LED backlight brightness detection device and detection method thereof

Abstract

The invention relates to the technical field of LED backlight source testing, in particular to an LED backlight source brightness detection device and a detection method thereof. The method comprises the following steps: firstly, acquiring a static frame sequence of a brightness image of an LED backlight source to be detected and a synchronous driving current time sequence signal; constructing a brightness disorder coefficient based on the static frame sequence, and generating a detection distortion value by combining edge curvature mutation analysis of the superpixel region; generating a brightness fluctuation index based on the static frame sequence and the driving current time sequence signal, and constructing a defect comprehensive factor; analyzing the time domain correlation between the brightness fluctuation index and the defect comprehensive factor, and determining a brightness anomaly dynamic index; fusing the detection distortion value and the brightness abnormity dynamic index to generate a key space-time dynamic index; and inputting the key time-space dynamic index and the brightness image into a neural network model, and outputting a brightness detection index of the LED backlight source to be detected. According to the invention, the recognition capability of highlight area details and dark area gradient during brightness detection of the LED backlight source is improved.

Description

Technical Field

[0001] This invention relates to the field of LED backlight testing technology, and specifically to an LED backlight brightness detection device and its detection method. Background Technology

[0002] LED backlights are a key component of LCD displays and other display devices, widely used in products such as televisions, mobile phones, and tablets. Brightness is one of the important indicators for measuring display quality, and it is affected by various factors, such as the quality of the LED chips, the stability of the power supply, and the heat dissipation design, leading to differences in brightness during the manufacturing process. To ensure that the brightness of each display panel meets the standard, accurate brightness testing is essential.

[0003] Currently, most LED backlight source brightness detection devices on the market use photoelectric sensors or spectrometers. Due to the limited bit width of analog-to-digital converters, they are forced to adopt linear or fixed nonlinear compression strategies in high dynamic range scenarios. This causes high-brightness areas to lose detailed features due to signal saturation, while the true brightness signal in low-brightness areas is submerged by quantization noise, resulting in irreversible degradation of brightness information in key areas. Consequently, the devices are not capable of capturing fine structures such as local micro-bright spots and gradual transitions in dark areas, leading to significant distortion in the detection results. Summary of the Invention

[0004] To address the technical problem of significant distortion in the brightness detection results when measuring the brightness of LED backlight sources, the present invention aims to provide an LED backlight brightness detection device and method, the specific technical solution of which is as follows: In a first aspect, embodiments of the present invention provide an LED backlight brightness detection method, the method comprising: Acquire the static frame sequence of the brightness image of the LED backlight under test and the synchronous driving current timing signal; Based on the static frame sequence, a brightness disorder coefficient is constructed by the local pixel feature distribution of the gridded sub-window, and a detection distortion value is generated by combining the edge curvature mutation analysis of the superpixel region. Based on the static frame sequence and the driving current timing signal, a brightness fluctuation index is generated; a defect comprehensive factor is constructed by combining the edge gradient direction change angle of the superpixel region with the brightness change of the neighboring region; the temporal correlation between the brightness fluctuation index and the defect comprehensive factor is analyzed to determine the dynamic index of brightness anomaly. The detected distortion value and the brightness anomaly dynamic index are fused to generate a key spatiotemporal dynamic index; the key spatiotemporal dynamic index and the brightness image are input into a trained neural network model to output the brightness detection index of the LED backlight under test.

[0005] Furthermore, the step of constructing a brightness disorder coefficient based on the local pixel feature distribution of the gridded sub-windows, according to the static frame sequence, includes: A brightness disorder coefficient is constructed based on the local brightness range and global variance of the gridded sub-windows; wherein the local brightness range is the pixel range in each gridded sub-window; and the global variance is the variance of the pixel mean in all gridded sub-windows.

[0006] Furthermore, the step of generating detection distortion values ​​by combining edge curvature abrupt changes in the superpixel region includes: Obtain the superpixel region in the brightness image of each frame in a static frame sequence; The degree of curvature change in the edge region is determined based on the curvature abrupt changes of adjacent edge points on the edge of the superpixel region. For each frame of the brightness image, the detection distortion value of each frame of the brightness image of the LED backlight under test is determined by combining the brightness disorder coefficient and the degree of abrupt change.

[0007] Further, generating the brightness fluctuation index based on the static frame sequence and the driving current timing signal includes: Construct a regional brightness mean sequence for each superpixel region in a static frame sequence within a complete period; the elements in the regional brightness mean sequence are the brightness mean of the same superpixel region in the brightness image of the static frame sequence arranged in time sequence; For each superpixel region, obtain the mean range of brightness in the mean brightness sequence of each superpixel region; For each superpixel region, the correlation between the region's average brightness sequence and the driving current timing signal is analyzed, and a correlation deviation weight is constructed. By combining the mean range of brightness and the correlation deviation weight, a brightness fluctuation index is generated for each superpixel region.

[0008] Furthermore, for each superpixel region, the correlation between the mean brightness sequence of the region and the driving current timing signal is analyzed, and a correlation deviation weight is constructed, including: For each superpixel region, the Pearson correlation coefficient between the region's mean brightness sequence and the driving current time-series signal sequence is calculated, and the Pearson correlation coefficient is negatively correlated. The result of the negative correlation mapping is used as the correlation deviation weight.

