Method and device for calibrating LED display screen

Abstract

The LED display system has an LED display screen coupled to a driving circuit. The driving circuit includes a scrambled pulse width modulation generator, a register and a memory. The scrambled pulse width modulation generator receives image data from an external source and, after certain compensation, is sent to the scrambled pulse width modulation generator to be distributed according to a set of new rules including compensated image data K. The compensated image data K can be an empirical value or obtained according to a formula using coefficients p and q, and the coefficients p and q can be obtained by calibration.

Description

[0001] This application is a continuation-in-part of U.S. patent application No. 15 / 945,497 filed on April 4, 2018, the contents of which are incorporated herein by reference in their entirety. Technical Field

[0002] The present invention generally relates to a method and apparatus for driving a display, and more particularly to a method and apparatus for compensating image data to improve the refresh rate and brightness uniformity of an LED display. Background Art

[0003] Modern LED (light-emitting diode) displays require higher grayscale to achieve higher color depth and higher visual refresh rates to reduce flicker. For example, the 16-bit grayscale of an RGB LED pixel allows 16-bit levels for the red, green, and blue LEDs (21 6 =65536). Such RGB LED pixels can display a total of 65536 3 colors. One of the commonly used methods to adjust the grayscale of an LED is pulse width modulation ("PWM"). Simply put, PWM generates a series of voltage pulses to drive an LED. When the pulse voltage is higher than the forward voltage of the LED, the LED is turned on. Therefore, when the pulse amplitude exceeds the threshold, the pulse duration (i.e., pulse width) of the PWM signal determines the on / off time of the LED. The percentage of the on time to the sum of the on time and the off time (i.e., the pulse width modulation period) is the duty cycle, which determines the brightness of the light-emitting diode. The configuration and operation of an exemplary LED display system (including LED topology, circuits, PWM engines, etc.) are described in detail in U.S. Patent No. 8,963,811 issued on February 24, 2015 and U.S. Patent Application 15 / 901,712 filed on February 21, 2018.

[0004] Another parameter of LED display is grayscale value, which is the brightness level of LED display. In 16-bit resolution LED display, grayscale value ranges from 0 (complete darkness) to 65535 (maximum brightness), corresponding to duty cycle from 0% to 100%. When the grayscale value is low, the brightness level of LED is low. Conversely, when the grayscale is high, the brightness level is also high. LED display often encounters performance problems at low grayscale values.

[0005] Another parameter of an LED display is its grayscale clock ("GCLK") frequency, which is related to the maximum number of GCLK cycles in a data frame and the refresh rate of the display. In addition, the frame rate refers to the number of times the video source inputs a full frame of new data to the display in one second. The refresh rate of an LED display is the number of times the LED display draws data per second. The refresh rate is equal to the frame rate multiplied by the number of segments.

[0006] One of the advantages of pulse width modulation (PWM) is the low power loss of the switching device. When the switch is off, there is almost no current. When the switch is on, there is almost no voltage drop across the switch. Therefore, the power loss in both cases is close to zero. On the other hand, pulse width modulation (PWM) is defined by the duty cycle, the switching frequency and the load characteristics. When the switching frequency is high enough, the pulse train can be smoothed and the average analog waveform can be restored. However, when the switching frequency is low, the off time of the light-emitting diode will be obvious and appear as flickering to the observer.

[0007] Scrambled pulse width modulation ("S-PWM") changes traditional pulse width modulation and achieves a higher visual refresh rate. To achieve this, S-PWM scrambles the on-time in one PWM cycle into multiple shorter PWM pulses, which drive each scan line in turn. In other words, the total grayscale value is scrambled into multiple PWM pulses within one PWM cycle. In traditional PWM schemes, there may be only one PWM pulse, causing the LED to light up continuously for a period of time, while the LED is not lit for the rest of the time. In contrast, S-PWM allows the LED to emit light in continuous short pulses during the PWM cycle so that the light pulses are more evenly distributed in the PWM cycle, avoiding or reducing flicker.

