Liquid crystal display device, method for controlling a liquid crystal display device, and program

Abstract

To enable the display of images with the desired contrast. [Solution] The liquid crystal display device comprises a liquid crystal panel, an input means for inputting data of a first image, a backlight module capable of changing the luminescence brightness for each of several divided regions, a generation means for inputting data generated from the first image into a neural network to generate the luminescence intensity of each divided region of the backlight module, an adjustment means for adjusting the generated luminescence intensity, a calculation means for calculating a correction value to correct the first image based on the adjusted luminescence intensity, a correction means for correcting the first image to a second image based on the calculated correction value, a luminescence control means for controlling the luminescence intensity of each divided region of the backlight module based on the adjusted luminescence intensity, and a display control means for controlling the transmittance of the liquid crystal panel based on data of the second image.

Description

[Technical Field]

[0001] The present invention relates to a liquid crystal display device in which the luminescence brightness of the backlight module can be changed. [Background technology]

[0002] There is a demand for higher contrast in display devices that show images with a relatively wide dynamic range, such as HDR (High Dynamic Range) images. One example of such a display device is a liquid crystal display (LCD) device.

[0003] To reduce black level issues and improve contrast in LCD displays, a technique called local dimming is generally used. Local dimming is a technique that reduces black level issues by controlling the luminescence of the backlight module for each divided area. LCD displays not only HDR images but also images with a standard dynamic range, such as SDR (Standard Dynamic Range) images. Therefore, it is desirable for LCD displays to be able to adjust the display contrast.

[0004] One technique for adjusting the display contrast in an LCD device is to adjust the brightness of the backlight module in the dark areas. Patent Document 1 discloses a technique for changing the brightness of the backlight module according to the size of the dark areas. [Prior art documents] [Patent Documents]

[0005] [Patent Document 1] Japanese Patent Publication No. 2015-176137 [Overview of the project] [Problems that the invention aims to solve]

[0006] However, the method disclosed in Patent Document 1 had the problem that it was not possible to display an image with the desired contrast.

[0007] Therefore, the present invention aims to provide a technology that allows control of the light emission intensity of a backlight module so that an image can be displayed with a desired contrast. [Means for solving the problem]

[0008] The liquid crystal display device comprises a liquid crystal panel, an input means for inputting data of a first image, a backlight module that irradiates the liquid crystal panel with light and whose luminescence brightness can be changed for each of a plurality of divided regions, a generation means for inputting data generated from the first image into a neural network to generate the luminescence intensity of each divided region of the backlight module, an adjustment means for adjusting the luminescence intensity generated by the generation means, a calculation means for calculating a correction value for correcting the first image based on the luminescence intensity adjusted by the adjustment means, a correction means for correcting the first image to a second image based on the correction value calculated by the calculation means, a light emission control means for controlling the luminescence intensity of each divided region of the backlight module based on the luminescence intensity adjusted by the adjustment means, and a display control means for controlling the transmittance of the liquid crystal panel based on data of the second image. [Effects of the Invention]

[0009] According to the present invention, an image can be displayed with a desired contrast. [Brief explanation of the drawing]

[0010] [Figure 1] This is a block diagram showing the functional blocks of a liquid crystal display device according to Example 1. [Figure 2] This is a flowchart of the parameter application process according to Example 2. [Figure 3] This is an example of the settings screen for the local dimming model according to Example 1. [Figure 4] This is an example of the display contrast setting screen for Example 1. [Figure 5] It is a diagram showing an example of a divided area of a backlight according to Example 1. [Figure 6] It is a diagram showing an example of a brightness estimation point of a backlight according to Example 1. [Figure 7] [Figure 8] It is a block diagram showing a functional block of an LD feature quantity generation unit according to Example 1. [Figure 9] It is an example of a checker pattern image. [Figure 10] It is an example of a graph related to local dimming according to a specific rule. [Figure 11] It is an example of a graph related to local dimming by the machine learning model of Example 1. [Figure 12] It is a block diagram showing a functional block of a liquid crystal display device according to Examples 2 and 3. [Figure 13] It is an example of a patch image. [Figure 14] It is a flowchart of a halo evaluation unit according to Example 2. [Figure 15] It is a flowchart of a halo evaluation unit according to Example 3.

Mode for Carrying Out the Invention

[0011] Hereinafter, embodiments of the present invention will be described with reference to the drawings. The technical scope of the present invention is determined by the scope of claims and is not limited to the embodiments illustrated below. Also, not all combinations of features described in the embodiments are essential to the present invention. The content described in this specification and the drawings is illustrative and should not be regarded as limiting the present invention. Various modifications are possible based on the gist of the present invention, and they are not excluded from the scope of the present invention. That is, all configurations combining each embodiment and its modification examples are also included in the present invention.

[0012] [Example 1] ​The following describes Embodiment 1 of the present invention. Figure 1 is a block diagram showing the functional blocks of the liquid crystal display device 100 according to Embodiment 1. The liquid crystal display device 100 includes an image input / conversion unit 101, a setting input unit 102, a local dimming control unit 103, a liquid crystal panel control unit 104, a liquid crystal panel 105, a backlight control unit 106, and a backlight module 107.

[0013] The image input / conversion unit 101 acquires image data (image data) from an external source. Specifically, the image input / conversion unit 101 has an input interface such as SDI (Serial Digital Interface) and inputs image data into the liquid crystal display device 100 from an external source via the input interface. The image input / conversion unit 101 then performs conversion processing such as grayscale conversion and signal format conversion on the acquired (input) image data and outputs the converted image data as the input image.

[0014] The gradation conversion is, for example, a gradation conversion using a one-dimensional lookup table (1D-LUT), and is a gradation conversion corresponding to the gamma value (panel gamma) of the liquid crystal panel 105. Here, we consider the case where the gamma characteristics (correspondence between gradation value and brightness; gradation characteristics) of the externally acquired image data are linear, where brightness increases linearly with increasing gradation value, and the panel gamma is 2.0. In this case, gradation conversion is performed using the inverse gamma of the panel gamma (i.e., 1 / 2.0). As a result, the acquired image data (image data with linear characteristics) is converted into image data with a gamma specification in which brightness is proportional to the gradation value raised to the power of 1 / 2.0. Note that the conversion processing in the image input / conversion unit 101 is not limited to gradation conversion using a 1D-LUT, but may also include conversion processing using a three-dimensional lookup table (3D-LUT), gain adjustment, offset adjustment, matrix conversion, etc.

[0015] Signal format conversion is the process of converting the signal format of image data, for example, from YCbCr or XYZ to RGB. Note that the signal formats before and after conversion are not limited to YCbCr, XYZ, and RGB.

[0016] Furthermore, the image input / conversion unit 101 synthesizes the menu screen generated by the setting input unit 102 into the image data.