[0009] Furthermore, the defect synthesis factor constructed by combining the abrupt change angle of the edge gradient direction of the superpixel region with the brightness change of the neighboring region includes: The angle formed by the gradient directions between adjacent pixels in the superpixel region is taken as the edge gradient direction abrupt change angle; The average gradient direction abruptness angle is obtained by fusing the angle values ​​of the edge gradient direction abruptness angle between all adjacent pixels in the superpixel region. Using the currently analyzed superpixel region as the central superpixel region, the information entropy of the average brightness of the central superpixel region and the neighboring regions is calculated as the brightness change of the neighboring regions; where the neighboring regions are the superpixel regions adjacent to the central superpixel region. A defect comprehensive factor is constructed by combining the mean gradient direction abrupt change angle and the brightness change in the neighborhood region.

[0010] Furthermore, the analysis of the time-domain correlation between the brightness fluctuation index and the defect comprehensive factor to determine the dynamic index of brightness anomalies includes: Based on the time sequence, construct the brightness fluctuation index sequence and the defect comprehensive factor sequence; Analyze the correlation between the brightness fluctuation index sequence and the defect comprehensive factor sequence; When the correlation between the brightness fluctuation index sequence and the defect comprehensive factor sequence is positive, the Pearson correlation coefficient between the brightness fluctuation index sequence and the defect comprehensive factor sequence is used as a correction factor to correct the sum of the brightness fluctuation index and the defect comprehensive factor in the current superpixel region, thus obtaining a dynamic index of brightness anomaly.

[0011] Furthermore, the process of fusing the detected distortion value and the brightness anomaly dynamic index to generate key spatiotemporal dynamic indicators includes: A positive correlation mapping is performed on the aforementioned brightness anomaly dynamic index to obtain the interference removal enhancement factor; By combining the detection distortion value and the interference removal enhancement factor, key spatiotemporal dynamic indicators are determined.

[0012] Furthermore, the training set of the neural network model consists of: key spatiotemporal dynamic indicators, the average brightness of the brightness image of the LED backlight, and brightness detection indicators.

[0013] Secondly, an LED backlight brightness detection device is provided, the device comprising the following modules: The optical acquisition module is used to acquire the static frame sequence of the brightness image of the LED backlight under test and the synchronous drive current timing signal. The data processing and analysis module, connected to the optical acquisition module, is used to receive and process the static frame sequence and the driving current timing signal. The processing includes: Acquire the static frame sequence of the brightness image of the LED backlight under test and the synchronous driving current timing signal; Based on the static frame sequence, a brightness disorder coefficient is constructed by the local pixel feature distribution of the gridded sub-window, and a detection distortion value is generated by combining the edge curvature mutation analysis of the superpixel region. Based on the static frame sequence and the driving current timing signal, a brightness fluctuation index is generated; a defect comprehensive factor is constructed by combining the edge gradient direction change angle of the superpixel region with the brightness change of the neighboring region; the temporal correlation between the brightness fluctuation index and the defect comprehensive factor is analyzed to determine the dynamic index of brightness anomaly. The detected distortion value and the brightness anomaly dynamic index are fused to generate a key spatiotemporal dynamic index; the key spatiotemporal dynamic index and the brightness image are input into a trained neural network model to output the brightness detection index of the LED backlight under test.

[0014] Thirdly, embodiments of the present invention provide an electronic device, including a memory and a processor, wherein the memory stores executable code, and when the processor executes the executable code, it implements the various possible implementations of the first aspect.

[0015] Fourthly, embodiments of the present invention provide a computer program product comprising: computer program code, which, when executed on a computer, causes the computer to perform the method described in the first aspect or any possible implementation thereof.

[0016] Fifthly, embodiments of the present invention provide a computer-readable storage medium having a computer program stored thereon, which, when executed in a computer, causes the computer to perform the various possible implementations of the first aspect.

[0017] The embodiments of the present invention have at least the following beneficial effects: This invention first analyzes the brightness disorder coefficient constructed by the gridded sub-window, and generates detection distortion values ​​by combining the edge curvature mutation analysis of the superpixel region, simultaneously achieving macroscopic uniformity assessment and microscopic bright spot / dark area localization. Second, it generates a brightness fluctuation index based on static frame sequences and driving current timing signals, and constructs a defect comprehensive factor by combining the pixel gradient direction mutation angle with the brightness change in the neighboring region, quantifying the continuous degradation of optical diffusion. Furthermore, it analyzes the strong positive correlation between the brightness fluctuation index and the defect comprehensive factor in the time domain, outputting a dynamic indicator of brightness anomalies. Finally, it fuses the detection distortion value and the dynamic indicator of brightness anomalies to generate key spatiotemporal dynamic indicators, accurately capturing the positive feedback loop of dynamic runaway induced by static defects. This invention significantly improves the ability to identify details in high-brightness areas and gradual changes in dark areas when detecting the brightness of LED backlights, avoiding significant distortion in detection results, and providing millisecond-level dynamic degradation criteria for backlight quality control. Attached Figure Description

[0018] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a flowchart of a method for detecting the brightness of an LED backlight according to an embodiment of the present invention; Figure 2 This is a schematic diagram of a device module for an LED backlight brightness detection device according to an embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of a computer device provided in one embodiment of the present invention. Detailed Implementation

[0020] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of an LED backlight brightness detection device and its detection method based on the present invention.