[0008] The number of GCLK cycles in one PWM cycle is equal to 2 raised to the power of the number of control bits:

[0009] Number of GCLK cycles = 2 control bits For example, 16-bit grayscale has 65536 GCLKs. Note that the number of GCLKs in one PWM cycle is equal to the grayscale value at its maximum brightness, that is, the maximum pulse width. In some S-PWMs, the total number of GCLKs can be divided into the MSB (most significant bit) and LSB (least significant bit) of the grayscale cycle. Each PWM cycle is divided into several segments (or sub-PWM cycles) according to the following formula:

[0010] Number of fragments = 2 least significant digits

[0011] For a video source with a 60 Hz frame rate and a PWM cycle length of 8000 GCLKs, the PWM cycle can be divided into 32 segments (LSB=5) so that each segment has a pulse duration of 250 GCLKs. Therefore, a total of 1600 GCLKs of grayscale values ​​can be distributed into 32 segments, 50 GCLKs per segment, potentially increasing the refresh rate by up to 32 times. However, when the PWM pulse duration (i.e., pulse width) in a segment is less than the time required to increase the LED voltage above the forward voltage, the LED will remain dark. U.S. Patent No. 9,390,647 provides a solution to extend the pulse duration by adding a fixed number of GCLKs to the pulse. However, such an S-PWM scheme results in a significant increase in light energy output at low brightness levels, which is explained in other parts of the present invention. Other technical solutions may require another power supply to provide additional drive current to extend the pulse duration, which increases the complexity and cost of the LED display electrical system.

[0012] Therefore, new systems and methods are needed to improve the image quality of LED display screens while eliminating the shortcomings of the existing technologies. Summary of the invention

[0013] An embodiment of the LED display system of the present invention includes an LED display screen coupled to a driving circuit. The driving circuit includes a scrambled pulse width modulation generator (S-PWM generator), a register and a memory. The S-PWM generator receives image data with a grayscale value of (X+K). X is the grayscale value of the data from an external image source, and K is a compensation value provided by the driving circuit.

[0014] According to one embodiment, the S-PWM generator distributes the grayscale value (X+K) to multiple segments according to the following rules:

[0015] When (X+K) is equal to or less than G0*S0,

[0016] S=ceil((X+K) / G0),R=mod(X+K,G0) (1)

[0017] When (X+K) is greater than G0*S0,

[0018] M=floor((X+K) / S0),L=mod(X+K,S0) (2)

[0019] In equations (1) and (2), G0 is the number of groups, S0 is the number of preset segments stored in the driving circuit, S is the number of output segments, where the pulse width of S-1 segments is G0 GCLKs, and the pulse width of the remaining segment is R.

[0020] In addition, L is the number of segments that receive pulse widths of M+1. The remaining S0-L segments receive pulse widths of M. Note that the unit of pulse width or grayscale value is GCLK. For example, a pulse width of M means that a pulse width has a time length of M GCLK.

[0021] The number of groups G0 can be predetermined empirically or obtained by calibrating the flicker of the LED display. It can be stored in a memory in the drive circuit. The compensation value K is related to the first set of calibration data obtained at high brightness and the second set of calibration data obtained at low brightness. For example, K = (floor (p*X) + q) - X, where p is derived from the first set of calibration data and q is derived from the second set of calibration data.

[0022] In some embodiments, the LED array in the light emitting diode (LED) display screen can be arranged in a common cathode configuration or a common anode configuration. The LED display screen can be a large wall display screen for indoor or outdoor use. The LED display screen can also be a micro display screen for handheld devices.

[0023] The present invention also provides an operating method of an LED display system. The LED display is coupled to a drive circuit having a scrambled pulse width modulation generator. Image data X is sent to the drive circuit. The image data X is compensated by multiplying the calibration coefficient p in a multiplier. The image data is further compensated by adding another constant q in an adder. Thus, a total compensation value K is added to X, so that the value of the compensated image data is (X+K).

[0024] The compensated image data (X+K) is then sent to the S-PWM generator, which scrambles the compensated image data into several segments, generates short pulse width modulated pulses, and sends them to the power supply or current source.

[0025] The present invention document further provides an image data compensation method for an LED display system. The LED display screen is driven by a driving circuit having an S-PWM generator. The driving circuit is connected to a video source. The image data input by the video source is X, and the compensated image data is floor(p*X)+q. The p value, or the q value, or both can be obtained by calibration. For example, the display screen is subjected to uniformity calibration at a high brightness level to determine the value of p, and is subjected to uniformity calibration at a low brightness level to determine the value of q. Alternatively, the value of p or the value of q, or both are predetermined empirical numbers.

[0026] For each individual LED in the LED display, the value of p, q, or both can be determined independently. Alternatively, q is a constant for the same color LEDs in the LED display, p is a constant for the same color LEDs in the LED display, or both are constants. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The teachings of the present invention can be readily understood by considering the following detailed description in conjunction with the accompanying drawings.