[0017] The setting input unit 102 outputs setting values ​​based on the setting operation to the parameter application unit 10302. The setting operation is, for example, a button operation on the liquid crystal display device 100. The setting input unit 102 generates a menu screen based on the button operation and outputs the menu screen to the image input / conversion unit 101. The menu screen generated by the setting input unit 102 is, for example, the screen shown in Figures 3 and 4. Figure 3 is a menu screen for switching trained models. When a trained model switching operation is performed on the menu screen in Figure 3, the setting input unit 102 outputs setting values ​​indicating the trained model (for example, an identifier and value indicating the trained model) to the parameter application unit 10302. Figure 4 is a menu screen for adjusting the contrast. When a contrast adjustment operation is performed on the menu in Figure 4, the setting input unit 102 outputs setting values ​​indicating the contrast (for example, an identifier and value indicating the contrast) to the parameter application unit 10302. The setting operation in the setting input unit 102 is not limited to button operations. The setting input unit 102 may, for example, obtain setting values ​​via HTTP (Hyper Text Transfer Protocol) over a network through a server (not shown). In this case, the setting input unit 102 may also send a web page of the setting menu to an operating terminal (not shown) over a network through a server (not shown).

[0018] The local dimming control unit 103 consists of a trained model holding unit 10301, a parameter application unit 10302, a pre-processing unit 10303, a backlight control value generation unit 10304, a post-processing unit 10305, a backlight brightness estimation unit 10306, a correction coefficient generation unit 10307, ​​and an image correction unit 10308.

[0019] The trained model holder 10301 holds a model (trained model) that has been trained to convert image features into backlight control values. The backlight control values ​​are control values ​​for controlling the backlight module 107. The trained model is a set of parameters applied to the neural network that converts image features into backlight control values, and includes weights and biases between neurons that make up the neural network. Note that the trained model held in the trained model holder 10301 may be one that converts an input image into backlight control values. The trained model held in the trained model holder 10301 is applied to the backlight control value generation unit 10304 via the parameter application unit 10302.

[0020] The trained model held in the trained model holding unit 10301 is associated with a target contrast value (target contrast) used during training. The target contrast associated with the trained model is equivalent to the ratio of the maximum brightness to the minimum brightness of the liquid crystal display device 100 when the trained model is applied to the backlight control value generation unit 10304. Here, the maximum brightness of the liquid crystal display device 100 is, for example, the brightness when displaying a completely white image, and the minimum brightness of the liquid crystal display device 100 is, for example, the brightness when displaying a completely black image.

[0021] The parameter application unit 10302 applies parameters to the backlight control value generation unit 10304 and the post-processing unit 10305 based on the setting values ​​output from the setting input unit 102. The parameter application process in the parameter application unit 10302 will be explained with reference to the flowchart in Figure 2.

[0022] In S11, the parameter application unit 10302 acquires the setting value output from the setting input unit 102.

[0023] In S12, the parameter application unit 10302 determines whether or not a setting value indicating a trained model was obtained in S11. If a setting value indicating a trained model was obtained in S11, in S13, the parameter application unit 10302 applies the trained model held in the trained model holding unit 10301 to the backlight control value generation unit 10304. The setting value indicating a trained model is specified by the user, for example, via the menu screen shown in Figure 3. If "Model 1" is selected, the parameter application unit 10302 applies a trained model with a target contrast of 1 million to the backlight control value generation unit 10304, and if "Model 2" is selected, it applies a trained model with a target contrast of 2000.

[0024] In S14, the parameter application unit 10302 sets (initializes) the target contrast value associated with the learned model applied to the backlight control value generation unit 10304 in S13, as the contrast value (display contrast) of the display image of the liquid crystal display device 100.

[0025] If a setting value indicating a trained model is not obtained in S11, in S15, the parameter application unit 10302 determines whether or not a setting value indicating contrast was obtained in S11. If a setting value indicating contrast was obtained in S11, in S16, the parameter application unit 10302 sets the setting value obtained in S11 as the display contrast value. The setting value indicating contrast is specified by the user, for example, via the menu screen shown in Figure 4.

[0026] In S17, the parameter application unit 10302 calculates the parameters to be applied to the post-processing unit 10305. The parameters to be applied to the post-processing unit 10305 are the gain value, the offset value, and the lower limit clipping value. The parameter application unit 10302 calculates the gain value, offset value, and lower limit clipping value to be applied to the post-processing unit 10305 based on the panel contrast, target contrast, and display contrast. The panel contrast is the contrast value of the liquid crystal panel 105. The target contrast is the contrast value associated with the learned model applied to the backlight control value generation unit 10304. The display contrast is the contrast value set in S14 or S16.

[0027] In S17, the parameter application unit 10302 calculates the gain value Ga, offset value Oa, and lower limit clip value Bmin to be applied to the post-processing unit 10305 according to the following formula (1). In the following formula (1), Cp represents the panel contrast, Ct represents the target contrast, and Cd represents the display contrast.

number

[0028] For example, if the panel contrast Cp is 1000, the target contrast Ct is 1 million, and the display contrast Cd is 10,000, then the gain value Ga, the offset value Oa, and the lower limit clipping value Bmin will be the values ​​shown in the following formula (2).

number

[0029] The parameters that the parameter application unit 10302 applies to the post-processing unit 10305 are not limited to gain value, offset value, and lower limit clipping value; other parameters such as 1D-LUT can also be applied.

[0030] In S18, the parameter application unit 10302 applies the parameters calculated in S17 to the post-processing unit 10305.

[0031] In the flowchart of Figure 2, steps S12-S14 describe an example in which the parameter application unit 10302 switches the trained model and initializes the display contrast value, but this process may be omitted. In this case, for example, the parameter application unit 10302 may also apply to the backlight control value generation unit 10304 a trained model that associates the display contrast set in S16 with the target contrast value closest to it.

[0032] The preprocessor 10303 generates feature quantities (image feature quantities) based on the input image output from the image input / conversion unit 101. In Embodiment 1, as shown in Figure 5, the preprocessor 10303 generates image feature quantities by calculating the maximum grayscale value and average grayscale value of the input image corresponding to multiple divided regions. That is, if the number of horizontal divisions of the backlight module 107 is n and the number of vertical divisions is m, the preprocessor 10303 generates n × m two-dimensional data as image feature quantities. The preprocessor 10303 then outputs the generated image feature quantities to the backlight control value generation unit 10304.

[0033] The backlight control value generation unit 10304 generates backlight control values ​​based on image features output from the preprocessing unit 10303. The backlight control values ​​are control values ​​for controlling the luminescence brightness of the backlight module 107 via the backlight control unit 106. As shown in Figure 5, the backlight module 107 has multiple divided regions that constitute the display surface, each of which has multiple light sources, and the luminescence brightness can be changed for each divided region. The light sources of the backlight module 107 are not particularly limited, but for example, they are LEDs (Light Emitting Diodes). The backlight control value generation unit 10304 generates backlight control values ​​for each divided region. The backlight control value generation unit 10304 then outputs the backlight control values ​​to the postprocessing unit 10305.