[0021] In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments may be combined in any suitable form.

[0022] In the description of the embodiments of the present invention, unless otherwise stated, " / " means "or". For example, A / B can mean A or B. The "and / or" in the text is merely a description of the relationship between related objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist simultaneously, and B exists alone. In addition, in the description of the embodiments of the present invention, "multiple" means two or more.

[0023] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature.

[0024] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0025] The embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art will recognize that, with technological advancements and the emergence of new scenarios, the technical solutions provided by the embodiments of the present invention are also applicable to similar technical problems.

[0026] This invention provides a specific implementation method for an LED backlight brightness detection device and method, applicable to LED backlight brightness detection scenarios. In this scenario, the LED backlight to be tested is placed on a vacuum adsorption stage, and its position is calibrated using a laser alignment instrument. A diffuser plate and an optical lens are placed between the detection device and the LED backlight. The diffuser plate is used to homogenize the light, while the optical lens is used to control the direction and angle of light propagation, and also to further improve uniformity.

[0027] The following description, in conjunction with the accompanying drawings, details the specific scheme of the LED backlight brightness detection device and its detection method provided by the present invention.

[0028] Please see Figure 1 The diagram illustrates a flowchart of an LED backlight brightness detection method according to an embodiment of the present invention, which includes the following steps: Step S100: Obtain the static frame sequence of the brightness image of the LED backlight under test and the synchronous driving current timing signal.

[0029] A CMOS area array sensor is used as the detection device, with a single pixel size of 3.45 micrometers (μm) × 3.45 micrometers (μm). It is arranged in an 8×8 grid array to cover the effective area of ​​the LED backlight under test. A telecentric lens is loaded simultaneously to eliminate viewing angle distortion, and a narrowband filter is switched according to the target spectral range. The wavelength range of the narrowband filter is 450~650 nanometers (nm).

[0030] The LED backlight under test is activated, and the sensor array is triggered simultaneously. Incident photons are converted into photocurrent signals by the CMOS photosensitive unit. The photocurrent is then converted into a voltage signal by a transimpedance amplifier. The analog-to-digital converter quantizes the voltage signal with 14-bit resolution to generate raw brightness data values, which in turn constitute complete backlight brightness image data. The pixel value of each pixel in the backlight brightness image represents the brightness data at that location.

[0031] Acquire sequential brightness images of the LED backlight under test and form a static frame sequence; the elements in the static frame sequence are brightness images ordered according to time sequence.

[0032] Furthermore, auxiliary data is acquired simultaneously, which is the driving current timing signal: based on the CMOS area array sensor, the driving current timing signal of the LED backlight under test is recorded synchronously using a Hall sensor during each exposure cycle.

[0033] The acquired static frame sequence and drive current timing signal are stored in the buffer area, marked with timestamps and location information; the data packets are transmitted to the host computer processing system through a high-speed communication interface to complete the data acquisition process.

[0034] Step S200: Based on the static frame sequence, a brightness disorder coefficient is constructed by the local pixel feature distribution of the gridded sub-window, and a detection distortion value is generated by combining the edge curvature mutation analysis of the superpixel region.

[0035] During the production process of LED backlights, due to the discreteness of LED chips, fluctuations in driving current, or uneven heat dissipation, uneven brightness distribution may occur on the surface of the LED backlight, manifested as local micro-bright spots or dark areas. Local micro-bright spots correspond to overly bright areas, and dark areas correspond to overly dark areas.

[0036] For any brightness image in a static frame sequence, the brightness image is divided into N sub-windows, with the top-left corner as the origin, in a left-to-right and top-to-bottom order. Each gridded sub-window has its own corresponding sub-window region. In this embodiment of the invention, the window size is 5×5mm with a step size of 3mm. This window size can be adjusted based on the actual spacing between LED units.

[0037] Based on the static frame sequence, a brightness disorder coefficient is constructed by the local pixel feature distribution of the gridded sub-windows. Specifically, the brightness disorder coefficient is constructed based on the local brightness range and global variance of the gridded sub-windows. The local brightness range is the pixel range in each gridded sub-window, and the global variance is the variance of the pixel mean in all gridded sub-windows.

[0038] A larger local brightness range indicates a greater local brightness difference in the LED backlight under test. The global variance is obtained more specifically by calculating the pixel mean of the sub-window region corresponding to each gridded sub-window. Then, for any frame of brightness image, the variance of the pixel mean of all gridded sub-windows in the brightness image is calculated as the global variance. A larger global variance indicates a more disordered pixel distribution in the brightness image, corresponding to a more uneven brightness of the LED backlight. This global variance characterizes the brightness disorder of the backlight.

[0039] In one embodiment of the present invention, the global variance is corrected by the local brightness range to obtain the brightness disorder coefficient of each frame of the brightness image. Specifically, the mean of the local brightness range of all gridded sub-windows in the brightness image is denoted as... The global variance of the brightness image is denoted as... Then the brightness disorder coefficient R of the corresponding brightness image is: .

[0040] The brightness disorder coefficient R reflects that the brightness of the LED backlight is not uniform, the brightness changes too drastically, the quality of the backlight is unstable, and the display effect is prone to dark areas or bright spots.