[0028] Figure 1 Schematic diagrams of prior art S-PWM schemes A and B are given.

[0029] Figure 2 The effect of the improved S-PWM scheme C is shown.

[0030] Figure 3 The operation of the prior art S-PWM scheme B is shown.

[0031] Figure 4 The operation of the improved S-PWM scheme C is shown.

[0032] Figure 5 A block diagram showing an LED display system of the present invention.

[0033] Figure 6 A flow chart of the iterative process for calibrating the LED array is given. DETAILED DESCRIPTION

[0034] Figure 1-6 The following descriptions are merely illustrative of embodiments of the present invention. It should be noted that from the following discussion, alternative embodiments of the structures and methods of the present invention will be readily identified as feasible alternatives that can be adopted without departing from the principles of the claimed invention.

[0035] Reference will now be made in detail to several embodiments of the present invention, examples of which are illustrated in the accompanying drawings. Please note that similar or like reference numerals may be used in the drawings, where applicable, and may represent similar or like functions. The drawings describe embodiments of the present invention for illustrative purposes only. It will be readily appreciated by those skilled in the art from the following description that alternative embodiments of the structures and methods described in the art may be used without departing from the inventive principles described in the art.

[0036] As used herein, the terms "coupled", "connected", and "connected" unless otherwise specified, may refer to either an indirect or direct electrical connection. Thus, if a first device is coupled or connected to a second device, that connection may be through a direct electrical connection or an indirect electrical connection via other devices or connections.

[0037] In the present invention, the term "low brightness" (i.e., low grayscale) generally refers to a case where the input signal length is short, for example, shorter than 4 times the LED boost time, or shorter than 3 times the LED boost time. Conversely, the term "high brightness" (i.e., high grayscale) refers to a case where the input signal length is long, for example, more than 4 times the LED boost time, or more than 6, 8, or 10 times the LED boost time.

[0038] Figure 1 Two existing S-PWM schemes are illustrated. The top panel shows that the grayscale value of one grayscale data input cycle is 320 GCLKs, that is, the total width of the PWM pulse in one grayscale data input cycle is 320 GCLKs. In S-PWM scheme A (such as Figure 1 In S-PWM scheme B (shown in the middle panel), 320 GCLKs are distributed in 32 segments (segment 0 to segment 31), with 10 GCLKs in each segment. Figure 1 In the S-PWM scheme B, an offset equal to N GCLKs is added to each PWM pulse to extend the PWM pulse width to N GCLKs, thereby obtaining a pulse with a width of (N+10) GCLKs. In the S-PWM scheme B, the extended PWM pulse width extends beyond the boost time to the forward voltage (Vf) of the LED, causing the LED to light up.

[0039] The present invention provides an innovative S-PWM scheme C. For the purpose of illustration, X is the grayscale value of the input image data in one grayscale input cycle; K is the compensation value added to the input image data; S0 is the number of segments; and G0 is the length of each segment.

[0040] In S-PWM scheme C, when (X+K) is equal to or less than G0*S0, S=ceil((X+K) / G0), R=mod(X+K,G0). S is the number of output segments, where the pulse width of S-1 segments is G0 GCLKs, and the pulse width of the remaining 1 segment is R, where R is a positive integer less than G0. The output segment used here is a segment with at least 1 GCLK pulse width, and a segment without an output pulse is called a "dark segment". Therefore, (S0-S) segments are dark segments.

[0041] Relatively speaking, when (X+K)=greater than G0*S0, M=floor(X+K) / S0, L=mod(X+K,S0). L is the number of segments with pulse width M+1, and the pulse width of the remaining S0-L segments is M.

[0042] This rule is applied to the scenario of distributing 1 to 320 GCLKs into 32 segments (S0=32), assuming that the number of groups is 8 GCLKs (G0=8), and the distribution of grayscale values ​​is shown in Tables 1 and 2. Table 1 shows the distribution of grayscale values ​​from 1 to 256 GCLKs (such as grayscale value ≦ S0*G0=256), and Table 2 shows the distribution of grayscale values ​​from 257 to 320 GCLKs.