[0034] In Example 1, the backlight control value generation unit 10304 is composed of at least a neural network that takes image features as input and outputs backlight control values. When the horizontal division of the backlight module 107 is n and the vertical division is m, the neural network of the backlight control value generation unit 10304 has a structure that inputs and outputs n × m two-dimensional data. Furthermore, the neural network of the backlight control value generation unit 10304 has a structure to which a trained model held by the trained model holding unit 10301 can be applied. Note that the network structure of the neural network of the backlight control value generation unit 10304 is not limited as long as it can generate backlight control values. In addition, the input to the neural network of the backlight control value generation unit 10304 is not limited to image features, but may also be the input image.

[0035] The backlight control value generation unit 10304 may include a configuration that generates backlight control values ​​based on specific rules in addition to a neural network. Furthermore, the backlight control value generation unit 10304 may include multiple neural networks or specific rules for generating backlight control values. In this case, the backlight control value generation unit 10304 can switch between neural networks or specific rules.

[0036] The post-processing unit 10305 adjusts the backlight control value output from the backlight control value generation unit 10304 based on the parameters output from the parameter application unit 10302. The parameters output from the parameter application unit 10302 are the gain value, offset value, and lower limit clip value. The post-processing unit 10305 calculates the adjusted backlight control value Badj according to the following formula (3). In the following formula (3), Borg represents the backlight control value output from the backlight control value generation unit 10304, Ga is the gain value, Oa is the offset value, Bmin is the lower limit clip value, and MAX represents a function that selects the maximum value. Badj = MAX(Bmin, Borg × Ga + Oa) ... Formula (3)

[0037] In the liquid crystal display device 100, when the panel contrast is 1000:1 and the backlight control value is 1.0, a full white image is displayed at 1000 nits and a full black image at 1 nit. Here, in the liquid crystal display device 100, when the backlight control value is 0.001, a full black image is displayed at 1 nit × 0.001 = 0.001 nits. The backlight control value generation unit 10304 outputs a backlight control value of 1.0 when displaying a full white image and 0.001 when displaying a full black image by applying a trained model with a target contrast of 1,000,000:1. In this case, the liquid crystal display device 100 displays a full white image at 1000 nits × 1.0 = 1000 nits and a full black image at 1 nits × 0.001 = 0.001 nits. That is, the contrast of the displayed image (display contrast) of the liquid crystal display device 100 is 1000 nits ÷ 0.001 nits = 1,000,000 (1,000,000).

[0038] Here, when the gain value Ga, offset value Oa, and lower limit clipping value Bmin shown in equation (2) are applied, the adjusted backlight control value Badj becomes as shown in equation (4). The gain value Ga, offset value Oa, and lower limit clipping value Bmin shown in equation (2) are parameters applied to the post-processing unit 10305 in order to adjust the display contrast from 1,000,000:1 to 10,000:1. Badj = MAX(0.1, Borg × 0.900900901 + 0.099099099) ... Formula (4) When displaying a full white image on the liquid crystal display device 100 at 1000 nits, the backlight control value Borg before adjustment is 1.0, so the backlight control value Badj after adjustment is MAX(0.1, 1.0 × 0.900900901 + 0.099099099) = 1.0. When displaying a full black image on the liquid crystal display device 100 at 0.001 nits, the backlight control value Borg before adjustment is 0.001, so the backlight control value Badj after adjustment is MAX(0.1, 0.001 × 0.900900901 + 0.099099099) = 0.1. In other words, by adjusting the backlight control value according to formula (4), the liquid crystal display device 100 will display a full white image at 1000 nits × 1.0 = 1000 nits and a full black image at 1 nit × 0.1 = 0.1 nits. In this way, by applying the parameters shown in formula (2) to the post-processing unit 10305, the display contrast of the liquid crystal display device 100 changes from 1,000,000:1 to 1,000 nit ÷ 0.1 nit = 10,000 (10,000:1).

[0039] Furthermore, the post-processing unit 10305 can disable the adjustment of the backlight control value by, for example, applying a gain value Ga of 1.0 and an offset value Oa of 0.0. Alternatively, the post-processing unit 10305 can also disable the adjustment of the backlight control value by providing a pass-through process.

[0040] The post-processing unit 10305 then outputs the adjusted backlight control value to the backlight brightness estimation unit 10306 and the backlight control unit 106.

[0041] The backlight brightness estimation unit 10306 performs an estimation calculation of the brightness of the light irradiated from the backlight module 107 onto the liquid crystal panel 105 (backlight brightness) based on the backlight control value output from the post-processing unit 10305. The backlight brightness estimation calculation estimates (calculates) the backlight brightness based on the backlight control value of each light source (each divided region) and the brightness distribution model of the light emitted from the light source (the part of the backlight module 107 corresponding to the divided region). Here, the backlight brightness estimation unit 10304 estimates (calculates) the backlight brightness for each brightness estimation point discretely arranged within the display surface as shown in Figure 6(A). Specifically, according to the following formula (5), the backlight brightness is estimated (calculated) B ij and weight W ij The backlight brightness L is calculated by performing a sum-of-products operation. In the following formula (5), n represents the number of horizontal divisions of the backlight module, and m represents the number of vertical divisions of the backlight module. Also, B ij The backlight control values ​​are for vertical position i and horizontal position j, W ij This is the backlight control value B ij The weight applied, L, represents the backlight brightness at the brightness estimation point. The backlight lit in each divided region decreases in intensity as the distance increases. Therefore, the backlight control value B ij Weight W applied to it ij The closer the distance from the brightness estimation point, the greater the value.

number

[0042] In the example shown in Figure 6(A), one luminance estimation point is placed for one divided region of the backlight module 107, but this is not limited to this. For example, points may be placed at the four corners of the divided region as shown in Figure 6(B), or one point may be placed for each of the four divided regions as shown in Figure 6(C). The backlight luminance estimation unit 10306 interpolates the backlight luminance between the luminance estimation points and scales it to the resolution of the input image. Therefore, when luminance estimation points are placed as shown in Figure 6(B), the number of luminance estimation points increases compared to Figure 6(A), suppressing interpolation errors but increasing computational load. On the other hand, when luminance estimation points are placed as shown in Figure 6(C), the number of luminance estimation points decreases compared to Figure 6(A), suppressing computational load but increasing interpolation errors. Methods for scaling backlight luminance include, for example, bicubic interpolation and bilinear interpolation, but are not limited to these methods as long as they can be scaled to the resolution of the input image. The backlight luminance estimation unit 10306 then outputs the estimated backlight luminance to the correction coefficient generation unit 10307. Alternatively, the backlight brightness estimation unit 10306 may output discretely arranged backlight brightness values ​​to the correction coefficient generation unit 10307 without scaling the calculated backlight brightness, as shown in Figure 6(A). In this case, the correction coefficient generation unit 10307 scales the calculated correction coefficient to the resolution of the input image.

