JP2016163099A - Image processor and image display device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To display an image of high quality while suppressing image dignity from decreasing by optimizing operation for gradation conversion processing according to a state of a noise component when performing processing to make an image light while emphasizing a local contrast by performing gradation conversion processing on an input image signal.SOLUTION: A gradation conversion part (15) of an image processor (100) includes: a first illumination light distribution calculation part which calculates a distribution of first illumination light intensities by averaging lightness of respective pixels in a first predetermined region of an input image signal; and a first correction coefficient calculation part which maintains a certain level in a region in which a gradation value of first illumination light intensity is less than a predetermined threshold for each pixel of the input image signal. A first correction coefficient calculation part varies a first correction coefficient less than the threshold according to average brightness that an average brightness detection part (13) detects.SELECTED DRAWING: Figure 3

Description

  The present invention relates to an image processing device and an image display device, and more particularly to an image processing device and an image display device.

In recent years, as part of improving the image quality of a liquid crystal display device using a liquid crystal panel, a 4K2K image (for example, 4096 ×) having a resolution higher than that of a so-called 2K1K image called a FHD (Full High Definition) image (for example, around 1920 × 1080 pixels). A liquid crystal display device using a high-resolution liquid crystal panel capable of displaying 2160 pixels) is provided. In addition, there is a strong demand for higher resolution, and a liquid crystal display device capable of displaying a so-called 8K4K image having pixels twice as long as a 4K2K image has been announced.
However, at present, the circuit configuration for enabling the display of 8K4K images is large, which causes an increase in cost. In order to solve such a problem, a liquid crystal display device capable of so-called pseudo 8K4K display, in which an image equivalent to an 8K4K image is displayed on a liquid crystal panel having a 4K2K resolution, has been studied.

  As the resolution of the liquid crystal display device on the image signal receiving side is increasing as described above, when looking at the situation on the transmitting side, there are few contents with realistic 4K and 8K resolutions, and images with resolutions below FHD The problem is how to enlarge the image so that it can be displayed with high image quality of 4K or 8K resolution.

  In response to such a problem, a high image quality process is performed by a so-called super resolution process. Super-resolution processing is equipped with an upscaler in the liquid crystal display device, which converts the resolution of the input image to interpolate the missing pixels, and at this time emphasizes the high frequency components to achieve high definition. The In addition to the super-resolution process, a so-called pixel deming gradation conversion process for increasing the image quality by brightening the display image and improving the contrast feeling has been put into practical use.

FIG. 18 is a diagram for explaining the concept of gradation conversion processing by pixel dimming. In the pixel dimming gradation conversion process, gradation conversion is performed in two stages to improve image quality. FIG. 18A schematically shows the gradation of the input image signal, the horizontal axis indicates the position of the pixel in the image, and the vertical axis indicates the pixel value (gradation value) of the image signal. ing.
Here, the gradation distribution L in the low gradation area and the gradation distribution H in the high gradation area in the image signal are shown simultaneously. The low gradation region is a low gradation region up to, for example, 0-255 gradation when an image signal is expressed in 4096 gradations of 12 bits, and the high gradation region is a high gradation of, for example, 2600 gradations or more. It is assumed that it is a key area.

  FIG. 18B shows an image after gradation conversion processing at the first stage of the two stages of gradation conversion. In gradation conversion at the first stage, gradation conversion is performed by multiplying the gradation of a pixel by a correction coefficient (first correction coefficient) to increase the gradation. The first correction coefficient is determined based on a value obtained by averaging pixel values of a pixel of interest and pixels in a predetermined region around the target pixel. At this time, for example, the largest first correction coefficient is applied in a predetermined low gradation region, and the first correction coefficient that decreases in value toward the highest gradation is applied. At the highest gradation, the correction coefficient is 1, and the gradation hardly changes near the highest gradation.

  In this case, as shown in FIG. 18B, the gradation distribution L in the low gradation area is raised by multiplying the input image signal by a predetermined amount of the first correction coefficient. On the other hand, the gradation distribution H of the high gradation area H hardly changes in this example. In the first-stage gradation conversion, it is possible to brighten the image while maintaining local contrast by maintaining the state of local fine pixel value fluctuation in the low gradation area.

