquire image information of an LED display device, where the LED display device includes LED light points;

[0048] A bright spot analysis module is used to perform grayscale processing on image information and analyze the bright spot information of the image information;

[0049] A bright spot screening module is used to screen the image information according to each bright spot information, determine the area boundary of the LED display device, and screen out valid bright spot information within the area boundary; and

[0050] The data extraction module is used to analyze and obtain the data of each LED light point within the area boundary based on the effective bright spot information.

[0051] In a third aspect, the present application further provides a computer device. The computer device includes a memory and a processor. The memory stores a computer program. When the processor executes the computer program, the following steps are performed:

[0052] Acquire image information of an LED display device, where the LED display device includes LED light points;

[0053] Perform grayscale processing on the image information and analyze the bright spot information of the image information;

[0054] Filtering the image information according to each bright spot information, determining the area boundary of the LED display device, and filtering to obtain valid bright spot information within the area boundary; and

[0055] Based on the effective bright spot information, the data of each LED light point within the area boundary is analyzed and obtained.

[0056] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the following steps:

[0057] Acquire image information of an LED display device, where the LED display device includes LED light points;

[0058] Perform grayscale processing on the image information and analyze the bright spot information of the image information;

[0059] Filtering the image information according to each bright spot information, determining the area boundary of the LED display device, and filtering to obtain valid bright spot information within the area boundary; and

[0060] Based on the effective bright spot information, the data of each LED light point within the area boundary is analyzed and obtained.

[0061] In a fifth aspect, the present application further provides a computer program product. The computer program product includes a computer program that, when executed by a processor, implements the following steps:

[0062] Acquire image information of an LED display device, where the LED display device includes LED light points;

[0063] Perform grayscale processing on the image information and analyze the bright spot information of the image information;

[0064] Filtering the image information according to each bright spot information, determining the area boundary of the LED display device, and filtering to obtain valid bright spot information within the area boundary; and

[0065] Based on the effective bright spot information, the data of each LED light point within the area boundary is analyzed and obtained.

[0066] The details of one or more embodiments of the present application are set forth in the accompanying drawings and the description below. Other features, objects, and advantages of the present application will become apparent from the description, drawings, and claims. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the conventional technology, the following briefly introduces the drawings required for use in the embodiments or the conventional technology descriptions. Obviously, the drawings described below are merely embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on the disclosed drawings without any creative work.

[0068] FIG1 is a diagram illustrating an application environment of a method for extracting LED light point data according to an embodiment of the present application;

[0069] FIG2 is a flow chart of a method for extracting LED light point data in one embodiment of the present application;

[0070] FIG3 is a flowchart illustrating the steps of filtering image information according to each bright spot information, determining the region boundary of the LED display device, and filtering to obtain valid bright spot information within the region boundary in one embodiment of the present application;

[0071] FIG4 is a flowchart illustrating steps of filtering image information according to each bright spot information, determining the region boundary of the LED display device, and filtering to obtain valid bright spot information within the region boundary in another embodiment of the present application;

[0072] FIG5 is a flowchart illustrating steps for analyzing and obtaining data of each LED light point within a region boundary based on effective bright spot information in one embodiment of the present application;

[0073] FIG6 is a schematic diagram of smoothing processing in one embodiment of the present application;

[0074] FIG7 is a schematic diagram showing a comparison between a histogram and a smoothed histogram in one embodiment;

[0075] FIG8 is a schematic diagram of a bright spot detection process in one embodiment of the present application;

[0076] FIG9 is a structural block diagram of an LED light point data extraction device according to an embodiment of the present application; and

[0077] FIG10 is a diagram showing the internal structure of a computer device in one embodiment of the present application. DETAILED DESCRIPTION

[0078] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0079] It will be understood that the terms "first," "second," etc., used herein may be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish a first element from another element. For example, a first resistor may be referred to as a second resistor, and similarly, a second resistor may be referred to as a first resistor without departing from the scope of this application. The first resistor and the second resistor are both resistors, but they are not the same resistor.

[0080] It can be understood that the “connection” in the following embodiments should be understood as “electrical connection”, “communication connection”, etc. if there is transmission of electrical signals or data between the connected circuits, modules, units, etc.

[0081] When used herein, the singular forms "a", "an", and "the" may also include the plural forms, unless the context clearly indicates otherwise. It should also be understood that the terms "include / comprise" or "have" and the like specify the presence of stated features, integers, steps, operations, components, parts, or combinations thereof, but do not preclude the possibility of the presence or addition of one or more other features, integers, steps, operations, components, parts, or combinations thereof.

[0082] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application pertains. The terms used herein in the specification of this application are for the purpose of describing specific embodiments only and are not intended to limit this application.

[0083] The LED light point data extraction method provided in the embodiments of the present application can be applied in the application environment shown in Figure 1. An LED display device 100 is connected to a control system 108, which is connected to a correction calculation module 106. The correction calculation module 106 is connected to an image acquisition device 102 via a processor 104. The LED display device 100 is the device under test and includes an LED light board, which includes multiple LED light points. The image acquisition device 102 is configured to correspond to the LED display device 100, ensuring that the LED display device 100 is within the acquisition range of the image acquisition device 102 and that the plane of the LED light board of the LED display device 100 is parallel to the plane of the acquisition lens of the image acquisition device 102.

[0084] The control system 108 controls the start-up of the driver module in the LED display device 100, controls the LED light points on the LED display device 100 to be in an operating state, and causes the LED light board to emit light. The acquisition lens of the image acquisition device 102 captures an image of the LED display device 100, which includes the image of the above-mentioned illuminated LED light board. The processor 104 obtains the image captured by the image acquisition device 102, obtains the data of each LED light point in the image by executing the LED light point data extraction method, and transmits it to the correction calculation module 106. The correction calculation module 106 can calculate the correction coefficient based on the LED light point data and transmit the correction coefficient to the control system 108. When the control system 108 performs display correction on the LED display device 100, it can complete the display correction function based on the correction coefficient.

[0085] The control system 108, calibration calculation module 106, and processor 104 can be, but are not limited to, various personal computers, laptop computers, smartphones, and tablet computers, including chips with certain computing capabilities. In some application environments, the control system 108 can include a calibration calculation module. In other application environments, the control system 108, calibration calculation module, and processor 104 can also be various personal computers, laptop computers, smartphones, and tablet computers that include a computing chip, with different running software on the computing chip being used to execute the operations of the control system 108, calibration calculation module 106, and processor 104, respectively. The image acquisition device 102 can be a camera, an imager, a photosensor array, etc.