[0043] Table 1

[0044]

[0045] Table 2

[0046]

[0047] Table 1 shows that when the gray value is less than or equal to S0*G0, the available gray data is first placed in a single segment until the PWM pulse width in the segment reaches G0, and then the remaining gray data is placed in another segment whose PWM pulse width is less than G0. Therefore, the maximum PWM pulse width in each segment is G0 (i.e., eight in this example). Therefore, at very low gray values, a single segment is preferentially filled until the segment has a pulse width of G0, while the remaining segments do not receive a signal and remain in a dark segment state. Note that when the gray value is equal to G0*S0, the pulse width of each segment is G0.

[0048] When the grayscale value is greater than G0*S0, the distribution pattern changes. As shown in Table 2, for the number of GCLKs exceeding G0*S0, 1 GCLK is allocated to each segment at a time until all 32 segments have (G0+1) GCLKs. Then, the excess GCLKs exceeding (G0+1)*S0 are allocated to each segment one GCLK at a time until all 32 segments have (G+2) GCLKs.

[0049] Therefore, in this embodiment, when the gray value is greater than S0*G0, the rule for allocating the gray value to each segment is the same as the traditional S-PWM scheme. However, when the gray value is lower, that is, less than S0*G0, this method maximizes the number of segments with at least G0 pulse width.

[0050] Figure 2 The effect of the innovative S-PWM scheme C is demonstrated. Figure 2Panels A, B, and C in show the output light energy (i.e., brightness) of a group of LEDs in response to input data length (i.e., input pulse width). Panel A shows the behavior of the LED without any compensation. The LED will not light up until the input pulse width exceeds a threshold level. Once the LED is lit, the energy output value of the LED generally increases linearly, but the rate of increase varies. Panel B shows the results of a first compensation for improving the brightness uniformity of the light-emitting diode at high brightness. Panel C shows the results of an embodiment of the current invention, which provides a second compensation in addition to the first compensation. After the second compensation, the LED emits light when the input pulse width is narrow.

[0051] Figure 3 for Figure 1 The light output of the LED in S-PWM scheme B is shown in the middle panel. Figure 3 In the bottom panel, when the PWM pulse in each segment is (t-1) GCLKs, the light energy output in one segment is e(t-1), and the total light energy output in 32 segments is 32*e(t-1). When the pulse width in the segment is extended by 1 GCLK to t GCLK values, the total light energy output in 32 segments is 32*e(t), as shown in Figure 3 As shown in the top panel of . Therefore, the difference in light output caused by one GCLK is 32*(e(t)-e(t-1)).

[0052] Figure 4 The light output of the light emitting diode in the inventive S-PWM scheme C of the present invention is shown. Figure 4 In the bottom panel of , the PWM pulse in segment 1 is t GCLKs, while each of the remaining segments receives (t-1) GCLKs and remains unlit. When the input PWM value increases by one GLCK, this GCLK is assigned to segment 2. Adding one GLCK to segment 2 is enough to light up the LED, as shown in Figure 4 As shown in the top panel of . Therefore, the difference in light output caused by one GCLK is 1*(e(t)-e(t-1)).

[0053] Since S-PWM Scheme B increases the PWM value in each of the 32 segments by the same number of GLCKs, the LED is either lit in all segments or remains off in all segments, not allowing fine-tuning at low brightness. In contrast, S-PWM Scheme C allows limited increases in the PWM values ​​of individual segments under certain conditions, so that the LED illuminates in at least some segments even at very low brightness levels. Therefore, S-PWM Scheme B results in a significant increase in light output, while S-PWM Scheme C allows fine-tuning of light output.

[0054] In some embodiments of the present invention, the compensation value K is obtained by calibration. For example, the calibration is performed by taking a photo and adjusting the brightness of a single LED in the LED display. Such calibration is usually performed at high brightness. Its purpose is to achieve uniformity of the brightness of the display. In such a calibration, each LED in the LED display receives the same image data - the same X value. Take a first photo of the LED display, which shows the change in LED brightness. The first data is added to the image data and sent to the LED. Take a second photo. Adjust the input image data and take photos until the brightness uniformity reaches a predetermined standard. This time, the compensation value K corresponding to the LED display is set.

[0055] In some embodiments, each LED pixel is an RGB LED pixel including a red LED, a blue LED, and a green LED, and each LED receives its own input image data X. i And obtain the coefficient p i , i = r, g or b. The coefficient p is obtained by calibrating each individual LED i , and then stored in a structure such as a lookup table in a memory (e.g., static random access memory). The memory can be built on the same chip as the driver circuit, or on a different chip coupled to the driver circuit chip. When needed, such as when the LED is powered on, the coefficients are retrieved to preload the calibration data into the registers in the driver circuit.