[0041] After obtaining the brightness disorder coefficient, the detection distortion value is generated by combining it with the edge curvature mutation analysis of the superpixel region.

[0042] Because uneven brightness of the LED backlight under test was found through rapid grid scanning, high-precision analysis was then performed to identify that the uneven brightness was due to significant dark areas or bright spots in the backlight.

[0043] Therefore, further, superpixel segmentation is performed on the luminance images in the static frame sequence. The SLIC superpixel segmentation algorithm is used to segment the luminance images into several superpixel regions with similar pixel features. The superpixel regions in the luminance images of each frame in the static frame sequence are then obtained.

[0044] In the brightness image of an LED backlight, the essence of micro-bright spots and dark areas is optical diffusion discontinuity. Due to the uneven coating of LED encapsulant bubbles or phosphors, light scattering abruptly occurs in local areas, which manifests as a sharp change in curvature at the image edge, such as a jagged outline.

[0045] For each superpixel region in the brightness image, the Soble operator is used to extract the edge of the superpixel region and the corresponding edge pixel.

[0046] Based on the curvature abrupt changes of adjacent edge points on the edge of the superpixel region, the degree of curvature abrupt change in the edge region is determined. More specifically: a third-order polynomial is used to fit the edge curve of the superpixel region, denoted as the edge fitting curve; the curvature of the edge fitting curve of the superpixel region is obtained based on the second derivative of the edge fitting curve, and the absolute value of the difference in curvature between adjacent edge points on the edge fitting curve is calculated, denoted as the curvature abrupt change rate. ; ;in, Let be the curvature of the s-th edge point of the edge fitting curve; Let be the curvature of the (s-1)th edge point of the edge fitting curve.

[0047] Calculate the mean curvature abrupt change rate of all adjacent edge points on the edge fitting curve of all superpixel regions. This represents the degree of abrupt change in the curvature of the edge region. A greater degree of abrupt change indicates a more drastic change in the edge fitting curve, potentially leading to significant microscopic defects in the region. Each brightness image corresponds to a specific degree of abrupt change.

[0048] By analyzing the spatial distribution characteristics and local region properties of the LED backlight under test, the detection distortion value of the LED backlight under test in brightness detection is determined. Specifically, after analyzing the spatial distribution characteristics of the LED backlight under test to obtain the brightness disorder coefficient and analyzing the local region properties of the LED backlight under test to obtain the abrupt change degree, the detection distortion value of the LED backlight under test in brightness detection is determined by combining the brightness disorder coefficient and the abrupt change degree. The brightness disorder coefficient and the abrupt change degree are positively correlated with the detection distortion value. The larger the detection distortion value, the more significant the micro-bright spots or dark areas caused by the uneven spatial distribution of brightness in the LED backlight under test. Brightness non-uniformity amplifies the influence of sensor physical limitations, such as dynamic range or quantization noise, causing irreversible information degradation and loss of the fine structure of micro-bright spots or dark areas during brightness detection.

[0049] In some specific embodiments, for each frame of brightness image, the detection distortion value of each frame of brightness image of the LED backlight under test is determined by combining the brightness disorder coefficient and the degree of abrupt change.

[0050] Step S300: Based on the static frame sequence and the driving current timing signal, generate a brightness fluctuation index; combine the edge gradient direction change angle of the superpixel region with the brightness change of the neighboring region to construct a defect comprehensive factor; analyze the temporal correlation between the brightness fluctuation index and the defect comprehensive factor to determine the dynamic index of brightness anomaly.

[0051] Extract all superpixel regions from the brightness image of the LED backlight under test, and calculate the average pixel value within each superpixel region as the region brightness value.

[0052] Based on the principle that the brightness of the LED backlight under test is positively correlated with the driving current, when local areas have abnormal brightness due to defects such as bubbles in the LED encapsulation colloid or uneven phosphor coating, the correlation between the brightness of the LED backlight under test and the driving current will deviate significantly. The synchronously recorded drive current timing signal is retrieved, the drive current period T is identified, and one drive current period is taken as a complete cycle.

[0053] For each superpixel region in the brightness image of the LED backlight under test, extract the average brightness sequence of the same superpixel region from the brightness images of the static frame sequence within the complete period. ;in, The average brightness of the superpixel region c in the first frame brightness image within the complete period; The average brightness of the superpixel region c in the second frame brightness image within the complete period; This represents the average brightness of superpixel region c in the last frame of the brightness image within a complete period. This can be understood as dynamically tracking the superpixel region c in the first frame of the brightness image to obtain the average brightness of superpixel region c in each frame of the brightness image. The elements in the average brightness sequence are the average brightness of the same superpixel region in the brightness images of a static frame sequence arranged in temporal order. It should be noted that optical flow algorithms can be used to obtain the same superpixel region in different frames of brightness images, achieving tracking and matching of superpixel regions in different frames of brightness images. Optical flow algorithms are well-known techniques in this field and will not be elaborated upon here.

[0054] Synchronously acquire the drive current timing signal sequence within the corresponding complete cycle. ,in, The driving current timing signal corresponds to the moment of the first frame brightness image within the complete cycle; The driving current timing signal corresponds to the moment of the second frame brightness image within the complete cycle; This is the driving current timing signal corresponding to the last frame of the brightness image within the complete cycle; where the elements in the driving current timing signal sequence are driving current timing signals arranged in time sequence.