[0043] The correction coefficient generation unit 10307 generates a correction coefficient to be applied to the input image output from the image input / conversion unit 101, based on the backlight brightness output from the backlight brightness estimation unit 10304. In Example 1, the correction coefficient Gt is calculated using the reciprocal of the backlight brightness L (a value normalized to 0.0 to 1.0) according to the following formula (6). For example, if the backlight brightness is reduced to 1 / 3, the reduction in backlight brightness can be compensated for by the image by increasing the input image by its reciprocal, i.e., by 3 times. Note that the method for calculating the correction coefficient is not limited to the reciprocal of the backlight brightness.

number

[0044] The image correction unit 10308 generates (calculates) the pixel values ​​of the corrected image by multiplying the correction coefficient output from the correction coefficient generation unit 10307 by the pixel values ​​of the input image output from the image input / conversion unit 101. In Example 1, the RGB values ​​(Vrc, Vgc, Vbc), which are the pixel values ​​of the corrected image, are generated by multiplying the RGB values ​​(R, G, B) = (Vr, Vg, Vb), which are the pixel values ​​of the input image, by the correction coefficient Gt according to the following formula (7). The image correction unit 10308 then outputs the generated corrected image. Vrc = Vr × Gt Vgc = Vg × Gt Vbc = Vb × Gt ...Formula (7)

[0045] The liquid crystal panel control unit 104 controls the transmittance of the liquid crystal panel 105 (transmittance distribution within the display surface) based on the corrected image output from the image correction unit 10308, so that the image based on the corrected image is displayed on the liquid crystal panel 105.

[0046] The liquid crystal panel 105 is controlled by the liquid crystal panel control unit 104 and displays an image on its display surface.

[0047] The backlight control unit 106 controls the luminescence brightness of the backlight module 107 (the light source of the backlight module 107) according to the backlight control value output from the post-processing unit 10305. For example, the backlight control unit 106 determines the duty cycle of PWM (Pulse Width Modulation) control according to the backlight control value, and controls the luminescence brightness of the backlight module 107 by PWM control at the determined duty cycle.

[0048] The backlight module 107 illuminates the back of the liquid crystal panel 105 with light. As described above, the luminescence brightness of the backlight module 107 is adjustable. In Embodiment 1, multiple divided regions constituting the display surface are pre-set, and the backlight module 107 has multiple light sources corresponding to each of the multiple divided regions, and the luminescence brightness can be changed for each divided region.

[0049] As explained with reference to Figure 1, in the liquid crystal display device 100 of Embodiment 1, the backlight module 107 can be controlled by applying a trained model to the neural network of the backlight control value generation unit 10304.

[0050] Figure 7 is a block diagram showing the functional blocks of the model learning unit 200 according to Embodiment 1. The model learning unit 200 includes an image input / conversion unit 201, a preprocessing unit 202, a backlight control value generation unit 203, a first setting unit 204, an LD feature generation unit 205, a second setting unit 206, a target feature generation unit 207, an error calculation unit 208, and a trained model output unit 209. The model learning unit 200 is, for example, a program that runs on a computer, but is not limited to that, and may also operate as a functional block constituting the liquid crystal display device 100.

[0051] The image input / conversion unit 201 acquires image data (image data) from an external source. Specifically, the image input / conversion unit 101 acquires image data from a group of files stored in the computer's memory, but is not limited to this; image data may also be input via an input interface such as SDI. The image input / conversion unit 101 can then perform the same conversion processing on the acquired (input) image data as the image input / conversion unit 101 of the liquid crystal display device 100, and outputs the converted image data as the input image. However, the image input / conversion unit 201 may also acquire (input) pre-converted image data. In this case, the image input / conversion unit 201 does not need to perform image data conversion processing.

[0052] The preprocessor 202 generates feature quantities (image feature quantities) based on the input image output from the image input / conversion unit 201. Specifically, the preprocessor 202 generates image feature quantities in the same way as the preprocessor 10303 of the liquid crystal display device 100. That is, if the number of horizontal divisions of the backlight module 107 of the liquid crystal display device 100 is n and the number of vertical divisions is m, the preprocessor 202 generates n × m two-dimensional data as image feature quantities. The preprocessor 202 then outputs the generated image feature quantities to the backlight control value generation unit 203, the LD feature quantity generation unit 205, the target feature quantity generation unit 207, and the error calculation unit 208, respectively.

[0053] The backlight control value generation unit 203 consists of a neural network that takes image features output from the preprocessing unit 202 as input and outputs backlight control values. The neural network of the backlight control value generation unit 203 has the same structure as the neural network of the backlight control value generation unit 10304 of the liquid crystal display device 100. That is, if the horizontal division number of the backlight module 107 of the liquid crystal display device 100 is n and the vertical division number is m, the neural network of the backlight control value generation unit 203 has a structure that inputs and outputs n × m two-dimensional data. The backlight control value generation unit 203 takes the parameters generated by the error calculation unit 208 as input and reflects them in the neural network. Then, the backlight control value generation unit 203 outputs the backlight control values ​​generated by the neural network to the LD feature generation unit 205.

[0054] The first setting unit 204 outputs parameters to the LD feature generation unit 205. Specifically, the first setting unit 204 outputs the panel contrast and luminance distribution model to the LD feature generation unit 205. The panel contrast is the contrast value of the liquid crystal panel 105 of the liquid crystal display device 100. The luminance distribution model is a model of the luminance distribution of light emitted from the light source of the liquid crystal display device 100 (the part of the backlight module 107 corresponding to the divided area). The parameters set by the first setting unit 204 are obtained from the computer's memory, but are not limited to this, and may also be input from an external source via the OSD (On Screen Display) menu or the like. Note that the parameters set by the first setting unit 204 to the LD feature generation unit 205 are not limited to the panel contrast and luminance distribution model, but may also include other parameters necessary for the LD feature generation unit 205.

[0055] The LD feature generation unit 205 generates LD (Local Dimming) features based on the image features output from the preprocessing unit 202 and the backlight control values ​​output from the backlight control value generation unit 203. In Example 1, the LD feature generation unit 205 generates LD features using parameters output from the first setting unit 204. The parameters output from the first setting unit 204 to the LD feature generation unit 205 are panel contrast and luminance distribution models, but are not limited to these; other parameters may be used, and external parameters may not be used at all. The LD features correspond to a simulated image of local dimming in the liquid crystal display device 100. The LD feature generation unit 205 then outputs the generated LD features to the error calculation unit 208. Details of the processing of the LD feature generation unit 205 will be described later.