[0086] In one embodiment, as shown in FIG2 , a method for extracting LED light point data is provided. The method is described by applying the method to the processor 104 in FIG1 as an example, and includes the following steps:

[0087] Step S202: Acquire image information of the LED display device.

[0088] The LED display device includes LED light points. When there are two or more LED light points, the LED light points form an LED light board. The LED light board maintains a plane parallel to the plane of the image acquisition device's camera lens. That is, during image acquisition, the plane of the LED light board of the LED display device is parallel to the plane of the image acquisition device's camera screen. Specifically, the image acquisition device captures the LED display device to acquire image information from the LED display device. The processor is connected to the image acquisition device and acquires image information from the LED display device through the image acquisition device.

[0089] Step S204: grayscale processing is performed on the image information to obtain bright spot information of the image information.

[0090] Grayscale is a method of displaying images using black as a base color, with varying degrees of black saturation. Grayscale processing is the process of converting a color image into a grayscale image. In grayscale processing, pixels in the image appear as varying degrees of black saturation depending on the grayscale value.

[0091] Specifically, grayscale processing is performed on the image information to obtain grayscale image information. Bright spot information of the image information can be obtained based on the grayscale values ​​of each pixel within the grayscale image information. For example, when the grayscale value of a pixel exceeds a set grayscale threshold, the pixel is marked as a bright spot region. The grayscale value and position information of the bright spot region are recorded and integrated into a single bright spot information. Alternatively, a bright spot center can be selected from each bright spot region, and the grayscale value and position information of the bright spot center, along with the previously obtained bright spot regions, can be integrated into a single bright spot information.

[0092] Step S206 , filtering the image information according to each bright spot information, determining the region boundary of the LED display device, and filtering to obtain valid bright spot information within the region boundary.

[0093] Specifically, the obtained bright spot information and image information are analyzed. The image information includes basic image information, including basic information such as the resolution of the image acquisition device, to analyze the data of each bright spot. Based on this calculation and analysis of the bright spot information and basic image information, the region boundary of the LED display device in the image information can be determined. The processor determines whether the bright spot corresponding to each bright spot information falls within the region boundary of the LED display device, and then filters the bright spots. Bright spot information outside the region boundary of the LED display device is filtered out to obtain valid bright spot information within the region boundary.

[0094] Optionally, the basic image information also includes the resolution of the LED light panel in the LED display device. The resolution of the LED light panel represents the maximum area that each LED light point can occupy in the image information, that is, the maximum number of pixels. By analyzing the information of each bright spot in combination with the basic image information, valid bright spot information within the region boundary can be filtered.

[0095] Step S208: Based on the effective bright spot information, analyze and obtain the data of each LED light point within the area boundary.

[0096] Specifically, the effective bright spot information is analyzed and sorted, and integrated into the data of each LED light point. The LED light point data includes the position information of the LED light point and the brightness information of the LED light point, which facilitates the subsequent control system to complete the correction processing.

[0097] The above-mentioned method for extracting LED light point data includes obtaining image information of an LED display device, grayscale processing the image information, and analyzing the bright spot information of the image information. The image information is filtered according to each bright spot information, the regional boundary of the LED display device is determined, and the effective bright spot information within the regional boundary is filtered. Based on the effective bright spot information, the data of each LED light point within the regional boundary is analyzed. Among them, the LED display device includes LED light points. By filtering the bright spot information, the present application can obtain the regional boundary of the LED display device through calculation and reduce the error of the effective bright spot information. Therefore, on the basis that the LED display device and the image acquisition device are parallel in surface, there is no need to strictly limit the parallelism of the LED display device and the image acquisition device line in the image of the image acquisition device, which can improve the convenience of LED light point identification.

[0098] In one embodiment, as shown in FIG3 , step S206 includes step S302 , step S304 , and step S306 .

[0099] Step S302 : Screening the maximum light point area according to each bright spot information and image information to obtain denoised image information.

[0100] The maximum light point area is the maximum pixel area that each LED light point can occupy in the image information. The maximum light point area can be calculated based on the resolution of the image acquisition device and the resolution of the LED light board. For example, if the resolution of the image acquisition device is CW×CH and the resolution of the LED light board is DW×DH, the maximum light point area can be TW×TH. Where: TW = MIN(CW / DW, CH / DH) (1) TH = TW (2)

[0101] In the above formula, CW is the horizontal resolution of the image acquisition device, CH is the vertical resolution of the image acquisition device, DW is the horizontal resolution of the LED light board, DH is the vertical resolution of the LED light board, TW is the horizontal resolution of the maximum light point area, and TH is the vertical resolution of the maximum light point area. MIN() represents the minimum value.

[0102] Since the aspect ratio of the resolution of the image acquisition device is not necessarily equal to the aspect ratio of the resolution of the LED light board, when calibrating the acquisition, it is usually selected to fill the image of the image acquisition device in one direction, either in length or width, and leave a gap in the other direction. The length and width of the resolution of the image acquisition device (i.e., the horizontal resolution and the vertical resolution) are divided by the length and width of the resolution of the LED light board, which can represent the length and width of the pixel point of an LED light point on the LED light board when imaging in the image acquisition device. Usually, the light point is circular, so the minimum aspect ratio value (Formula 1) is taken as the length and width value of the light point area (Formula 2). The positive direction pixel point area formed by the length and width of the light point area is the maximum light point area. This can improve the accuracy of the calculated maximum light point area and facilitate subsequent processing.

[0103] Specifically, the maximum light point area is calculated based on the basic image information in the image information, and each bright spot information is screened based on the maximum light point area. Bright spots that do not meet the maximum light point area are screened out to obtain denoised image information. The screening rule for the maximum light point area is: obtain the length and width data of each bright spot information and compare it with the maximum light point area. When the length and width of a bright spot are greater than the length and width of the maximum light point area, the processor determines that the bright spot is not an LED light point, and may be a noise point generated by factors such as noise, poor contact, or light reflection. The processor then screens out the bright spot and performs the screening of the next bright spot until all the bright spot information is traversed. The bright spot information that has not been screened out is called denoised image information.

[0104] Step S304: performing boundary detection on the denoised image information to determine the area boundary of the LED display device.

[0105] Specifically, before performing boundary detection on the denoised image information, the denoised image information must be processed. Specifically, the bright spot information in the denoised image information is statistically analyzed based on the horizontal and vertical directions, resulting in horizontal and vertical bright spot distribution statistics. The horizontal bright spot distribution statistics are used to determine the left and right boundaries of the LED display device, while the vertical bright spot distribution statistics are used to determine the upper and lower boundaries of the LED display device.