[0056] In another embodiment, calibration is performed under high brightness conditions to obtain a first set of coefficients p i , calibrated under low brightness conditions to obtain the second set of coefficients q i In some embodiments, the performance characteristic at low brightness is the flicker of the LED display, which can be monitored by visual inspection. Assume that, under low brightness conditions, a single LED receives input image data Xi and assigns coefficients q after a calibration process i Or, q i can be stored in a memory in the drive circuit. Therefore, the coefficient p i ,q i , or both are assigned to each individual LED.

[0057] For a 1920x1080 pixel color LED display, there can be up to six calibration data matrices - P r , P b , P g , Q r , Q b and Q g , each matrix is ​​of size 1920x1080. The elements in these matrices are p imnor q imn , where i = r, g, or b, m is the row number of the color pixel in the LED array, and n is the column number.

[0058] The p and p used here i and p imn can be used interchangeably. Each of them is a coefficient assigned to an LED in the LED array, while p refers to the general coefficient assigned to an LED. i Emphasize the color of the LED, p imn represents the color and position of the LED. The coefficient p may vary from one LED to another. Alternatively, the same coefficient p may be applied to all LEDs of the same color in the LED array. Similarly, q,q i, and q imn can also be used in the same way. In addition, p,p i , or p imn The matrix of is designated as the coefficient matrix P, and q,q i , or q imn The matrix of is designated as the coefficient matrix Q.

[0059] One of the calibration methods in the present invention is an open-loop process that uses two or more images of the LED array to derive coefficients p and q. First, calibration is performed at a high brightness level. The LED array has a uniform input data set for each red, green, or blue color. For example, when pwm_R=65535, pwm_G=0, and pwm_B=0, the red LEDs in the LED array are turned on. An image of the LED array is taken using a camera. The image is then processed to derive the brightness of each red LED to obtain an intensity matrix A. The lighting efficiency of each red LED is proportional to the corresponding matrix element. The red LED in row m and column n has A rmn The intensity of the red LEDs in the LED array is the average intensity (A). The corresponding coefficient p imn This can be derived from the following equation:

[0060] p imn =Armn / mean(A) (3)

[0061] The same process is repeated when the input to the LED array is green or blue. Therefore, red, green, and blue each have a corresponding coefficient p imn The coefficient matrix P is the coefficient p imn The matrix of .

[0062] The second step of the calibration process can be performed at low brightness levels. Again, for each red, green, or blue color, a unified data set is sent to all LEDs of the same color in the LED array but at a lower brightness. For example, the data set (pwm_R=200, pwm_G=0, pwm_B=0) lights up all red LEDs in the array at a low brightness. An image of the LED array is taken and the intensity matrix B is extracted from the image. Again, each red, green, and blue color is calibrated to obtain the corresponding matrix B.

[0063] Assume that the high brightness input data is X H , low brightness input data is X L . The following equation can be obtained:

[0064] B imn =(X L –q imn )*p imn *X L / X H (4)

[0065] And q imn =X L -B imn *X H / X L / p imn (5)

[0066] q imn is the coefficient obtained by the color LED at low intensity. The coefficient matrix q imn is designated as the matrix Q.

[0067] The above calculations assume that the imaging efficiency is the same at high and low brightness levels, which may not be entirely accurate. The term "imaging efficiency" as used herein refers to the ratio between the brightness extracted from the image of the LED and the actual brightness of the LED. To address this issue, in some embodiments, two different but lower brightness LED array images are used to obtain q imn Two images are taken at two different low brightness levels with input data X1 and X2, from which two intensity matrices B1 and B2 are derived. In this case, q can be obtained by solving two linear equations for each color LED imn , as shown below:

[0068] B1i mn =(X1-q imn )*p imn (6)

[0069] B2 imn =(X2-q imn )*p imn (7)

[0070] q imn =X1-B1 imn (X1-X2) / (B1 imn -B2 imn ) (8).

[0071] Another typical calibration method is a closed loop or iterative process that uses an imaging system to adjust the coefficient q of each LED to obtain uniform and consistent brightness. The process uses an adjustment cycle in which the value of q is modified to reduce the variation of the measured image until the displayed image is at or below a predetermined level. The details of this embodiment are described below, and reference Figure 6 .