[0056] The second setting unit 206 outputs parameters to the target feature generation unit 207. Specifically, the second setting unit 206 outputs the target contrast to the target feature generation unit 207. The target contrast is a contrast value used to convert the image features output from the preprocessing unit 202 to the target contrast. The parameters set by the second setting unit 206 are obtained from the computer's memory, but are not limited to this; they may also be input externally via an OSD (On Screen Display) menu or the like. Note that the parameters set by the second setting unit 206 to the target feature generation unit 207 are not limited to the target contrast and may include other parameters necessary for the target feature generation unit 207.

[0057] The target feature generation unit 207 generates target features based on the image features output from the preprocessing unit 202. The target features correspond to the target display image when local dimming control is performed on the liquid crystal display device 100. In Embodiment 1, the target feature generation unit 207 generates target features by converting the image features into a target contrast output from the second setting unit 206. Specifically, the data value Vt of the target feature data is calculated by performing a gain / offset calculation on the image feature data value V based on the target contrast Ct, according to the following formula (8). The target contrast is obtained from the first setting unit 204. If the contrast ratio of the target contrast is 1,000,000:1, then 1,000,000 is entered for the target contrast Ct.

number

[0058] In Example 1, the target feature generation unit 207 generates target features by converting image features to a target contrast, but is not limited to this; any target feature corresponding to the target display image when local dimming control is performed on the liquid crystal display device 100 is acceptable. For example, the target feature generation unit 207 may acquire pre-prepared target features from an external source. Alternatively, the target feature generation unit 207 may output the image features output from the preprocessing unit 202 as the target features. The target feature generation unit 207 then outputs the generated target features to the error calculation unit 208.

[0059] The error calculation unit 208 generates the parameters for the neural network of the backlight control value generation unit 203 based on the difference between the LD features output from the LD feature generation unit 205 and the target features output from the target feature generation unit 207. The difference between the LD features and the target features is, for example, L1 (the sum of the absolute values ​​of the differences). In this case, the error calculation unit 208 calculates the parameters for the neural network of the backlight control value generation unit 203 so as to minimize the difference between the LD features and the target features. The backlight control value generation unit 203 then outputs the generated parameters to itself.

[0060] The method for generating parameters by the error calculation unit 208 is not limited to one based on the difference between the LD feature and the target feature, but may be based on other methods. For example, the error calculation unit 208 may generate the parameters of the neural network of the backlight control value generation unit 203 based on the truth value determination of the LD feature and the target feature by the classifier. The classifier is composed of a neural network. In this case, the error calculation unit 208 is trained (updated) so that when a combination of image feature and target feature is input, the output value approaches 1 (true), and when a combination of image feature and LD feature is input, the output value approaches 0 (fake). Furthermore, the error calculation unit 208 calculates (updates) the parameters of the neural network of the backlight control value generation unit 203 so that the output value of the classifier when a combination of image feature and LD feature is input approaches 1 (true).

[0061] Furthermore, the parameter generation method by the error calculation unit 208 may be a combination of multiple methods. For example, a method based on the difference between the LD feature and the target feature may be combined with a method based on the truth value determination of the LD feature and the target feature by the classifier.

[0062] The trained model output unit 209 outputs the neural network parameters of the backlight control value generation unit 203 to the outside as a trained model. Furthermore, the trained model output unit 209 outputs the target contrast set by the second setting unit 206 to the outside in association with the trained model. For example, the trained model output unit 209 outputs the trained model and target contrast to the trained model holding unit 10301 of the liquid crystal display device 100 via external memory or the like.

[0063] As explained with reference to Figure 7, the model learning unit 200 of Example 1 can be trained to output a backlight control value so that the displayed image due to local dimming approaches the target value.

[0064] Figure 8 is a block diagram showing the functional blocks of the LD feature generation unit 205 according to Example 1. The LD feature generation unit 205 includes a backlight brightness estimation unit 20501, a correction coefficient generation unit 20502, a correction unit 20503, a contrast conversion unit 20504, and a synthesis unit 20505.

[0065] The backlight brightness estimation unit 20501 estimates the brightness of the light illuminating the liquid crystal panel 105 when the backlight module 107 of the liquid crystal display device 100 is illuminated using the backlight control value output from the backlight control value generation unit 203 (backlight brightness estimation unit 20501). Similar to the backlight brightness estimation unit 10306, the estimation calculation of the backlight brightness estimation unit 20501 estimates (calculates) the backlight brightness for each brightness estimation point discretely arranged within the display surface, as shown in Figure 6(A), based on the backlight control value and the brightness distribution model. Here, the backlight brightness estimation unit 20501 estimates (calculates) one backlight brightness for each divided region of one backlight module. The backlight brightness estimation unit 20501 obtains the brightness distribution model used for the estimation calculation from the first setting unit 204. The backlight brightness estimation unit 20501 then outputs the estimated backlight brightness to the correction coefficient generation unit 20502 and the synthesis unit 20505.

[0066] The correction coefficient generation unit 20502 generates a correction coefficient to be applied to the image features based on the backlight brightness output from the backlight brightness estimation unit 20501. The method for generating the correction coefficient by the correction coefficient generation unit 20502 is the same as that of the correction coefficient generation unit 10307 of the liquid crystal display device 100, for example, by calculating it using the reciprocal of the backlight brightness. The correction coefficient generation unit 20502 then outputs the generated correction coefficient to the correction unit 20503.

[0067] The correction unit 20503 generates (calculates) corrected features by multiplying the image features output from the preprocessing unit 206 by the correction coefficient output from the correction coefficient generation unit 20502. Specifically, the data value Vc of the corrected features is calculated by multiplying the data value V of the image features by the correction coefficient Gt according to the following formula (9). Here, the data value V of the features is a value normalized to between 0.0 and 1.0. Vc = V × Gt ... Formula (9)

[0068] The correction unit 20503 then outputs the generated correction feature quantity to the contrast conversion unit 20504.

[0069] The contrast conversion unit 20504 generates panel features by converting the correction features output from the correction unit 20503 into panel contrast. The panel features correspond to a simulated image when the backlight module 107 of the liquid crystal display device 100 is made to emit light uniformly and the correction features output from the correction unit 20503 are displayed on the liquid crystal panel 105. Specifically, the data value Vp of the panel features is calculated by performing a gain / offset calculation on the data value Vc of the correction features based on the panel contrast Cp, according to the following formula (10). The panel contrast is the contrast value of the liquid crystal panel 105 of the liquid crystal display device 100 and is obtained from the first setting unit 204. If the contrast of the liquid crystal panel 105 of the liquid crystal display device 100 is 1000:1, then 1000 is entered for the panel contrast Cp.

number

[0070] The contrast conversion unit 20504 then outputs the corrected feature quantities (panel feature quantities) converted to panel contrast to the synthesis unit 20505.