[0106] Perform convolution calculations on the horizontal and vertical bright spot distribution statistics, and combine the resulting left and right boundaries and upper and lower boundaries to form the region boundaries. For example, call the edge detection module to perform convolution operations to identify boundaries. This convolution calculation determines the region boundaries of the LED display device.

[0107] Step S306 , filtering the bright spot information based on the region boundary to obtain valid bright spot information within the region boundary.

[0108] Specifically, the bright spot information is filtered according to the obtained area boundary, and the processor compares the position information of each bright spot information with the area boundary of the LED display device. When the position information of the bright spot information is outside the area boundary, the bright spot information is filtered out; when the position information of the bright spot information is within the area boundary, the bright spot information is valid bright spot information.

[0109] In this embodiment, the maximum light point area is screened based on each bright spot information and image information to obtain denoised image information, and then the denoised image information is subjected to boundary detection to determine the regional boundary of the LED display device. The bright spot information is screened based on the regional boundary to obtain valid bright spot information within the regional boundary. By removing excessive noise points and noise points outside the regional boundary of the LED display device, the bright spot information is accurately screened, so that the effective bright spot information obtained is accurate and has low error. The regional boundary of the LED display device can be identified. Even without strictly restricting the parallelism of the LED display device and the image acquisition device line in the image of the image acquisition device, the bright spot information can be accurately distinguished, noise errors are removed, and the convenience and accuracy of LED light point identification are improved.

[0110] In order to avoid the presence of noise points in the bright spot information near the region boundary of the LED light panel, which may lead to incorrect identification of the region boundary of the LED light panel, in one embodiment, as shown in FIG4 , after step S302 and before step S304 , further steps S402 , S404 and S406 are included.

[0111] Prior to step S402, the denoised image information must be processed. Specifically, the bright spot information in the denoised image information is statistically analyzed based on the horizontal and vertical directions to obtain horizontal and vertical bright spot distribution statistics. The horizontal bright spot distribution statistics are used to determine the left and right boundaries of the LED display device, while the vertical bright spot distribution statistics are used to determine the upper and lower boundaries of the LED display device.

[0112] Step S402 : performing smoothing processing on the bright spot information in the denoised image information to obtain smoothed denoised bright spot information.

[0113] Specifically, the horizontal bright spot distribution statistical results and the vertical bright spot distribution statistical results are smoothed respectively, and a smoothing core with a set radius is used to smooth the horizontal bright spot distribution statistical results and the vertical bright spot distribution statistical results respectively, so as to make the horizontal bright spot distribution statistical results and the vertical bright spot distribution statistical results data smooth, thereby avoiding the error in the regional boundary judgment of the LED light board due to the discontinuous brightness distribution of the bright spots.

[0114] Optionally, the smoothing kernel radius may be 3, and the smoothing kernel radius may be adjusted when approaching an edge area to prevent the smoothing width from exceeding the boundaries of the horizontal bright spot distribution statistical results and the vertical bright spot distribution statistical results.

[0115] For example, assuming that the statistical result of the horizontal bright spot distribution is HMY and the statistical result of the vertical bright spot distribution is VMY, the expression for the smoothing process is:

[0116] Where n is the horizontal or vertical position of the reference point, and i is the radius of the smoothing kernel rounded down to an integer. i is 0 at the boundary between SHY and SVY.

[0117] Step S404 , obtaining an erosion threshold value based on the smoothed denoised bright spot information; performing erosion processing on the smoothed denoised bright spot information based on the erosion threshold value to obtain eroded denoised bright spot information.

[0118] Specifically, the average of the horizontal and vertical distribution statistics of the smoothed denoised bright spot information is calculated, and the resulting average is used as the erosion threshold. The smoothed denoised bright spot information is eroded using an erosion kernel with the same radius as the smoothing kernel. Let the erosion threshold be FTH, the erosion result of the smoothed horizontal bright spot distribution statistics SHY be denoted as FHY, and the erosion result of the smoothed vertical bright spot distribution statistics SVY be denoted as FVY. The erosion process is then expressed as:

[0119] Step S406 , performing dilation processing based on the eroded denoised bright spot information to obtain dilated denoised bright spot information; and updating the denoised image information based on the dilated denoised bright spot information.

[0120] Specifically, a dilation process is performed based on the denoised bright spot information after corrosion. The dilation process result of the horizontal bright spot distribution statistical result FHY after corrosion is recorded as PHY, and the dilation process result of the vertical bright spot distribution statistical result FVY after corrosion is recorded as PVY. The dilation process expression is:

[0121] In this embodiment, by performing smoothing, corrosion and expansion processing before performing boundary detection on the denoised image information, the bright spot information in the denoised image information is optimized, which is conducive to accurately obtaining the regional boundary of the LED display device during subsequent boundary detection, and reduces the error in the regional boundary identification of the LED light board due to the presence of noise points in the bright spot information located near the regional boundary of the LED light board, which is conducive to accurately identifying the regional boundary of the LED display device.

[0122] In one embodiment, as shown in FIG5 , step S208 includes step S502 , step S504 , and step S506 .

[0123] Step S502: Determine empty points within the region boundary based on the valid bright spot information.

[0124] Among them, the empty point refers to the damaged LED light point in the LED light board. The damaged LED light point does not emit light when the LED display device is working, forming a empty point.

[0125] Specifically, before obtaining the valid bright spot information, the noise points outside the region boundary of the LED display device have been removed. However, there may be empty points within the region boundary of the LED display device. In this case, the processor needs to exclude the empty points within the region boundary of the LED display device based on the valid bright spot information.

[0126] In one embodiment, step S502 includes step S602 and step S604.

[0127] Step S602: Calculate the light point spacing based on the effective bright spot information.

[0128] Specifically, a valid bright spot information closest to the center point of the image captured by the image acquisition device is set as the initial point P, and then the horizontal bright spot distribution statistical results and the vertical bright spot distribution statistical results are differentially processed as required. When in the horizontal detection state, the horizontal bright spot distribution statistical results are differentially processed, and when in the vertical detection state, the vertical bright spot distribution statistical results are differentially processed. Taking the horizontal detection state as an example, the differential formula is as follows: l = ABS(MY[n].px - MY[n-1].px) (9)

[0129] In the formula, l represents the horizontal distance between two adjacent valid bright spot information, ABS represents the absolute value, MY is the set of valid bright spot information, n is the selected valid bright spot information, and px is the horizontal coordinate of the valid bright spot information. At the same time, when in the vertical detection state, px in formula (9) can be replaced with py. py represents the vertical coordinate of the valid bright spot information. In this case, l represents the vertical distance between two adjacent valid bright spot information.