[0072] like Figure 6 As shown, in S1, the coefficient matrix P is obtained from the image taken at a high brightness level according to equation (3) using the same process as described above. In S2, the coefficient matrix Q is either assigned initial values ​​or assigned the values ​​calculated in S8. In S3, a data matrix X' is applied to the LED array.

[0073] X'=(X+Q)*P (9)

[0074] The matrix X is a uniform matrix. Therefore, Q adjusts the matrix X. A uniform matrix as used here is a matrix where all elements have the same value.

[0075] After the adjustment, in S4, an image is captured and used to extract the intensity matrix B. According to equation (8), in S5, the difference between the matrix B and the uniform matrix is ​​calculated as an error matrix E.

[0076] E = B - mean (B) (10)

[0077] In formula (10), mean (B) is the matrix of average intensity values ​​of LEDs in the LED array, which is a uniform matrix.

[0078] In S6, the error matrix E is compared with a predetermined threshold. If E is equal to or less than the threshold, Q is output as the result of the calibration. The threshold can be a fraction of the mean (B), for example, 1% of the mean (B) or 0.5% of the mean (B). Each element q in the matrix Q is a coefficient obtained at low brightness or its equivalent. If E is greater than the threshold, the process continues to S7.

[0079] The compensation matrix C is calculated using the error matrix E in S7.

[0080] C=-k*E (11)

[0081] k is a constant that is less than the lighting efficiency of the LED array, ie, 50% of the average lighting efficiency of the LED array.

[0082] In S8, the compensation matrix C is added to the matrix Q to obtain the adjusted Q new .

[0083] Q new =Q+C=Qk*E=Qk*(B-mean(B)) (12)

[0084] Q new Assign it to Q in S2 to calculate the new input data matrix X' of the LED array and start a new iteration.

[0085] When the observed intensity of an LED pixel in the array is high, the corresponding element in the error matrix E of the LED array will also be large. Therefore, for this pixel (C = -k*E), the LED pixel needs a more significant compensation value c. Therefore, the input to the pixel q new The new data is smaller (Q new =Q+C), resulting in a lower output brightness. Further adjustments to Q are made using the new output intensity matrix B and a smaller k value. Therefore, each iteration requires modifying the input data to ensure that the brighter LED pixels have smaller input values, making the output image more uniform. Iterations continue until the change in the image reaches a predetermined level or there is no further significant reduction. The resulting Q matrix provides the q value used to calculate the compensation value k.

[0086] In some embodiments, for example, when the light emitted by the LEDs is consistent and uniform, it may not be necessary to apply a different q to each individual LED. Instead, all LEDs of the same color in the LED array can use one set of calibration data at low brightness, high brightness, or both. That is, at low brightness, all red LEDs use the same q r , all blue LEDs use the same q b , all green LEDs use the same q g , so that q imn (The three matrices q r ,q b and q g , each matrix size is 1920×1080) is reduced to three numbers. Regardless of the q used for low brightness r ,q b and q g At high brightness, all red LEDs can use the same p r , all blue LEDs use the same p b , all green LEDs use the same p g, so that p imn Three matrices p r 、p b and p g , each matrix size is 1920x1080) is reduced to three numbers. This simplification reduces the size of the memory required to store the calibration data. In these embodiments, the values ​​of q and p can be selected based on experience or obtained through calibration.

[0087] Both q and p are used to determine the compensation value K in order to achieve optimal compensation for the LED over the entire range of brightness levels.

[0088] In another embodiment of the present invention, the number of groups G0 and the number of segments S0 can be determined based on experience or obtained through calibration. S0 and G0 are stored in the driver circuit of the light-emitting diode display screen, for example, in a register. During the calibration process, an initial G0 value (e.g., 8) and / or an initial S0 value (e.g., 32) are set in the driver circuit, and the LED display screen is operated at different brightness levels, especially low brightness levels, to test performance characteristics such as flicker and brightness uniformity. G0 and S0 can be adjusted until the performance meets or exceeds the predetermined standard.

[0089] Note that the values ​​of pi, qi, G0 and S0 may be obtained by calibration of the LED array or may be directly predetermined without calibration, for example, based on experience.