[0071] The synthesis unit 20505 generates LD (Local Dimming) features by multiplying the panel features output from the contrast conversion unit 20504 by the backlight brightness output from the backlight brightness estimation unit 20501. Specifically, the data value Vl of the LD features is calculated by multiplying the data value Vp of the panel features by the backlight brightness L according to the following formula (11). Vl = Vp × L ... Formula (11)

[0072] As explained with reference to Figure 8, the LD feature generation unit 205 of Example 1 can generate data equivalent to a simulation image of local dimming for calculating the error with the target feature.

[0073] Refer to Figures 9-11 to explain the difference between local dimming using specific rules and local dimming using the machine learning model in Example 1.

[0074] Figure 9(A) is an example of an 8x8 checkerboard pattern image, and Figure 9(B) is an example of a 16x16 checkerboard pattern image. The checkerboard pattern image in Figure 9 is an image in which white and black rectangles are arranged alternately.

[0075] Figures 10-11 are examples of graphs obtained by sampling data related to the checker pattern image in Figure 9 horizontally from the area indicated by the arrow. The solid lines in Figures 10-11 are graphs normalized from 0 to 100% for the feature quantities of the checker pattern image in Figure 9. The dotted lines in Figures 10-11 are graphs of backlight control values ​​generated based on the feature quantities shown by the solid lines. The dashed lines in Figures 10-11 are graphs of the brightness of the light irradiated onto the liquid crystal panel (backlight brightness) when the backlight module is illuminated with the backlight control values ​​shown by the dotted lines. Note that in the backlight brightness shown by the dashed lines in Figures 10-11, the backlight brightness that displays the image with the maximum gradation value input to the liquid crystal panel at a predetermined white brightness (e.g., 1000 nits) is defined as 100%.

[0076] Figure 10(A) is a graph showing the results of local dimming applied to the checkerboard pattern image in Figure 9(A) according to a specific rule. Figure 10(B) is a graph showing the results of local dimming applied to the checkerboard pattern image in Figure 9(B) according to a specific rule. Figure 11(A) is a graph showing the results of local dimming applied to the checkerboard pattern image in Figure 9(A) using the machine learning model of Example 1. Figure 11(B) is a graph showing the results of local dimming applied to the checkerboard pattern image in Figure 9(B) using the machine learning model of Example 1.

[0077] First, let's explain local dimming using specific rules, referring to Figure 10. In local dimming using specific rules, a predetermined spatial filter is applied to the image features to generate backlight control values, for example, to prevent a decrease in brightness in the bright areas of the image. Spatial filters include, for example, HPF (High Pass Filter) and LPF (Low Pass Filter). By applying a predetermined spatial filter, i.e., a predetermined HPF or LPF, to the image features, the values ​​in areas with a difference in brightness are amplified, as shown by the dotted line in Figure 10. Furthermore, as shown by the dotted line in Figure 10, when a predetermined spatial filter is applied to statistics (features) to generate backlight control values, the degree of increase in backlight brightness varies depending on the image pattern (image characteristics). Therefore, when a predetermined spatial filter is applied to statistics (features) to generate backlight control values, the backlight brightness illuminating the liquid crystal panel may increase excessively depending on the image pattern (image characteristics), as shown by the dashed line in Figure 10. In Figure 10, within the dark areas (areas where the feature quantity is close to 0%, indicated by solid lines), areas where the backlight brightness is excessively high (for example, areas exceeding 100%), indicated by dashed lines, are visible as black level issues.

[0078] Next, with reference to Figure 11, local dimming using the machine learning model of Example 1 will be explained. In local dimming using the machine learning model of Example 1, a machine learning model is trained to generate backlight control values ​​so that the brightness of the displayed image due to local dimming approaches the target value. As a result, the machine learning model of Example 1 generates backlight control values ​​(dotted lines in Figure 11) according to the image pattern (image characteristics) so that the minimum necessary backlight brightness (dashed lines in Figure 11) is secured in the bright areas (white rectangles in Figure 9). Furthermore, the machine learning model of Example 1 generates backlight control values ​​(dotted lines in Figure 11) according to the image pattern (image characteristics) so that the backlight brightness (dashed lines in Figure 11) is reduced as much as possible in the dark areas (black rectangles in Figure 9).

[0079] As described above, in Example 1, in a liquid crystal display device in which the luminous intensity of the backlight module can be changed, a machine learning model can be used to generate control values ​​for the backlight module. By training the machine learning model so that the brightness of the displayed image due to local dimming approaches the target value, the luminous intensity of the backlight module can be controlled according to the characteristics of the image, thereby improving the display quality. Furthermore, in Example 1, since the control values ​​of the backlight module generated using the machine learning model are adjusted, the contrast of the displayed image can be changed without changing the machine learning model.

[0080] [Example 2] The following describes Embodiment 2 of the present invention. Embodiment 1 described an example of adjusting the contrast of a displayed image without changing the machine learning model in order to adjust the control values ​​of the backlight module generated using a machine learning model. Embodiment 2 describes an example of adjusting the contrast of a displayed image according to the input image data. Specifically, Embodiment 2 describes an example of adjusting the contrast of a displayed image so that halos are less visible. A halo is a phenomenon in liquid crystal display devices that perform local dimming control in which light leakage (black floating) of the backlight is visible in the dark area surrounding the bright area. The configuration of the liquid crystal display device 100 in Figure 12 according to Embodiment 2 is the same as the liquid crystal display device 100 in Figure 1, with the addition of a halo evaluation unit 10310 and some differences in the processing of the parameter application unit 10302. In Embodiment 2, the same reference numerals and names used in Embodiment 1 are used for components equivalent to those in Embodiment 1, and detailed explanations are omitted as appropriate.

[0081] The halo evaluation unit 10310 calculates a halo evaluation value based on the input image output from the image input / conversion unit 101, the backlight brightness output from the backlight brightness estimation unit 10306, and the corrected image output from the image correction unit 10308. The halo evaluation value is a numerical representation of the degree of halo occurrence. The method for calculating the halo evaluation value will be described later with reference to the flowchart in Figure 14. The halo evaluation unit 10310 outputs the calculated halo evaluation value to the parameter application unit 10302.