[0130] The l obtained after the differential processing is statistically analyzed, and the l with the largest value is selected as the light point spacing L. If the light point spacings obtained in the horizontal detection state and the vertical detection state are different, the horizontal light point spacing LH and the vertical light point spacing LV need to be recorded separately.

[0131] Step S604: Analyze the effective bright spot information according to the light spot spacing and the preset deflection angle to determine the empty points within the area boundary.

[0132] Specifically, each bright spot in the effective bright spot information is calculated and analyzed based on the light point spacing and the preset deflection angle, the theoretical bright spot position is obtained based on the light point spacing, the theoretical bright spot position is matched with the effective bright spot information, and the theoretical bright spot position that does not match the effective bright spot information is determined as an empty point within the area boundary.

[0133] Optionally, when the matching rate between the effective bright spot information and the theoretical bright spot position falls below a set matching threshold, the preset deflection angle is adjusted based on the set accuracy. For example, if the matching threshold is set to 90% and the accuracy is set to 5°, when the matching rate between the effective bright spot information and the theoretical bright spot position falls below 90%, the preset deflection angle of 0° is adjusted to 5° based on the set accuracy of 5°.

[0134] The search range is determined based on the distance between the light points. A search is performed based on the initial point and the search range, with theoretical light point locations where no valid light point information exists being identified as empty points. For example, starting from the initial point P, the positions of each theoretical light point are sequentially determined based on the light point distance L (including the horizontal light point distance LH and the vertical light point distance LV). This means that starting from the initial point P, the theoretical light point locations are searched for any bright spots in the surrounding area, with the search range being LH × LV. During the search, theoretical light points where no valid light point information exists are marked as empty points.

[0135] Furthermore, since the present application does not require that the LED display device in the image capture device be parallel to the image capture device line, a preset deflection angle can be set to analyze the effective bright spot information when searching for the theoretical light point position. Let the preset deflection angle be Φ, and the value of Φ can be 0°. When the current light point position is CP, the retrieved new theoretical light point position is NP, and the calculation expression is as follows: NP.x=CP.x±LH×sinΦ (10) NP.y=CP.y±LV×cosΦ (11)

[0136] Wherein, x represents the horizontal coordinate of the light point in the image information, and y represents the vertical coordinate of the light point in the image information. The deviation between the actual deflection angle and the preset deflection angle Φ cannot be greater than atan(LV / LH) degrees.

[0137] Traverse and search for locations within the LED display's area boundary where theoretical light points may exist, and identify empty points within the area boundary. You can choose to search for the theoretical light point location horizontally first and then vertically, or you can search for the theoretical light point location vertically first and then horizontally, and then determine whether the theoretical light point location is empty based on whether there is bright spot information within the theoretical light point location.

[0138] Step S504: Sort the LED light points according to the information of the empty points and the effective bright points within the area boundary to obtain the sequence number of each LED light point.

[0139] Specifically, according to the information of empty points and effective bright spots within the region boundary, after excluding or marking the empty points within the region boundary, the LED light points in the effective bright spot information are sorted to obtain the sequence number of each LED light point.

[0140] For example, suppose the LEDS array includes the light point position information and light point type information, which is: {y, x, row, col, flag}, where x and y represent the horizontal and vertical coordinates of the pixel position of the light point in the image, row represents the row number of the light point on the LED light board, col represents the column number of the light point on the LED light board, the row and col initialization values ​​are 0, and flag represents the current light point type, which includes empty point (0) and light point (1).

[0141] Taking the initial point P as an example (Φ = 0, LH = LV = L), first search for the light points horizontally, then vertically, and record them in the light point array LEDS. Record the light point information of point P as col = 0, row = 0, flag = 1. And x and y can be obtained from px and py in the effective bright point information MY. In summary, the light point information of point P can be recorded as LEDS[0] = {px, py, 0, 0, 1}.

[0142] The search range of the light points cannot exceed the boundary of the LED display device. Let the upper boundary of the LED display device be MU, the lower boundary be MD, the left boundary be ML, and the right boundary be MR. First, search in one direction. For example, first search for the light points to the right. The current light point position CP = P, and the new light point position NP is:

[0143] NP.x = CP.x + L

[0144] NP.y = CP.y

[0145] NP.col = CP.col + 1

[0146] NP.row = CP.row

[0147] Where the search range cannot exceed the right boundary of the LED display device, that is, NP.x < MR, otherwise end the search to the right. When searching to the left, the new light point NP is:

[0148] NP.x = CP.x - L

[0149] NP.y = CP.y

[0150] NP.col = CP.col - 1

[0151] NP.row = CP.row

[0152] Where the search range cannot exceed the left boundary of the LED display device, that is, NP.x > ML, otherwise end the search to the left.

[0153] Traverse all the bright points in this row within the valid bright point information MY in this order. When no bright point is found within the L×L range centered on the point NP, then this point is an empty point, and it is recorded as LEDS[k] = {NP.x, NP.y, NP.row, NP.col, 0}, where k is a natural number. Conversely, if a bright point is found, then record LEDS[k] = {NP.x, NP.y, NP.row, NP.col, 1}. After completing the data recording of the new light points, assign the new light points to the old light points and continue to search for the next light point until the left or right boundary is retrieved and then end. All the points in the row where the initial point P is located are recorded in the array LEDS, including light points and empty points.

[0154] Then search upward and downward. When searching upward for sorting, the new light point NP is:

[0155] NP.x = CP.x

[0156] NP.y = CP.y - L

[0157] NP.col = CP.col

[0158] NP.row = CP.row - 1

[0159] Where the search range cannot exceed the upper boundary of the LED display device, that is, NP.y > MU, otherwise end the upward search. When searching downward, the new light point NP is:

[0160] NP.x = CP.x

[0161] NP.y = CP.y + 1

[0162] NP.col = CP.col

[0163] NP.row = CP.row + 1

[0164] Where the search range cannot exceed the lower boundary of the LED display device, that is, NP.x < MD, otherwise end the downward search.

[0165] Furthermore, when any boundary in any direction of the searched area boundary is an empty point, the area boundary of the LED display device is corrected again to further improve the accuracy of the area boundary of the LED display device.