[0055] For each superpixel region, the correlation between the mean brightness sequence and the driving current timing signal is analyzed, and a correlation deviation weight is constructed. Specifically, for each superpixel region, the Pearson correlation coefficient between the mean brightness sequence and the driving current timing signal sequence is calculated, and a negative correlation mapping is performed on the Pearson correlation coefficient. The result of the negative correlation mapping is used as the correlation deviation weight. The closer the Pearson correlation coefficient between the two sequences is to 1, the more significant the brightness change in the current superpixel region is due to the driving current, indicating normal brightness performance. Conversely, the closer the Pearson correlation coefficient is to -1 or 0, the more likely the brightness change in the current superpixel region is affected by non-current factors, and the brightness may be abnormal due to structural defects in the LED backlight itself.

[0056] The Pearson correlation coefficient is negatively correlated and converted into a correlation deviation weight. The larger the correlation deviation weight, the more severe the decoupling between brightness and current.

[0057] For each superpixel region, the mean range of brightness is obtained from the mean brightness sequence of that region. This mean range represents the maximum shift in brightness within a complete cycle, reflecting the absolute intensity of sudden brightness jumps, such as flickering caused by poor contact.

[0058] By combining the mean range of brightness and the correlation deviation weight, a brightness fluctuation index is generated for each superpixel region.

[0059] In some embodiments, the product of the mean range of brightness and the correlation deviation weight is used as the brightness fluctuation index. The larger the brightness fluctuation index, the higher the degree of brightness anomaly in the superpixel region, and the less effective the driving current is at controlling the brightness of the superpixel region.

[0060] In some embodiments, the formula for calculating the luminance fluctuation index F is: ;in, The average brightness is extremely poor; The Pearson correlation coefficient is the relationship between the mean regional brightness sequence and the driving current timing signal sequence. The weighting factor is the correlation deviation weight. It should be noted that in the formula for calculating this brightness fluctuation index, the negative correlation mapping to the Pearson correlation coefficient is achieved by calculating the difference between the constant 1 and the Pearson correlation coefficient between the two sequences.

[0061] For the superpixel region with a high brightness fluctuation index on the LED backlight under test, the degradation of optical diffusion inside the superpixel region is further analyzed in order to construct a defect comprehensive factor.

[0062] Specifically, a defect synthesis factor is constructed by combining the abrupt change angle of the edge gradient direction of the superpixel region with the brightness change of the neighboring region.

[0063] For the superpixel region in the brightness image of the LED backlight under test, the Soble operator is used to calculate the brightness gradient of each pixel within the superpixel region, and the gradient direction of each pixel is obtained. It should be noted that since the pixel value of each pixel in the brightness image is the brightness value, the brightness gradient of a pixel is also the pixel gradient of that pixel.

[0064] Traverse all pixels within the superpixel region to obtain the gradient direction of each pixel. Use the angle formed by the gradient directions of adjacent pixels within the superpixel region as the edge gradient direction abrupt change angle.

[0065] The larger the abrupt change angle of the edge gradient direction, the more it indicates that the direction of light propagation changes abruptly at that location, such as scattering caused by bubbles.

[0066] The mean gradient direction abrupt change angle is obtained by fusing the angle values ​​of the edge gradient direction abrupt change angles between all adjacent pixels in the superpixel region. More specifically, the average value of the edge gradient direction abrupt change angles between all adjacent pixels in the superpixel region is calculated as the mean gradient direction abrupt change angle. A larger mean gradient direction abrupt change angle reflects abrupt changes in the scattering path of light at the microscale within the superpixel region, directly indicating a more severe local optical distortion caused by bubbles or coating defects.

[0067] Using the currently analyzed superpixel region as the central superpixel region, the information entropy of the average brightness of the central superpixel region and its neighboring regions is calculated as the brightness change of the neighboring regions. The neighboring regions are the superpixel regions adjacent to the central superpixel region. A smaller value for the brightness change of the neighboring regions indicates a highly concentrated brightness distribution in the neighborhood of the current central superpixel region, suggesting a higher probability of either being entirely bright or entirely dark. This indicates a discontinuous brightness distribution in the neighborhood of the central superpixel region, i.e., discontinuous light diffusion.

[0068] Then, by combining the mean gradient direction abrupt change angle and the brightness change in the neighborhood region, a defect comprehensive factor is constructed. Specifically, the product of the mean gradient direction abrupt change angle and the brightness change in the neighborhood region is used as the defect comprehensive factor.

[0069] The larger the value of the defect synthesis factor, the more likely it is that the superpixel region simultaneously exhibits microscopic optical ray scattering path distortion and macroscopic brightness distribution breakage, with the two working together to degrade the continuity of optical diffusion.

[0070] After obtaining the brightness fluctuation index and defect comprehensive factor for each superpixel region, the temporal correlation between the brightness fluctuation index and the defect comprehensive factor is analyzed to determine the dynamic index of brightness anomalies. First, based on temporal sequence, a brightness fluctuation index sequence and a defect comprehensive factor sequence are constructed. The elements in the brightness fluctuation index sequence represent the brightness fluctuation index of the same superpixel region in different frames of brightness images within a complete period; similarly, the elements in the defect comprehensive factor sequence represent the defect comprehensive factor of the same superpixel region in different frames of brightness images within a complete period.