FIG. 18C shows an image after the second-stage gradation change process. In the second-stage gradation conversion processing, the gradation of the image subjected to gradation conversion with the first image correction coefficient is multiplied by the second correction coefficient. The second correction coefficient is a value of 1 or less that decreases the input pixel value, and has a smaller value as the input pixel becomes lower in gradation. In the second gradation conversion, the local contrast maintained by the first gradation conversion is enhanced to improve visibility.
Through the processing as described above, image quality improvement processing is performed to improve the visibility of the dark area and improve the contrast.

  As a technique related to image quality enhancement processing, for example, in Patent Document 1, enhancement processing for noise components can be suppressed as compared to the conventional technology, and sufficient enhancement processing is performed on an input image with less noise components. A video signal processing apparatus for the purpose is disclosed. This video signal processing apparatus includes a noise reduction processing unit that performs noise reduction processing on an input video signal, a noise reduction amount detection unit that detects a noise reduction amount indicating the degree of noise reduction processing, and a video after the noise reduction processing A residual noise amount detection unit that detects a residual noise amount remaining in the signal, and an enhancement amount determination unit that determines an enhancement amount indicating a degree of enhancement processing based on at least the noise reduction amount of the noise reduction amount and the residual noise amount; And an enhancement processing unit that performs enhancement processing on the video signal after the noise reduction processing based on the enhancement amount determined by the enhancement amount determination unit.

  Patent Document 2 discloses a video correction circuit and a video display device that are intended to correct contrast while maintaining the sharpness and S / N ratio of an image. The video correction circuit is provided with an average luminance level calculation circuit and a histogram calculation circuit. A noise reduction intensity value that defines the degree of noise component reduction of the video signal by the NR circuit according to the average brightness level value calculated by the average brightness level calculation circuit and the brightness level histogram calculated by the histogram calculation circuit; The tone correction coefficient used for contrast correction by the contrast correction circuit is set by the video processing controller.

  In Patent Document 3, it is determined whether a displayed image is a moving image or a still image, and image processing suitable for the moving image or the still image, in particular, the contrast of the still image is improved to obtain a high-quality still image. An image display device or the like for performing obtained image processing is disclosed. The image processing apparatus includes: a luminance distribution change amount calculation unit that calculates a change amount of the luminance distribution of the image between frames to be compared; and the image is either a moving image or a still image based on the change amount of the luminance distribution. A moving image still image determining unit that determines whether or not, a gradation correction processing unit that performs predetermined gradation correction processing based on a determination result of the moving image still image determination unit, and a display unit that displays an image subjected to the gradation correction processing Have

JP 2009-44487 A JP 2010-22030 A Japanese Patent No. 3660142

  Due to the characteristics of pixel demming, the contrast is expanded after the low gradation area of the image signal is lifted once, so if noise is included in the input image signal, unlike normal sharpness processing, The noise component is further expanded, and the video quality is lowered. Therefore, it is necessary to sufficiently remove noise from the image signal before a circuit that performs gradation conversion processing by pixel deming.

As a specific example, in the gradation conversion of the low gradation area, an input image signal represented by 4096 gradations is multiplied by a first correction coefficient to the low gradation area in the range of 0 to 255 gradations. Tone conversion and raising the gradation value. At this time, the noise component in the vicinity of the 0th gradation is low in level and hardly noticeable. However, by raising the low gradation area by gradation conversion, the gradation of the noise component is also raised and is easily visible. Since the noise component is expanded while maintaining the local contrast, the noise component becomes more conspicuous.
That is, if the input image signal includes a noise component, the noise component in the low gradation region is further expanded by performing pixel deming processing, and there is a problem that the image quality is lowered.

  Patent Documents 1 to 3 disclose image processing techniques aimed at improving the image quality of displayed images. As described above, the brightness of an image in a low gradation region is optimized to enhance local contrast. Therefore, it is not applied to a technique for improving image quality.

  The present invention has been made in view of the above-described circumstances, and in performing a process of brightening an image while performing a gradation conversion process on an input image signal to enhance a local contrast, the noise component state is obtained. Accordingly, it is an object of the present invention to provide an image processing apparatus and an image display apparatus that can perform high-quality image display while suppressing deterioration in image quality by optimizing the operation of gradation conversion processing accordingly. .