[0166] Traverse the members of the LEDS array, find the minimum and maximum values of col and row, denoted as MinCol (negative value), MaxCol (positive value), MinRow (negative value), and MaxRow (positive value), for screening out boundary empty points.

[0167] Left boundary screening: Find all points with col = MinCol in LEDS. When all these points are empty points, increment MinCol by 1 and continue repeating the above steps until a non-empty point is found. At this time, MinCol is the left boundary of the light points.

[0168] Right boundary screening: Find all points with col = MaxCol in LEDS. When all these points are empty points, decrement MaxCol by 1 and continue repeating the above steps until a non-empty point is found. At this time, MaxCol is the right boundary of the light points.

[0169] Upper boundary screening: Find all points with row = MinRow in LEDS. When all these points are empty points, increment MinRow by 1 and continue repeating the above steps until a non-empty point is found. At this time, MinRow is the upper boundary of the light points.

[0170] Lower boundary screening: Find all points with row = MaxRow in LEDS. When all these points are empty points, decrement MaxRow by 1 and continue repeating the above steps until a non-empty point is found. At this time, MaxRow is the lower boundary of the light points.

[0171] After correcting the regional boundary of the LED display device, traverse the sorted array of all LED light points in LEDS again. When the row and col in the sorted array of LED light points exceed the regional boundary of the LED display device, then this sorted array of LED light points is screened out and removed from the LEDS array. That is, when the condition row > MaxRow or row < MinRow or col > MaxCol or col < MinCol is satisfied, this sorted array of LED light points is removed from the array LEDS. Finally, subtract MinCol from the col of all sorted arrays of LED light points in the array LEDS, and subtract MinRow from the row. The resulting array set LEDS is the sorted result of the LED light points, including the serial numbers of each LED light point.

[0172] Step S506: Analyze and obtain the data of each LED light point within the regional boundary by combining the effective bright point information and the serial numbers of each LED light point.

[0173] Specifically, according to the serial numbers of each LED light point, associate the gray values in the effective bright point information with the serial numbers of the LED light points and integrate them into the LED light point data.

[0174] The array set LEDS records the pixel positions of the image information of the light points and empty points, as well as the positions of the light points on the light board. Based on the pixel positions in the image information and the valid bright point information, the brightness value of each light point is calculated according to a specified method. This specified method can be to take the maximum grayscale value near the light point as the brightness value of the light point, or to take the average grayscale value of the pixels above the maximum grayscale g% near the light point (g ranges from [0, 100], and g = 75 is selected in this embodiment) as the brightness value of the light point. According to the selected bright point value calculation method, all members of the LEDS array are traversed, and the brightness value of each light point is calculated in turn. Combined with the array LEDS, the brightness value is integrated into the light point data array LEDY. The LEDY array includes the light board light point position (row, col) and brightness value LY in the LEDS array. Each LED light point has corresponding row and col position data and brightness value LY data.

[0175] In one embodiment, step S204 includes step S702 , step S704 and step S706 .

[0176] Step S702: grayscale processing is performed on the image information, and a histogram is obtained by statistics.

[0177] Specifically, grayscale processing is performed on the image information to obtain grayscale image information, and the grayscale value of each pixel in the grayscale image information is counted to obtain a histogram.

[0178] Optionally, if the light points in the grayscale image are too dense, forming patchy bright spots, an image opening operation may be performed on the grayscale image. This operation involves sequentially eroding and dilating the image, eliminating small, bright areas without significantly changing the areas of other objects. After the image opening operation, the grayscale values ​​of each pixel in the grayscale image are statistically analyzed to generate a histogram. This operation can be repeated until the light points are no longer densely packed.

[0179] Step S704: smoothing the histogram to obtain a smoothed histogram.

[0180] Specifically, a smoothing kernel with a set radius is used to smooth the histogram to eliminate the influence of burrs on the histogram and obtain a smoothed histogram. Under normal circumstances, the statistical curve of the histogram obtained from the image information should have two peaks and one trough. The two peaks represent the background area and the LED light point area of ​​the LED display device, respectively, and the trough between the two peaks is the threshold for distinguishing the background area from the LED light point area. However, due to the accuracy of the image acquisition equipment and the influence of the external environment, the statistical curve of the histogram may fluctuate abnormally and not be smooth enough, resulting in multiple troughs, which affects the subsequent LED light point data extraction. Therefore, it is necessary to smooth the statistical curve of the histogram to remove the influence of burrs and improve the accuracy of LED light point data extraction.

[0181] For example, suppose that in the statistical histogram, the statistical result is Y[n], n is the grayscale value, and Y[n] represents the number of times the pixel point with grayscale value n appears Y times. For example, the grayscale statistical length is 10 (i.e., ten different n) and the radius of the smoothing kernel is 2. As shown in Figure 6, the grayscale values ​​in the histogram statistics include ten values ​​0-9, and the smoothing kernel is represented by s. The grayscale value statistical result in the smoothed histogram is SY[n], and the calculation formula is:

[0182] Where SY[n] is the statistical result of the grayscale value n in the smoothing histogram, and m is the radius of the smoothing kernel. Since the radius of the smoothing kernel will be adjusted at the edge of the histogram to keep it within the statistical range of the histogram, that is, within the grayscale value range counted in the histogram, special edge processing is performed on the grayscale n in the [0, m-1] and [Max-m+1, Max] regions: m is temporarily assigned to m=i (i∈[0, m-1]) or m=Max-i (i∈[Max-m+1, Max]). That is, the radius of the smoothing kernel becomes smaller when approaching the edge, so that the smoothing area does not exceed the upper and lower grayscale boundaries, and the minimum value of m is 0.

[0183] As shown in Figure 6, when smoothing grayscale values ​​at the edge, the radius of the smoothing kernel decreases. For example, when smoothing grayscale values ​​0 and 9, the radius of the smoothing kernel is 0, while when smoothing grayscale values ​​1 and 8, the radius of the smoothing kernel is 1. A comparison of the histogram and the smoothed histogram is shown in Figure 7. The left side of Figure 7 shows the unsmoothed histogram, and the right side shows the smoothed histogram. Smoothing can remove the influence of burrs and improve the accuracy of LED light point data extraction.

[0184] Step S706 , analyzing the smoothed histogram to obtain a light point threshold, and performing bright spot recognition based on the light point threshold to obtain bright spot information of the image information.