[0082] The parameter application unit 10302 applies parameters to the post-processing unit 10305 based on the halo evaluation value output from the halo evaluation unit 10310. The parameter application unit 10302 calculates the contrast (display contrast) Cd of the display image of the liquid crystal display device 100 based on the halo evaluation value HR using the linear function shown in the following formula (12). In the following formula (12), a and b represent variables. By changing the variables a and b in formula (12), the adjustment range of the display contrast can be changed. By applying a negative value to the constant a in formula (12), the parameter application unit 10302 can decrease the display contrast Cd in accordance with an increase in the halo evaluation value HR. The variables a and b in formula (12) may be changed according to the setting value input by the user via the setting input unit 102 or according to the target contrast associated with the learned model applied to the backlight control value generation unit 10304. The method for calculating the display contrast Cd based on the halo evaluation value HR is not limited to the linear function shown in formula (12), but may also be a quadratic function, an exponential function, or a 1D-LUT. The parameter application unit 10302 calculates the parameters to be applied to the post-processing unit 10305 by applying the display contrast Cd calculated using the following formula (12) to the display contrast Cd in formula (1) mentioned above. Cd = a × HR + b ... Formula (12)

[0083] Figure 13(A) shows an example of a patch image with a white patch placed on a black background. When the liquid crystal display device 100 displays the patch image shown in Figure 13(A) by local dimming, a halo is visible around the white patch, as shown in Figure 13(B). The parameter application unit 10302 can reduce the display contrast when the degree of halo occurrence is high. As a result, as shown in Figure 13(C), the brightness of the black background area increases, and the brightness difference with the area where the halo occurs decreases, making the halo less visible.

[0084] FIG. 14 is a flowchart showing the calculation process of the halo evaluation value in the halo evaluation unit 10310.

[0085] In S21, the halo evaluation unit 10310 creates a display simulation image based on the backlight luminance output from the backlight luminance estimation unit 10306, the corrected image output from the image correction unit 10308, the panel gamma, and the panel contrast. The panel gamma is the gamma characteristic of the liquid crystal panel 105. The panel contrast is the contrast value of the liquid crystal panel 105. The display simulation image is an image that simulates the image displayed on the liquid crystal display device 100. Let the pixel value of the corrected image be Vc xy , the panel gamma be pg, the panel contrast be Cp, and the backlight luminance be BL xy . When this is the case, the halo evaluation unit 10310 calculates the luminance value Ls of the display simulation image according to, for example, the following formula (13). For Vc xy and BL xy in formula (13), and Ls xy , x represents the horizontal pixel position and y represents the vertical pixel position.

Equation

[0086] <0OO0377>In S22, the halo evaluation unit 10310 creates an ideal image based on the input image output from the image input / transformation unit 101, the panel gamma, and the target contrast. The target contrast is the contrast value targeted in the local dimming control of the liquid crystal display device 100. The ideal image is an image when the input image is displayed with the target contrast. Let the pixel value of the input image be Vi xy , the panel gamma be pg, and the target contrast be Ct. When this is the case, the halo evaluation unit 10310 calculates the luminance value Lr of the ideal image according to, for example, the following formula (14). For Vi xy and Lr xy in formula (14), x represents the horizontal pixel position and y represents the vertical pixel position.

Equation

[0087] In S23, the halo evaluation unit 10310 calculates a halo evaluation value based on the displayed simulation image and the ideal image. The halo evaluation value is calculated, for example, using a series of formulas from formula (15) to formula (19) below.

[0088] Formula (15) is used to calculate the number of pixels Nd in the dark region (luminance less than Lth) of an ideal image. In formula (15), n represents the number of horizontal pixels and m represents the number of vertical pixels.

[0089]

number

[0090] Equation (16) shows that in the dark region of the ideal image (luminance less than Lth), the luminance value Ls of the display simulation image is calculated. xy and the brightness value Lr of the ideal image xy This is a formula for calculating the sum of the differences between the two values, Sd. In formula (16), n represents the number of horizontal pixels and m represents the number of vertical pixels.

number

[0091] Equation (17) shows that in the dark region of the ideal image (luminance less than Lth), the luminance value Ls of the display simulation image is calculated. xy and the brightness value Lr of the ideal image xy This is a formula for calculating Sds, which is the sum of the squares of the differences between the two values. In formula (17), n represents the number of horizontal pixels and m represents the number of vertical pixels.

number

[0092] Formula (18) is used to calculate the standard deviation SDd of the difference (luminance error) between the luminance value of the display simulation image and the luminance value of the ideal image in the dark areas of the ideal image. Formula (18) allows for the quantification of the degree of variation in luminance error in the dark areas.

number

[0093] Formula (19) calculates the halo evaluation value HR by multiplying the standard deviation SDd of the luminance error in the dark region calculated in formula (18) by the adjustment coefficient Hcoe. The halo becomes more noticeable as the area of ​​the dark region increases. Therefore, in formula (19), for example, the area ratio of the dark region is applied as the adjustment coefficient Hcoe. Note that the method of using formulas (13) to (19) to calculate the halo evaluation value is just one example; other methods that can quantify the halo may also be used. HR = SDd × Hcoe ... Formula (19)

[0094] In S24, the halo evaluation unit 10310 outputs the halo evaluation value calculated in S23 to the parameter application unit 10302.

[0095] As explained above, in Example 2, the contrast of the displayed image can be adjusted so that the halo is less visible. In Example 2, the contrast of the displayed image is controlled according to the input image data, but the control can also be toggled on or off.

[0096] [Example 3] The following describes Embodiment 3 of the present invention. Embodiment 2 described an example of adjusting the contrast of a display image according to a halo evaluation value based on a display simulation image and an ideal image. Embodiment 3 describes an example of adjusting the contrast of a display image according to the feature quantities of the input image data. The configuration of the liquid crystal display device 100 in Embodiment 3 is the same as that of the liquid crystal display device 100 in Figure 12, but the processing of the halo evaluation unit 10310 is slightly different. In Embodiment 3, the same reference numerals and names used in Embodiment 2 are used for components equivalent to those in Embodiment 2, and detailed explanations are omitted as appropriate.

[0097] The halo evaluation unit 10310 calculates a halo evaluation value based on the input image output from the image input / conversion unit 101. The method for calculating the halo evaluation value will be described later with reference to the flowchart in Figure 15. The halo evaluation unit 10310 outputs the calculated halo evaluation value to the parameter application unit 10302.

[0098] Figure 15 is a flowchart showing the calculation process of the halo evaluation value in the halo evaluation unit 10310.

[0099] In S31, the halo evaluation unit 10310 calculates image features based on the input image output from the image input / conversion unit 101. The image features calculated by the halo evaluation unit 10310 in S31 are, for example, the maximum pixel value (full-screen maximum value) or the average pixel value (full-screen average value) calculated from the entire input image, but are not limited to these; any data obtainable from the input image is acceptable.

[0100] In S32, the halo evaluation unit 10310 calculates a halo evaluation value based on the image features calculated in S31. Halos are more likely to occur in images that are generally dark and have small areas of brightness. For example, this characteristic can be captured by dividing the maximum value of the entire screen by the average value of the entire screen. Therefore, the halo evaluation unit 10310 calculates the halo evaluation value HR based on the maximum value of the entire screen Smax and the average value of the entire screen Savg according to the following formula (20). In formula (20), the image features are transformed by the function fh. In formula (20), the function fh is logarithmic. In the function fh, a and b are constants. Note that the method for calculating the halo evaluation value from image features is not limited to the method using formula (20), and other methods that can quantify the halo may be used.

number

[0101] In S33, the halo evaluation unit 10310 outputs the halo evaluation value calculated in S32 to the parameter application unit 10302.