  In order to solve the above-described problem, the first technical means of the present invention averages the brightness of each pixel in the first predetermined area around the pixel for each pixel of the input image signal, and reflects the subject. A first illumination light distribution calculating unit that calculates a first illumination light intensity serving as an index indicating light intensity and calculating a distribution of the first illumination light intensity of the input image signal; and for each pixel of the input image signal Furthermore, a constant level is maintained in a region where the gradation value of the first illumination light intensity is less than a predetermined threshold value, and the first illumination light intensity is maintained in a region where the gradation value of the first illumination light intensity is greater than or equal to the threshold value. A first correction coefficient calculation unit that calculates a first correction coefficient that decreases as the gradation value increases, and a second predetermined smaller than the first predetermined area from the image signal corrected by the first correction coefficient. The brightness of each pixel in the region is averaged to calculate the second illumination light intensity distribution A second illumination light distribution calculating section; a second correction coefficient calculating section for calculating a second correction coefficient that decreases as the gradation value of the second illumination light intensity decreases; the first correction coefficient; A gradation conversion processing unit that performs gradation conversion by applying a second correction coefficient to the input image signal, and an average luminance detection unit that detects an average luminance for each frame of the input video signal. The first correction coefficient calculation unit changes the value of the first correction coefficient less than the threshold according to the average luminance detected by the average luminance detection unit.

  According to a second technical means, in the first technical means, the first correction coefficient calculation unit relatively sets a value of the first correction coefficient that is less than the threshold for an image signal having a relatively low average luminance. It is characterized by lowering.

  According to a third technical means, in the second technical means, the first correction coefficient calculation unit divides the average luminance level detected by the average luminance detection unit into a plurality of levels, and the lower the average luminance, the higher the level. The value of the first correction coefficient less than the threshold value is lowered.

  The fourth technical means averages the brightness of each pixel in the first predetermined area around the pixel for each pixel of the input image signal, and serves as an index indicating the intensity of the reflected light of the subject. A first illumination light distribution calculator that calculates illumination light intensity and calculates a distribution of the first illumination light intensity of the input image signal; and a gradation of the first illumination light intensity for each pixel of the input image signal A constant level is maintained in a region where the value is less than a predetermined threshold value, and the value decreases as the gradation value of the first illumination light increases in a region where the gradation value of the first illumination light intensity is greater than or equal to the threshold value. The brightness of each pixel in a second predetermined area smaller than the first predetermined area is averaged from a first correction coefficient calculating unit that calculates a first correction coefficient and an image signal corrected by the first correction coefficient. A second illumination light distribution calculation unit for calculating a second illumination light intensity distribution; A second correction coefficient calculation unit that calculates a second correction coefficient that decreases as the gradation value of the illumination light intensity decreases, and applies the first correction coefficient and the second correction coefficient to the input image signal. An image processing apparatus having a gradation conversion processing unit that performs gradation conversion, and detects an average luminance for each frame of the input image signal, and detects a peak luminance for each frame of the input image signal The first correction coefficient calculation unit is configured to reduce the first correction less than the threshold according to the average luminance detected by the average luminance detection unit and the peak luminance detected by the peak detection unit. It is characterized by changing the value of the coefficient.

  According to a fifth technical means, in the fourth technical means, the first correction coefficient calculating unit has an average luminance detected by the average luminance detecting unit lower than a predetermined level and the peak detected by the peak luminance detecting unit. Except for the case where the luminance level is equal to or higher than a predetermined level, in the image signal having a relatively low average luminance, the value of the first correction coefficient less than the threshold value is relatively low. .

  According to a sixth technical means, in the fifth technical means, when the average luminance is lower than a predetermined level and the peak luminance level is equal to or higher than the predetermined level, the value of the second correction coefficient less than the threshold value is relatively set. It is characterized by making it higher.

  According to a seventh technical means, in the sixth technical means, the first correction coefficient calculating unit divides the average luminance level detected by the average luminance detecting unit into a plurality of stages, and the divided average luminance levels. In the stage where the average brightness level is the lowest, the peak brightness is divided into a stage where the peak brightness is higher than a predetermined level and a stage where the peak brightness is lower than the predetermined level, and the average brightness level is the lowest and the peak brightness level is higher than the predetermined level. Except for certain cases, the lower the average luminance, the lower the value of the first correction coefficient that is less than the threshold, when the average luminance level is the lowest and the peak luminance level is equal to or higher than a predetermined level, The value of the first correction coefficient that is less than the threshold value is set to be as high as that at the highest average luminance.