[0071] The correlation between the brightness fluctuation index sequence and the defect comprehensive factor sequence is analyzed. When the correlation between the brightness fluctuation index sequence and the defect comprehensive factor sequence is positive, the Pearson correlation coefficient between the brightness fluctuation index sequence and the defect comprehensive factor sequence is used as a correction factor to correct the sum of the brightness fluctuation index and the defect comprehensive factor of the current superpixel region, thus obtaining the dynamic index of brightness anomaly.

[0072] More specifically: Calculate the Pearson correlation coefficient between the brightness fluctuation index series and the defect comprehensive factor series. The Pearson correlation coefficient The closer the value of is to 1, the stronger the positive correlation between brightness fluctuation and optical diffusion degradation in the time domain. This indicates a strong positive correlation between the Pearson correlation coefficient and the Pearson correlation coefficient. When the value is greater than 0, the brightness fluctuation index sequence and the defect comprehensive factor sequence are positively correlated. (This refers to a dynamic index of brightness anomalies in superpixel regions.) The calculation formula is: ;in, The brightness fluctuation index of the superpixel region; is the defect synthesis factor for the superpixel region; max is the function to maximize the value; is the Pearson correlation coefficient between the brightness fluctuation index sequence and the defect comprehensive factor sequence.

[0073] in, This indicates that negative correlations are ignored, and only positive correlations are retained as defect characteristics; this is the dynamic index of abnormal brightness. The larger the value, the stronger the correlation between brightness fluctuation and optical diffusion degradation in the time domain, indicating that the optical performance degradation of the superpixel region is more severe.

[0074] Step S400: Fuse the detected distortion value and the brightness anomaly dynamic index to generate a key spatiotemporal dynamic index; input the key spatiotemporal dynamic index and the brightness image into the trained neural network model to output the brightness detection index of the LED backlight under test.

[0075] By combining the detection distortion value obtained through static distribution anomaly detection and the brightness anomaly dynamic index obtained through multi-scale dynamic quantization, key spatiotemporal dynamic indicators for the brightness detection of the LED backlight under test are generated. Specifically, this involves fusing the detection distortion value and the brightness anomaly dynamic index to generate key spatiotemporal dynamic indicators. Specifically, a positive correlation mapping is performed on the brightness anomaly dynamic index to obtain a de-interference enhancement factor; the key spatiotemporal dynamic indicators are then determined by combining the detection distortion value and the de-interference enhancement factor.

[0076] In some embodiments, key spatiotemporal dynamic indicators The calculation formula is: ;in, The dynamic index of brightness anomalies in all superpixel regions on the brightness image of the LED backlight under test; is the detected distortion value of the brightness image of the LED backlight under test; ln is the natural logarithm function.

[0077] in, This is used to suppress random noise interference while preserving the impact of dynamic fluctuations, and to enhance the synergistic amplification effect of static defects and dynamic runaway in scenarios highlighting severe defects. Key spatiotemporal dynamic indicators The larger the value of , the more it indicates that the static defects of the LED backlight under test induce local scattering distortion. Static defects, such as colloidal bubbles, cause the driving current to lose its ability to regulate brightness. The brightness jump and the breakage of the optical diffusion path form a positive feedback loop, and the defect area enters an irreversible accelerated deterioration state.

[0078] The key spatiotemporal dynamic indicators and the brightness image are input into a trained neural network model, which outputs a brightness detection index for the LED backlight under test. This machine learning neural network can be a temporal convolutional network or a graph neural network. The trained neural network model is a multilayer perceptron regression model. Its input layer receives a feature vector composed of the key spatiotemporal dynamic indicators and basic brightness parameters, where the basic brightness parameters are the average brightness values ​​of the LED backlight brightness image. Through offline training, the neural network model learns the nonlinear mapping relationship between the concatenated feature vector and the backlight brightness detection index obtained from a high-precision photometer, and uses this brightness detection index as the ground truth label for training. The training set of the neural network model consists of: the key spatiotemporal dynamic indicators, the basic brightness parameters obtained from the LED backlight brightness image, and the brightness detection index. It should be noted that during network training, the brightness detection index in the training set is the light intensity value obtained from a high-precision photometer. In some embodiments, the brightness detection index can also be manually labeled data values. The key spatiotemporal dynamic indicators in the training set are obtained by analyzing the LED backlight brightness image. In this embodiment of the invention, the basic brightness parameter of the LED backlight is the average brightness of all pixels in the brightness image of the LED backlight.

[0079] Please see Figure 2 , Figure 2 This invention provides a schematic diagram of an LED backlight brightness detection device module, which includes: The optical acquisition module is used to acquire the static frame sequence of the brightness image of the LED backlight under test and the synchronous drive current timing signal. The data processing and analysis module, connected to the optical acquisition module, is used to receive and process the static frame sequence and the driving current timing signal. The processing includes: Acquire the static frame sequence of the brightness image of the LED backlight under test and the synchronous driving current timing signal; Based on the static frame sequence, a brightness disorder coefficient is constructed by the local pixel feature distribution of the gridded sub-window, and a detection distortion value is generated by combining the edge curvature mutation analysis of the superpixel region. Based on the static frame sequence and the driving current timing signal, a brightness fluctuation index is generated; a defect comprehensive factor is constructed by combining the edge gradient direction change angle of the superpixel region with the brightness change of the neighboring region; the temporal correlation between the brightness fluctuation index and the defect comprehensive factor is analyzed to determine the dynamic index of brightness anomaly. The detected distortion value and the brightness anomaly dynamic index are fused to generate a key spatiotemporal dynamic index; the key spatiotemporal dynamic index and the brightness image are input into a trained neural network model to output the brightness detection index of the LED backlight under test.