[0185] The smoothed histogram is analyzed to determine the light point threshold, which is used to distinguish the LED light point area from the background area. The LED light point area is identified and marked as bright spot information. Specifically, the statistical curve in the smoothed histogram is analyzed, and the grayscale value corresponding to the trough of the background area peak and the LED light point area peak component is used as the light point threshold. Pixels with grayscale values ​​greater than the light point threshold are considered bright spot areas, and the bright spot center is selected from the bright spot area. Bright spot information includes the bright spot area and the bright spot center.

[0186] Exemplarily, the trough is selected by the following criteria:

[0187] SY[nk]≥SY[n] and SY[n+k]≥SY[n] and SY[n-k+1]≥SY[n] and SY[n+k-1]≥SY[n] and…SY[n-1]≥SY[n] and SY[n+1]≥SY[n].

[0188] Where SY[nk] represents the number of occurrences of the gray level k-th to the left of gray level n. The above determination is: the gray level n with the fewest occurrences within the range ±k centered on gray level n is used to determine whether gray level n is a valley within the range ±k.

[0189] The processing flow of valley detection is similar to the above-mentioned smoothing process, except that the smoothing process is changed to valley detection judgment, where k is the radius of the valley detection kernel, and the boundary is specially treated. k is temporarily assigned to k = i (i∈[0,k-1]) or k = Max-i (i∈[Max-k+1,Max]), that is, the smoothing width becomes smaller when approaching the edge, so that the smoothing area does not exceed the upper and lower boundaries of the grayscale. The minimum value of k is 0. When k=0, the detection result is invalid, that is, the minimum grayscale and maximum grayscale detection results are invalid.

[0190] To prevent interference, the grayscale values ​​corresponding to all troughs are counted as a preselected threshold value PTH. The light point threshold value TH is selected from the preselected threshold values. Specifically, when a grayscale value meets the trough condition, PTH[j] = n and is recorded as the preselected threshold value. Where j∈[0,Max], the initial value of j is 0 and is incremented by 1 each time a trough value is found. n is the grayscale value corresponding to the detected trough. TH is then selected from PTH based on the grayscale value. That is, the grayscale value closest to the average grayscale value within PTH is selected as the light point threshold value TH.

[0191] After determining the light point threshold, bright spot recognition is performed on the image information collected by the image acquisition device. Bright spot recognition is performed based on the grayscale value of the pixel in the image information, including: taking the pixel points with grayscale values ​​greater than the light point threshold as the bright spot area, and selecting the position with the largest grayscale value within the set bright spot range from the bright spot area as the bright spot center.

[0192] The bright spot detection process diagram is shown in Figure 8, where each square is a bright spot area, one square represents one pixel, and the bright spot range is set to h = 1. The pixels within the set bright spot range are the detection area, and the edge points are invalid and not detected. The grayscale values ​​of the detection point (the rectangle with a cross mark in Figure 8) and the points within the detection area (the rectangle with a slash mark in Figure 8) are compared and judged. When the grayscale of the detection point is the largest value in the detection area and is greater than the light point threshold TH, the detection point is defined as the bright spot center. As shown in Figure 8, you can choose to traverse the bright spot area from left to right and from top to bottom to determine the bright spot center, or you can choose other traversal methods to determine the bright spot center of the bright spot area.

[0193] The detection point is judged to be the brightest point among the pixels within the range of ±kx above and below the center (i.e., the set bright spot range h). Let the pixels in the bright spot area be marked as IMG[px][py], and the pixel information greater than the light point threshold is counted. The statistical result is defined as MY[j]. The MY[j] array includes the pixel position and the grayscale value of the pixel. The recognition condition expression is:

[0194] IMG[py][px]≥IMG[py-ky][px-kx] AND IMG[py][px]≥IMG[py-ky][px+kx] AND IMG[py][px]≥IMG[py-ky][px-kx+1] AND IMG[py][px]≥IMG[py-ky][px+kx-1] and IMG[py][px]≥IMG[py+ky][px+kx-1] and IMG[py][px]≥IMG[py+ky][px+kx]

[0195] Where py and px represent the horizontal and vertical coordinates of the current detection point in the image information, and ky and kx represent half the width and height of the detection area, rounded down, corresponding to the set bright spot range. When the bright spot detection conditions are met, the statistics MY[j] = {py, px, IMG[py][px]} are recorded as bright spot information. Where j∈[0,Max], the initial value of j is 0 and the value of j is increased by 1 each time a bright spot is found.

[0196] In this embodiment, the image information is grayscale processed to obtain a histogram, which is then smoothed to obtain a smoothed histogram. A light threshold is selected from the smoothed histogram, and bright spot identification is performed based on the light threshold to obtain bright spot information. This accurately determines the bright spot information of the LED display device in the image information, ensuring accurate subsequent extraction of LED light spot data.

[0197] To better understand the above solution, the following detailed explanation is provided using a specific embodiment, in conjunction with the application scenario shown in Figure 1. In one embodiment, the processor performs a seven-step LED light point data extraction method, including: histogram statistics, histogram smoothing, light point threshold calculation, bright spot identification, noise point screening, light point sorting, and brightness extraction.

[0198] The processor obtains image information from the LED display device, performs grayscale processing on the image information, generates a statistical histogram, smoothes the histogram to obtain a smoothed histogram, analyzes the smoothed histogram to determine a light threshold, and identifies bright spots based on the light threshold to obtain bright spot information in the image information. The specific steps for performing histogram statistics, histogram smoothing, light threshold calculation, and bright spot identification are described in detail above and will not be repeated here.

[0199] The noise removal step removes noise outside the LED display's area boundaries. The processor filters the maximum light point area based on each bright spot information and image information to obtain denoised image information. The bright spot information within the denoised image information is then smoothed, eroded, and expanded. Boundary detection is then performed to determine the LED display's area boundaries. Based on the area boundaries, the bright spot information is filtered to obtain valid bright spot information within the area boundaries.

[0200] The light point sorting and brightness extraction steps integrate LED light point information, facilitating subsequent calibration of the LED display device. The processor calculates the distance between light points based on the valid bright spot information. The valid bright spot information is analyzed based on the distance between light points and the preset deflection angle to determine the empty points within the area boundary. Based on the empty points and valid bright spot information within the area boundary, each LED light point is sorted to obtain a sequence number for each LED light point. Combining the valid bright spot information with the sequence number of each LED light point, the data for each LED light point within the area boundary is analyzed.