[0102] As explained above, in Example 3, similar to Example 2, the contrast of the displayed image can be adjusted so that the halo is less visible. Furthermore, in Example 3, since feedback processing is not required as in Example 2, processing delay is reduced.

[0103] [Other embodiments] The present invention can also be realized by supplying a program that implements one or more of the functions of the above-described embodiments to a system or device via a network or storage medium, and by having one or more processors in the computer of that system or device read and execute the program. It can also be realized by a circuit (e.g., an ASIC) that implements one or more functions.

[0104] It should be noted that the present invention is not limited to the embodiments described above, and the components can be modified and implemented in practice without departing from the spirit of the invention. Furthermore, various inventions can be formed by appropriately combining the multiple components disclosed in the above embodiments. For example, some components may be deleted from all the components shown in the embodiments. Moreover, components from different embodiments may be appropriately combined.

[0105] In the above embodiment, at least one of A and B may be A alone, B alone, or A and B.

[0106] Furthermore, the disclosure of this embodiment includes the following configurations and methods.

[0107] [Configuration 1] LCD panel and An input means for inputting data from the first image, A backlight module that illuminates the aforementioned liquid crystal panel with light, and whose luminescence brightness can be changed for each of the multiple divided regions, A generation means that inputs the data generated from the first image into a neural network to generate the light emission intensity of each divided region of the backlight module, An adjustment means for adjusting the luminescence intensity generated by the generation means, A calculation means for calculating a correction value for correcting the first image based on the light emission intensity adjusted by the adjustment means, Correction means for correcting the first image to a second image based on the correction value calculated by the calculation means, A light emission control means that controls the light emission intensity of each divided region of the backlight module based on the light emission intensity adjusted by the adjustment means, Display control means for controlling the transmittance of the liquid crystal panel based on the data of the second image, A liquid crystal display device characterized by having the following features.

[0108] [Configuration 2] The liquid crystal display device according to configuration 1, characterized in that the data generated from the first image is a feature quantity of the first image.

[0109] [Configuration 3] The liquid crystal display device according to configuration 1 or 2, characterized in that the trained model applied to the neural network in the generation means can be changed.

[0110] [Structure 4] It also includes a setting mechanism that accepts user input, The liquid crystal display device according to any one of configurations 1 to 3, characterized in that the adjustment means adjusts the trained model applied to the neural network, the target contrast associated with the trained model, and the luminescence intensity generated by the generation means in accordance with the user operation.

[0111] [Composition 5] The liquid crystal display device according to any one of configurations 1 to 3, characterized in that the adjustment means adjusts the luminescence intensity generated by the generation means according to the trained model applied to the neural network, the target contrast associated with the trained model, and the first image.

Claims

1. LCD panel and An input means for inputting data from the first image, A backlight module that illuminates the aforementioned liquid crystal panel with light, and whose luminescence brightness can be changed for each of the multiple divided regions, A generation means that inputs the data generated from the first image into a neural network to generate the light emission intensity of each divided region of the backlight module, An adjustment means for adjusting the luminescence intensity generated by the generation means, A calculation means for calculating a correction value for correcting the first image based on the light emission intensity adjusted by the adjustment means, Correction means for correcting the first image to a second image based on the correction value calculated by the calculation means, A light emission control means that controls the light emission intensity of each divided region of the backlight module based on the light emission intensity adjusted by the adjustment means, Display control means for controlling the transmittance of the liquid crystal panel based on the data of the second image, A liquid crystal display device characterized by having the following features.

2. The liquid crystal display device according to claim 1, characterized in that the data generated from the first image is a feature of the first image.

3. The liquid crystal display device according to claim 1, characterized in that the trained model applied to the neural network in the generation means is changeable.

4. It also includes a setting mechanism that accepts user input, The liquid crystal display device according to claim 1, characterized in that the adjustment means adjusts the trained model applied to the neural network, the target contrast associated with the trained model, and the luminescence intensity generated by the generation means in accordance with the user operation.

5. The liquid crystal display device according to claim 1, characterized in that the adjustment means adjusts the luminescence intensity generated by the generation means according to the trained model applied to the neural network, the target contrast associated with the trained model, and the first image.

6. LCD panel and An input means for inputting data from the first image, A backlight module that illuminates the aforementioned liquid crystal panel with light, and whose luminescence brightness can be changed for each of the multiple divided regions, A method for controlling a liquid crystal display device having the following characteristics: A generation step of inputting data generated from the first image into a neural network to generate the light emission intensity of each divided region of the backlight module, An adjustment step to adjust the luminescence intensity generated by the above generation step, A calculation step for calculating a correction value to correct the first image based on the light emission intensity adjusted by the adjustment step, A correction step in which the first image is corrected to a second image based on the correction value calculated in the calculation step, A light emission control step that controls the light emission intensity of each divided region of the backlight module based on the light emission intensity adjusted by the adjustment step, A display control process that controls the transmittance of the liquid crystal panel based on the data of the second image, A control method characterized by having the following features.

7. LCD panel and An input means for inputting data from the first image, A backlight module that illuminates the aforementioned liquid crystal panel with light, and whose luminescence brightness can be changed for each of the multiple divided regions, A computer with a liquid crystal display device having, A generation step of inputting data generated from the first image into a neural network to generate the light emission intensity of each divided region of the backlight module, An adjustment step to adjust the luminescence intensity generated by the above generation step, A calculation step for calculating a correction value to correct the first image based on the light emission intensity adjusted by the adjustment step, A correction step in which the first image is corrected to a second image based on the correction value calculated in the calculation step, A light emission control step that controls the light emission intensity of each divided region of the backlight module based on the light emission intensity adjusted by the adjustment step, A display control process that controls the transmittance of the liquid crystal panel based on the data of the second image, A program for causing a control method to be executed.

8. The input process involves inputting data from the input image, A first generation step involves inputting data generated from the aforementioned input image into a neural network to generate the luminescence intensity of the backlight module. A second generation step generates LD feature quantities based on the data generated from the input image and the emission intensity generated in the generation step, A third generation step generates target features based on the data generated from the input image and the target contrast, A calculation step of calculating the difference between the LD feature and the target feature, An update step to update the parameters of the neural network based on the aforementioned difference, The output process involves mapping the trained model to the target contrast and outputting the result. A learning method characterized by having the following features.

9. The learning method according to claim 8, characterized in that in the second generation step, the LD feature quantity is generated based on the data generated from the input image, the luminescence intensity generated in the generation step, and the contrast of the liquid crystal panel.

10. The method for manufacturing a trained model according to claim 9, characterized in that the contrast of the liquid crystal panel can be changed in the second generation step.

11. The learned method according to claim 8, characterized in that the target contrast can be changed in the third generation step.

Images