The eighth technical means averages the brightness of each pixel in the first predetermined area around the pixel for each pixel of the input image signal, and serves as an index indicating the intensity of reflected light from the subject. A first illumination light distribution calculation unit that calculates illumination light intensity and calculates a distribution of the first illumination light intensity of the input image signal; and a gradation of the first illumination light intensity for each pixel of the input image signal A constant level is maintained in a region where the value is less than a predetermined threshold value, and the value decreases as the gradation value of the first illumination light increases in a region where the gradation value of the first illumination light intensity is greater than or equal to the threshold value. The brightness of each pixel in a second predetermined area smaller than the first predetermined area is averaged from a first correction coefficient calculating unit that calculates a first correction coefficient and an image signal corrected by the first correction coefficient. A second illumination light distribution calculation unit for calculating a distribution of the second illumination light intensity; A second correction coefficient calculation unit that calculates a second correction coefficient that decreases as the gradation value of the bright light intensity decreases, and applies the first correction coefficient and the second correction coefficient to the input image signal. An image processing apparatus having a gradation conversion processing unit for performing tone conversion,
A luminance histogram detection unit that detects a luminance histogram obtained by integrating the number of pixels for each luminance value for each frame of the input image signal, wherein the first correction coefficient calculation unit is a luminance histogram detected by the luminance histogram detection unit; The value of the first correction coefficient that is less than the threshold value is changed according to the above.

  According to a ninth technical means, in the eighth technical means, the first correction coefficient calculating unit has an area of a predetermined low gradation region and a predetermined high gradation region of the luminance histogram detected by the luminance histogram detection unit. Based on this, in the predetermined range where the predetermined low gradation region is the smallest, the value of the first correction coefficient less than the threshold is relatively largest, and in the predetermined range where the predetermined low gradation region is the intermediate level, the predetermined value is determined. When there is a high gradation area, the value of the first correction coefficient less than the threshold is set to a relatively intermediate level, and when there is no predetermined high gradation area, the first correction coefficient less than the threshold When the predetermined low gradation area is the highest and the predetermined low gradation area is the highest, and the predetermined high gradation area is present, the value of the first correction coefficient less than the threshold value is relatively largest. And there is no predetermined high gradation area To come, in which it is characterized in that the value of the first correction coefficient less than the threshold value relatively lowest.

  The tenth technical means averages the brightness of each pixel in the first predetermined area around the pixel for each pixel of the input image signal, and serves as an index indicating the intensity of the reflected light of the subject. A first illumination light distribution calculation unit that calculates illumination light intensity and calculates a distribution of the first illumination light intensity of the input image signal; and a gradation of the first illumination light intensity for each pixel of the input image signal A constant level is maintained in a region where the value is less than a predetermined threshold value, and the value decreases as the gradation value of the first illumination light increases in a region where the gradation value of the first illumination light intensity is greater than or equal to the threshold value. The brightness of each pixel in a second predetermined area smaller than the first predetermined area is averaged from a first correction coefficient calculating unit that calculates a first correction coefficient and an image signal corrected by the first correction coefficient. A second illumination light distribution calculation unit for calculating a second illumination light intensity distribution; A second correction coefficient calculation unit that calculates a second correction coefficient that decreases as the gradation value of the illumination light intensity decreases, and applies the first correction coefficient and the second correction coefficient to the input image signal. An image processing apparatus comprising: a gradation conversion processing unit that performs gradation conversion; and a noise reduction processing unit that performs noise removal processing on an image signal input to the gradation conversion processing unit, wherein the input image signal For each frame, calculate the luminance difference between adjacent pixels, which is the absolute value of the luminance difference between adjacent pixels, and generate an edge histogram that integrates the number of pixels for each luminance difference between adjacent pixels. And an edge histogram detection unit that detects a change amount between frames of the generated edge histogram, wherein the first correction coefficient calculation unit detects the edge histogram detected by the edge histogram detection unit. The value of the first correction coefficient less than the threshold value is changed according to the amount of change in the ram, and the noise reduction processing unit changes the noise removal intensity according to the edge histogram detected by the edge histogram detection unit. It is what.