[0201] In this embodiment, there is no need to strictly limit the parallelism between the LED display device and the image capture device in the image capture device's frame. By determining the region boundaries of the LED display device, precise positioning of the LED light points can be achieved. Large noise points and noise points outside the region boundaries are eliminated, resulting in accurate LED light point data, which facilitates subsequent display calibration of the LED display device.

[0202] It should be understood that, although the steps in the flowcharts of the above embodiments are shown in sequence as indicated by the arrows, these steps are not necessarily performed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be performed in other orders. Moreover, at least a portion of the steps in the flowcharts of the above embodiments may include multiple steps or multiple stages, and these steps or stages are not necessarily performed at the same time, but can be performed at different times. The execution order of these steps or stages is not necessarily to be performed in sequence, but can be performed in turn or alternately with other steps or at least a portion of steps or stages in other steps.

[0203] Based on the same inventive concept, embodiments of the present application also provide an LED light point data extraction device for implementing the aforementioned LED light point data extraction method. The solution provided by this device is similar to the solution described in the aforementioned method. Therefore, the specific limitations of one or more embodiments of the LED light point data extraction device provided below can be found in the above-described limitations of the LED light point data extraction method and will not be further elaborated here.

[0204] In one embodiment, as shown in FIG9 , a device for extracting LED light point data is provided, including: an image acquisition module 902 , a bright spot analysis module 904 , a bright spot screening module 906 , and a data extraction module 908 , wherein:

[0205] The image acquisition module 902 is used to acquire image information of an LED display device, where the LED display device includes LED light points.

[0206] The bright spot analysis module 904 is used to perform grayscale processing on the image information and obtain bright spot information of the image information through analysis.

[0207] The bright spot screening module 906 is used to screen the image information according to each bright spot information, determine the area boundary of the LED display device, and screen to obtain the effective bright spot information within the area boundary.

[0208] The data extraction module 908 is used to analyze and obtain the data of each LED light point within the area boundary based on the effective bright spot information.

[0209] In one embodiment, the bright spot screening module 906 is further configured to screen the maximum light spot region based on each bright spot information and image information to obtain denoised image information. Boundary detection is performed on the denoised image information to determine the region boundaries of the LED display device. Based on the region boundaries, the bright spot information is screened to obtain valid bright spot information within the region boundaries.

[0210] In one embodiment, the bright spot screening module 906 is further configured to obtain a maximum light spot area based on basic image information in the image information, and to screen each bright spot information based on the maximum light spot area to obtain denoised image information.

[0211] In one embodiment, the bright spot screening module 906 is further configured to obtain horizontal and vertical bright spot distribution statistics based on the denoised image information. Convolution calculations are performed on the horizontal and vertical bright spot distribution statistics, and the obtained left and right boundaries and upper and lower boundaries are integrated into the region boundaries.

[0212] In one embodiment, the bright spot screening module 906 is further configured to, after performing maximum light spot area screening based on each bright spot information and image information to obtain denoised image information, perform boundary detection on the denoised image information to determine the region boundary of the LED display device, smooth the bright spot information within the denoised image information to obtain smoothed denoised bright spot information. Determine an erosion threshold based on the smoothed denoised bright spot information, perform erosion processing on the smoothed denoised bright spot information based on the erosion threshold to obtain eroded denoised bright spot information. Perform dilation processing based on the eroded denoised bright spot information to obtain dilated denoised bright spot information. Finally, update the denoised image information based on the dilated denoised bright spot information.

[0213] In one embodiment, the data extraction module 908 is further configured to determine empty points within the region boundary based on the valid bright point information. The LED light points are sorted based on the empty points and valid bright point information within the region boundary to obtain a sequence number for each LED light point. The valid bright point information and the sequence number of each LED light point are combined to analyze and obtain data for each LED light point within the region boundary.

[0214] In one embodiment, the data extraction module 908 is further configured to calculate the light point spacing based on the effective light point information, analyze the effective light point information according to the light point spacing and the preset deflection angle, and determine the empty points within the area boundary.

[0215] In one embodiment, the data extraction module 908 is further configured to obtain a theoretical bright spot position based on the light spot spacing, match the theoretical bright spot position with valid bright spot information, and determine a theoretical bright spot position that does not match the valid bright spot information as a null point.

[0216] In one embodiment, the data extraction module 908 is further configured to determine a search range based on the distance between light points, perform a search based on the initial point and the search range, and determine a theoretical bright spot position where no valid bright spot information exists as a null point.

[0217] In one embodiment, the bright spot analysis module 904 is further configured to perform grayscale processing on the image information to generate a histogram, smooth the histogram to generate a smoothed histogram, analyze the smoothed histogram to obtain a bright spot threshold, and perform bright spot identification based on the bright spot threshold to obtain bright spot information of the image information.

[0218] Each module in the aforementioned LED light point data extraction device can be implemented in whole or in part through software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in hardware form, or can be stored in a computer device memory in software form, so that the processor can call and execute the corresponding operations of each module.

[0219] In one embodiment, a computer device is provided, which may be a terminal. Its internal structure may be as shown in Figure 10. The computer device includes a processor, memory, an input / output interface, a communication interface, a display unit, and an input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are connected to the system bus via the input / output interface. The processor of the computer device provides computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operating system and computer program stored in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with external terminals via wired or wireless means, where the wireless means may be implemented via Wi-Fi, a mobile cellular network, NFC (near-field communication), or other technologies. When executed by the processor, the computer program implements a method for extracting data from LED light points. The display unit of the computer device is used to produce a visually visible image and may be a display screen, a projection device, or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device casing, or an external keyboard, touchpad or mouse.

[0220] Those skilled in the art will understand that the structure shown in FIG10 is merely a block diagram of a portion of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.

[0221] In one embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented:

[0222] Acquire image information of an LED display device, where the LED display device includes LED light points;

[0223] Perform grayscale processing on the image information and analyze the bright spot information of the image information;

[0224] Filtering the image information according to each bright spot information, determining the area boundary of the LED display device, and filtering to obtain valid bright spot information within the area boundary; and

[0225] Based on the effective bright spot information, the data of each LED light point within the area boundary is analyzed and obtained.

[0226] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0227] Based on the information of each bright spot and the image, the maximum light spot area is filtered to obtain the denoised image information. The denoised image information is subjected to boundary detection to determine the area boundary of the LED display device. The bright spot information is filtered based on the area boundary to obtain the valid bright spot information within the area boundary.