[0080] Alternatively, the transmission medium may be a wired link, such as, but not limited to, coaxial cable, fiber optic cable and digital subscriber line, or a wireless link, such as, but not limited to, wireless Fidelity (WIFI), Bluetooth and mobile device networks.

[0081] It should be noted that the device provided in the above embodiments is only an example of the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device can be divided into different functional modules to complete all or part of the functions described above.

[0082] Figure 3 This is a schematic diagram of the structure of a computer device provided in an embodiment of the present invention. For example, as shown... Figure 3 As shown, the computer device 500 includes: a memory 510, a processor 520, and a computer program 530 stored in the memory 510 and running on the processor 520, wherein when the processor 520 executes the computer program 530, the computer device can execute any of the LED backlight brightness detection methods described above.

[0083] Furthermore, embodiments of the present invention also protect an apparatus that may include a memory and a processor, wherein the memory stores executable program code, and the processor is used to call and execute the executable program code to execute an LED backlight brightness detection method provided in embodiments of the present invention.

[0084] In this embodiment of the invention, the device can be divided into functional modules according to the above method example. For example, each module can correspond to a separate function, or two or more functions can be integrated into one processing module. The integrated module can be implemented in hardware. It should be noted that the module division in this embodiment is illustrative and is only a logical functional division. In actual implementation, there may be other division methods.

[0085] When each module is divided according to its function, the device may also include a signal uploading module, a determination module, and an adjustment module. It should be noted that all relevant content of each step involved in the above method embodiments can be referenced from the functional descriptions of the corresponding functional modules, and will not be repeated here.

[0086] It should be understood that the apparatus provided in this embodiment of the invention is used to perform the above-described LED backlight brightness detection method, and therefore can achieve the same effect as the above-described implementation method.

[0087] When using integrated units, the device may include a processing module and a storage module. When applied to a device, the processing module can be used to control and manage the device's operations. The storage module can be used to support the device in executing program code, etc. The processing module may be a processor or a controller, which can implement or execute various exemplary logic blocks, modules, and circuits as described in this disclosure. The processor may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of Digital Signal Processing (DSP) and a microprocessor, etc., and the storage module may be a memory.

[0088] In addition, the device provided in the embodiments of the present invention may specifically be a chip, component or module. The chip may include a connected processor and a memory. The memory is used to store instructions. When the processor calls and executes the instructions, the chip can execute the LED backlight brightness detection method provided in the above embodiments.

[0089] This invention also provides a computer-readable storage medium storing computer program code. When the computer program code is run on a computer, the computer executes the aforementioned method steps to implement the LED backlight brightness detection method provided in the above embodiments.

[0090] This invention also provides a computer program product that, when run on a computer, causes the computer to perform the aforementioned steps to implement the LED backlight brightness detection method provided in the above embodiments.

[0091] In this invention, the apparatus, computer-readable storage medium, computer program product, or chip provided in the embodiments are all used to execute the corresponding methods described above. Therefore, the beneficial effects they achieve can be referred to the beneficial effects in the corresponding methods described above, and will not be repeated here. Through the above description of the embodiments, those skilled in the art will understand that, for the sake of convenience and brevity, only the division of the above functional modules is used as an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In the embodiments provided by this invention, it should be understood that the disclosed apparatus and method can be implemented in other ways.

[0092] The device embodiments described above are merely illustrative. For example, the division of modules or units is only a logical functional division. In actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0093] It should also be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or terminal device. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or terminal device that includes said element.

[0094] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0095] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0096] The above content is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the protection scope of the present invention.

Claims

1. A method for detecting the brightness of an LED backlight, characterized in that, The method includes the following steps: Acquire the static frame sequence of the brightness image of the LED backlight under test and the synchronous driving current timing signal; Based on the static frame sequence, a brightness disorder coefficient is constructed by the local pixel feature distribution of the gridded sub-window, and a detection distortion value is generated by combining the edge curvature mutation analysis of the superpixel region. Based on the static frame sequence and the driving current timing signal, a brightness fluctuation index is generated; a defect comprehensive factor is constructed by combining the edge gradient direction change angle of the superpixel region with the brightness change of the neighboring region; the temporal correlation between the brightness fluctuation index and the defect comprehensive factor is analyzed to determine the dynamic index of brightness anomaly. The detected distortion value and the brightness anomaly dynamic index are fused to generate a key spatiotemporal dynamic index; the key spatiotemporal dynamic index and the brightness image are input into a trained neural network model to output the brightness detection index of the LED backlight under test.

2. The LED backlight brightness detection method according to claim 1, characterized in that, The step of constructing a brightness disorder coefficient based on the local pixel feature distribution of a gridded sub-window, based on the static frame sequence, includes: A brightness disorder coefficient is constructed based on the local brightness range and global variance of the gridded sub-windows; wherein the local brightness range is the pixel range in each gridded sub-window; and the global variance is the variance of the pixel mean in all gridded sub-windows.