  According to an eleventh technical means, in the tenth technical means, the first correction coefficient calculation unit is less than the threshold value for an image signal in which the change amount of the edge histogram detected by the edge histogram detection unit is relatively small. The value of the first correction coefficient is set relatively high, and the noise reduction processing unit relatively increases the strength of the noise removal processing for an image signal in which the change amount of the edge histogram detected by the edge histogram detection unit is relatively small. It is characterized by weakening.

  According to a twelfth technical means, in the eleventh technical means, the amount of change in the edge histogram detected by the edge histogram detection unit is divided into a plurality of stages, and the first correction coefficient calculation unit includes the edge histogram detection unit. The smaller the change amount of the detected edge histogram, the higher the value of the first correction coefficient less than the threshold value, and the noise reduction processing unit is a step where the change amount of the edge histogram detected by the edge histogram detection unit is small. As described above, the strength of the noise removal process is reduced.

  The thirteenth technical means averages the brightness of each pixel in the first predetermined area around the pixel for each pixel of the input image signal, and serves as an index indicating the intensity of reflected light from the subject. A first illumination light distribution calculation unit that calculates illumination light intensity and calculates a distribution of the first illumination light intensity of the input image signal; and a gradation of the first illumination light intensity for each pixel of the input image signal A constant level is maintained in a region where the value is less than a predetermined threshold value, and the value decreases as the gradation value of the first illumination light increases in a region where the gradation value of the first illumination light intensity is greater than or equal to the threshold value. The brightness of each pixel in a second predetermined area smaller than the first predetermined area is averaged from a first correction coefficient calculating unit that calculates a first correction coefficient and an image signal corrected by the first correction coefficient. A second illumination light distribution calculation unit for calculating a second illumination light intensity distribution; A second correction coefficient calculation unit that calculates a second correction coefficient that decreases as the gradation value of the illumination light intensity decreases, and applies the first correction coefficient and the second correction coefficient to the input image signal. An image processing apparatus having a gradation conversion processing unit that performs gradation conversion, an average luminance detection unit that detects an average luminance for each frame of the input video signal, and a motion that detects movement of the input image signal A motion detection determination unit that determines the degree of noise and a noise reduction processing unit that performs noise removal processing on the input image signal, and the noise reduction processing unit includes a two-dimensional noise reduction function and a three-dimensional noise reduction function. And the first correction coefficient calculation unit has a value of the first correction coefficient less than the threshold value according to the average luminance detected by the average luminance detection unit and the degree of movement determined by the motion detection determination unit. The noise reduction processing unit is configured to change the intensity of noise removal processing according to the average luminance detected by the average luminance detection unit and the degree of movement determined by the motion detection determination unit. Is.

  In a fourteenth technical means according to the thirteenth technical means, the first correction coefficient calculation unit is configured to output the first correction coefficient less than the threshold value for an image signal having a relatively large degree of motion determined by the motion detection determination unit. The noise reduction processing unit increases the noise reduction strength by the two-dimensional noise reduction function and weakens the noise reduction strength by the three-dimensional noise reduction function. In the case where the average brightness level is the same level, an image signal having a relatively low average brightness detected by the average brightness detection unit has a relatively low value of the first correction coefficient less than the threshold value. Is.

  According to a fifteenth technical means, in the fourteenth technical means, the level of the average luminance detected by the average luminance is divided into a plurality of levels, and the level of the degree of motion determined by the motion detection determination unit is determined in a plurality of levels. The first correction coefficient calculation unit is configured such that, in the same average luminance stage, the first correction that is less than the threshold for the image signal having a relatively large degree of motion determined by the motion detection determination unit. While reducing the value of the coefficient, the noise reduction processing unit increases the noise reduction strength by the two-dimensional noise reduction function, and decreases the noise reduction strength by the three-dimensional noise reduction function. In the stage, the first correction coefficient calculation unit is less than the threshold value for an image signal having a relatively low average luminance detected by the average luminance detection unit. Is obtained by said lowering the value of the first correction coefficient.

  The sixteenth technical means includes an image processing apparatus according to any one of the first to fifteenth technical means, and an image display unit that displays an image based on an image signal subjected to gradation conversion by the image processing apparatus. It is a display device.