[0228] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0229] The bright spot information in the denoised image information is smoothed to obtain smoothed denoised bright spot information. An erosion threshold is obtained based on the smoothed denoised bright spot information, and the smoothed denoised bright spot information is eroded based on the erosion threshold to obtain eroded denoised bright spot information. An expansion process is performed based on the eroded denoised bright spot information to obtain expanded denoised bright spot information; and the denoised image information is updated based on the expanded denoised bright spot information.

[0230] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0231] Based on the valid bright spot information, empty spots within the region boundary are determined. Based on the empty spots and valid bright spot information within the region boundary, each LED light point is sorted to obtain a sequence number for each LED light point. Combining the valid bright spot information and the sequence number of each LED light point, the data for each LED light point within the region boundary is analyzed.

[0232] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0233] Based on the effective bright spot information, the distance between the light points is calculated. The effective bright spot information is analyzed according to the distance between the light points and the preset deflection angle to determine the empty points within the area boundary.

[0234] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0235] The image information is grayscale processed and statistically analyzed to obtain a histogram. The histogram is smoothed to obtain a smoothed histogram. The smoothed histogram is analyzed to obtain a light point threshold, and bright spot identification is performed based on the light point threshold to obtain bright spot information of the image information.

[0236] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0237] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0238] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, database or other media used in the embodiments provided in this application may include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory may include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, distributed databases based on blockchains. The processor involved in the various embodiments provided herein may be, but are not limited to, a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic unit, a data processing logic unit based on quantum computing, and the like.

[0239] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0240] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present application. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present application, and these modifications and improvements fall within the scope of protection of the present application. Therefore, the scope of protection of the present application shall be determined by the appended claims.

Claims

1. A method for extracting LED light point data, the method comprising: Acquire image information of an LED display device, wherein the LED display device includes LED light points; Performing grayscale processing on the image information to analyze and obtain bright spot information of the image information; Filtering the image information according to each of the bright spot information, determining the area boundary of the LED display device, and filtering to obtain valid bright spot information within the area boundary; as well as Based on the effective bright spot information, the data of each LED light point within the area boundary is analyzed and obtained.

2. The method according to claim 1, wherein: The step of screening the image information according to each of the bright spot information, determining the region boundary of the LED display device, and screening to obtain the effective bright spot information within the region boundary includes: Screening the maximum light spot area according to each of the bright spot information and the image information to obtain denoised image information; Performing boundary detection on the denoised image information to determine the area boundary of the LED display device; and The bright spot information is screened based on the region boundary to obtain valid bright spot information within the region boundary.

3. The method according to claim 2, wherein: The step of screening the maximum light spot area according to each of the bright spot information and the image information to obtain denoised image information includes: Obtaining the maximum light point area according to basic image information in the image information; and Based on the maximum light spot area, each of the bright spot information is screened to obtain the denoised image information.

4. The method according to claim 2, wherein: The performing boundary detection on the denoised image information to determine the area boundary of the LED display device includes: Based on the denoised image information, obtaining horizontal bright spot distribution statistics and vertical bright spot distribution statistics; and Convolution calculations are performed on the horizontal bright spot distribution statistics results and the vertical bright spot distribution statistics results respectively, and the obtained left and right boundaries and upper and lower boundaries are integrated into the region boundaries.

5. The method according to claim 2, wherein: After the maximum light spot area is screened according to each of the bright spot information and the image information to obtain the denoised image information, and before the boundary detection is performed on the denoised image information to determine the area boundary of the LED display device, the method further includes: Smoothing the bright spot information in the denoised image information to obtain smoothed denoised bright spot information; Obtaining a corrosion threshold according to the smoothed denoised bright spot information; Performing corrosion processing on the smoothed denoised bright spot information according to the corrosion threshold to obtain the denoised bright spot information after corrosion; Performing expansion processing based on the eroded denoised bright spot information to obtain expanded denoised bright spot information; and The denoised image information is updated based on the expanded denoised bright spot information.

6. The method according to claim 1, wherein: The analyzing and obtaining the data of each LED light point within the area boundary based on the effective bright spot information includes: Based on the effective bright spot information, determining empty points within the region boundary; Sort the LED light points according to the empty points within the region boundary and the valid bright point information to obtain the sequence number of each LED light point; and The effective bright spot information and the serial number of each LED light point are combined to analyze and obtain the data of each LED light point within the boundary of the area.

7. The method according to claim 6, wherein: The step of determining the empty points within the region boundary based on the effective bright spot information includes: Based on the effective bright spot information, calculating the light spot spacing; and The effective bright spot information is analyzed according to the light spot spacing and the preset deflection angle to determine the empty points within the area boundary.

8. The method according to claim 7, wherein the step of analyzing the effective bright spot information according to the light spot spacing and the preset deflection angle to determine the empty points within the region boundary comprises: According to the light point spacing, a theoretical bright spot position is obtained; as well as The theoretical bright spot position is matched with the effective bright spot information, and the theoretical bright spot position that does not match the effective bright spot information is determined as the empty point.

9. The method according to claim 8, wherein determining the theoretical bright spot position that does not match the valid bright spot information as the empty point comprises: Determine the search range based on the distance between light points; as well as A search is performed based on the initial point and the search range, and the theoretical bright spot position where the valid bright spot information does not exist is determined as the empty point.

10. The method according to claim 1, wherein: The grayscale processing of the image information to obtain the highlight information of the image information through analysis includes: Performing grayscale processing on the image information and obtaining a histogram by statistics; Smoothing the histogram to obtain a smoothed histogram; and The smoothed histogram is analyzed to obtain a light point threshold, and bright spot identification is performed according to the light point threshold to obtain bright spot information of the image information.

11. The method according to claim 1, wherein: The LED light point data includes the position information of the LED light point and the brightness information of the LED light point.

12. A device for extracting LED light point data, the device comprising: An image acquisition module, used to acquire image information of an LED display device, wherein the LED display device includes an LED light point; A bright spot analysis module, used for performing grayscale processing on the image information to obtain bright spot information of the image information through analysis; A bright spot screening module, used for screening the image information according to each bright spot information, determining the area boundary of the LED display device, and screening to obtain the effective bright spot information within the area boundary; as well as The data extraction module is used to analyze and obtain the data of each LED light point within the area boundary based on the effective bright spot information.

13. A computer device comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the method according to any one of claims 1 to 11 when executing the computer program.

14. A computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the method according to any one of claims 1 to 11.

15. A computer program product, comprising a computer program, which, when executed by a processor, implements the steps of the method according to any one of claims 1 to 11.

Images