3. The LED backlight brightness detection method according to claim 1, characterized in that, The generation of detection distortion values ​​by combining edge curvature abrupt change analysis of superpixel regions includes: Obtain the superpixel region in the brightness image of each frame in a static frame sequence; The degree of curvature change in the edge region is determined based on the curvature abrupt changes of adjacent edge points on the edge of the superpixel region. For each frame of the brightness image, the detection distortion value of each frame of the brightness image of the LED backlight under test is determined by combining the brightness disorder coefficient and the degree of abrupt change.

4. The LED backlight brightness detection method according to claim 1, characterized in that, The step of generating a brightness fluctuation index based on the static frame sequence and the driving current timing signal includes: Construct a regional brightness mean sequence for each superpixel region in a static frame sequence within a complete period; the elements in the regional brightness mean sequence are the brightness mean of the same superpixel region in the brightness image of the static frame sequence arranged in time sequence; For each superpixel region, obtain the mean range of brightness in the mean brightness sequence of each superpixel region; For each superpixel region, the correlation between the region's average brightness sequence and the driving current timing signal is analyzed, and a correlation deviation weight is constructed. By combining the mean range of brightness and the correlation deviation weight, a brightness fluctuation index is generated for each superpixel region.

5. The LED backlight brightness detection method according to claim 4, characterized in that, For each superpixel region, the correlation between the region's average brightness sequence and the driving current timing signal is analyzed, and a correlation deviation weight is constructed, including: For each superpixel region, the Pearson correlation coefficient between the region's mean brightness sequence and the driving current time-series signal sequence is calculated, and the Pearson correlation coefficient is negatively correlated. The result of the negative correlation mapping is used as the correlation deviation weight.

6. The LED backlight brightness detection method according to claim 1, characterized in that, The defect synthesis factor, which combines the abrupt change angle of the edge gradient direction of the superpixel region with the brightness change of the neighboring region, includes: The angle formed by the gradient directions between adjacent pixels in the superpixel region is taken as the edge gradient direction abrupt change angle; The average gradient direction abruptness angle is obtained by fusing the angle values ​​of the edge gradient direction abruptness angle between all adjacent pixels in the superpixel region. Using the currently analyzed superpixel region as the central superpixel region, the information entropy of the average brightness of the central superpixel region and the neighboring regions is calculated as the brightness change of the neighboring regions; where the neighboring regions are the superpixel regions adjacent to the central superpixel region. A defect comprehensive factor is constructed by combining the mean gradient direction abrupt change angle and the brightness change in the neighborhood region.

7. The LED backlight brightness detection method according to claim 1, characterized in that, The analysis of the time-domain correlation between the brightness fluctuation index and the defect comprehensive factor, to determine the dynamic index of brightness anomalies, includes: Based on the time sequence, construct the brightness fluctuation index sequence and the defect comprehensive factor sequence; Analyze the correlation between the brightness fluctuation index sequence and the defect comprehensive factor sequence; When the correlation between the brightness fluctuation index sequence and the defect comprehensive factor sequence is positive, the Pearson correlation coefficient between the brightness fluctuation index sequence and the defect comprehensive factor sequence is used as a correction factor to correct the sum of the brightness fluctuation index and the defect comprehensive factor in the current superpixel region, thus obtaining a dynamic index of brightness anomaly.

8. The LED backlight brightness detection method according to claim 1, characterized in that, The process of fusing the detected distortion value with the brightness anomaly dynamic index to generate key spatiotemporal dynamic indicators includes: A positive correlation mapping is performed on the aforementioned brightness anomaly dynamic index to obtain the interference removal enhancement factor; By combining the detection distortion value and the interference removal enhancement factor, key spatiotemporal dynamic indicators are determined.

9. The LED backlight brightness detection method according to claim 1, characterized in that, The training set for the neural network model consists of: key spatiotemporal dynamic indicators, the average brightness of the LED backlight brightness image, and brightness detection indicators.

10. An LED backlight brightness detection device, characterized in that, The device includes the following modules: The optical acquisition module is used to acquire the static frame sequence of the brightness image of the LED backlight under test and the synchronous drive current timing signal. The data processing and analysis module, connected to the optical acquisition module, is used to receive and process the static frame sequence and the driving current timing signal. The processing includes: Acquire the static frame sequence of the brightness image of the LED backlight under test and the synchronous driving current timing signal; Based on the static frame sequence, a brightness disorder coefficient is constructed by the local pixel feature distribution of the gridded sub-window, and a detection distortion value is generated by combining the edge curvature mutation analysis of the superpixel region. Based on the static frame sequence and the driving current timing signal, a brightness fluctuation index is generated; a defect comprehensive factor is constructed by combining the edge gradient direction change angle of the superpixel region with the brightness change of the neighboring region; the temporal correlation between the brightness fluctuation index and the defect comprehensive factor is analyzed to determine the dynamic index of brightness anomaly. The detected distortion value and the brightness anomaly dynamic index are fused to generate a key spatiotemporal dynamic index; the key spatiotemporal dynamic index and the brightness image are input into a trained neural network model to output the brightness detection index of the LED backlight under test.

Images