[0090] Figure 5 is a block diagram of an exemplary LED display system of the present invention. A video source sends video data (8-bit, 10-bit or 12-bit) to an LED display system having an LED display screen having an LED array and an LED driving circuit. The video data is gamma corrected and converted to 16-bit data in a color depth converter. The 16-bit data stream enters a multiplier, where a first set of calibration data is combined into the data stream. The first set of calibration data is obtained under high brightness conditions, i.e., high brightness calibration. Assume that the input data is X i , high brightness calibration multiplies the input data by the calibration factor p i For example, the output data of the multiplier is equal to a Floor function: Floor(p i *X i ). This calibration adjusts the 16-bit data for pixel efficiency. Figure 2 The first compensation shown in panel B is an exemplary result of this high brightness calibration.

[0091] The data from the multiplier enters the adder where the second set of calibration data q iThe second set of calibration data is obtained under low brightness conditions, that is, low brightness calibration. Assume that the calibration data will be q i GCLK is added to N1, then the output data N2 of the adder is equal to (N1+q i ) or (floor(p i *X)+q i ). Therefore, the compensation value K i Determined by high brightness calibration and low brightness calibration, corresponding to Figure 2 The curve shown in panel C.

[0092] The calibrated image data (X+K) is sent to the S-PWM engine, which receives the preset number of segments S0 and the preset number of groups G0 from the register and generates a digital PWM signal. The digital PWM signal is sent to multiple power supplies. These power supplies in turn drive the scanning type LED display, which can be a common anode configuration or a common cathode configuration.

[0093] In the common anode configuration, the LED display has an array of RGB LED pixels arranged in rows and columns. The LED array has multiple common anode nodes. Each of the multiple common anode nodes operably connects the anodes of the same color LED pixels in a row to a corresponding scan switch. The cathodes of the LED pixels in the same column are connected to a power supply.

[0094] In the common cathode configuration, the LED array has a plurality of common cathode nodes. Each of the plurality of common cathode nodes operably connects the cathodes of LED pixels in a row to a corresponding scan switch. The anodes of LEDs of the same color in a column of LED pixels are connected to a current source.

[0095] In view of the contents of the present invention presented in the foregoing description and the related drawings, many modifications and other embodiments of the present invention listed here will occur to those skilled in the art. For example, the drive circuit can be used to drive an LED array in a common cathode or common anode configuration. The elements in the LED array can be monochrome LEDs or RGB units or any other form of LEDs. The drive circuit can be scaled up or down to drive LED arrays of various sizes. Multiple drive circuits can be used to drive multiple LED arrays in an LED display system. The elements in the drive circuit can be integrated on a single chip, or on multiple chips or printed circuit boards. In addition, the display screen can be any suitable display, including large outdoor display panels or small micro displays for mobile phones. Such changes are within the scope of the present invention. It should be understood that the present invention is not limited to the specific embodiments invented, and modifications and embodiments are intended to be included within the scope of the dependent claims.

Claims

1. A calibration method for an LED display system, characterized in that: include: obtaining a first image of the LED array at a high brightness level; deriving a coefficient matrix P using the first image, where P is a matrix consisting of matrix elements p, wherein each matrix element p corresponds to an LED in the LED array; obtaining one or more images of the LED array at one or more low brightness levels; and deriving a coefficient matrix Q using one or more images, wherein each matrix element q corresponds to an LED in the LED array, wherein obtaining Q comprises: Step a: Apply the image data matrix X′ to the LED array, where X′=(X+Q)*P, and X is a uniform matrix; Step b: acquiring an image of the LED array and extracting a brightness intensity matrix from the image; Step c: Obtain an error matrix E, where E=B-mean(B), where mean(B) is a uniform matrix representing the average brightness intensity of the LED array; Step d: Get the new matrix Q new , where Q new =Qk*E, k is a constant; Step e: Assign Q to Q new , repeat steps a to e, and when the error matrix E is equal to or less than the threshold, output Q as the result of the calibration process.

2. The method of claim 1, wherein p is a ratio between the intensity of the corresponding LED and the average intensity of LEDs in an LED array of the same color.

3. The method according to claim 2, wherein q = X L -B*X H / X L / p, where X L is the image data input at a high brightness level, X L is image data input at a low brightness level, and B is the brightness intensity of the LED extracted from the LED image taken at the low brightness level.

4. The method according to claim 2, wherein q = X1-B1(X1-X2) / (B1-B2), wherein B1 is the brightness intensity of the LED extracted from the LED image taken at a first low brightness level, and B2 is the brightness intensity of the LED extracted from the LED image taken at a second low brightness level. The method of claim 1 , wherein k is a constant that is less than a lighting efficiency of the LED array.

6. The method of claim 1, wherein the LED array is an RGB LED array.

Images