  According to the present invention, when the tone conversion process is performed on the input image signal to brighten the image while enhancing the local contrast, the operation of the tone conversion process is optimized according to the state of the noise component. By doing so, it is possible to provide an image processing apparatus and an image display apparatus that can perform high-quality image display while suppressing deterioration in image quality.

1 is a block diagram showing a configuration of a first embodiment of an image processing apparatus according to the present invention. It is a figure for demonstrating the function structural example of a gradation conversion process part. It is a figure which shows the example of the 1st correction coefficient calculated by the 1st correction coefficient calculation part. It is a figure which shows the example of the state of the pixel value of a local area | region when gradation conversion is performed using the 1st correction coefficient (alpha). It is an example of the 2nd correction coefficient which a 2nd correction coefficient calculation part computes. It is a figure which shows the example of the state of the pixel value of a local area | region when gradation conversion is performed using the 2nd correction coefficient (beta). It is a figure which shows the example of a setting of the 1st correction coefficient converted based on average brightness | luminance. It is a figure which shows the relationship between the average brightness | luminance performed in this embodiment, and the intensity | strength of the gradation conversion process by a 1st correction coefficient. It is a figure which shows the structural example of the image processing apparatus which concerns on the 2nd Embodiment of this invention. It is a figure which shows the relationship between the average brightness | luminance and peak brightness which are performed in 2nd Embodiment, and the intensity | strength of the gradation conversion process by a 1st correction coefficient. It is a figure which shows the structural example of the image processing apparatus which concerns on the 3rd Embodiment of this invention. It is a figure which shows an example of a brightness | luminance histogram. It is a figure which shows the relationship between the area of the low gradation area | region of a brightness | luminance histogram performed in 3rd Embodiment, the area of a high gradation area | region, and the intensity | strength of the gradation conversion process by a peak brightness | luminance and a 1st correction coefficient. It is a figure which shows the structural example of the image processing apparatus which concerns on the 4th Embodiment of this invention. It is a figure which shows the relationship between the variation | change_quantity of the edge histogram performed in 4th Embodiment, and the conversion process by noise reduction intensity | strength and a 1st correction coefficient. It is a figure which shows the structural example of the image processing apparatus which concerns on the 4th Embodiment of this invention. It is a figure which shows the relationship between the average brightness | luminance performed in this embodiment, the grade of the motion of an image, and the gradation conversion process by a noise reduction intensity | strength and a 1st correction coefficient. It is a figure for demonstrating the concept of a gradation conversion process by pixel deming.

(First embodiment)
FIG. 1 is a block diagram showing a configuration of a first embodiment of an image processing apparatus according to the present invention. The image processing apparatus of the present embodiment is applied to an image display apparatus provided with a display unit, and specifically, is configured as a television apparatus.
The present invention can be configured as an image processing device and an image display device including the image processing device.
The image display apparatus 100 of this example includes an image display unit 20 using a liquid crystal panel in which four RGBY sub-pixels (sub-pixels) are arranged at a resolution equivalent to 4K2K. In addition, a high-resolution image equivalent to 8K4K is displayed with high image quality.

  First, an analog SD (Standard Definition) image signal input from the tuner 1 or externally to the video input terminal 2 is selected by an analog SD input switching SW (switch) 7. The selected image signal is separated into a color signal and a luminance signal by the 3D (three-dimensional) YC separation unit 10, converted into a digital signal by the AD conversion unit 11, and input to the color space conversion unit 12.

In addition, an analog HD (High Definition) signal by a component signal input from the outside to the YPbPr terminal 3 or an RGB signal input from the outside to the PC input terminal 4 is selected by the analog HD input switching SW8. . The selected image signal is converted into a digital signal by the AD converter 11 and input to the color space converter 12.
A digital signal input from the outside to the HDMI (registered trademark) input terminals 5 and 6 is selected by the digital input switching SW 9, and the selected digital signal is input to the color space conversion unit 12.

The above-mentioned analog SD signal, analog HD signal, and digital signal are optimized for each signal path, and the color space conversion unit 12 performs color space conversion such as color space conversion of the RGB signal into a luminance color difference signal. Processing is performed and the result is output to the average luminance detection unit 13.
The average luminance detection unit 13 calculates an average value (average luminance) of luminance gradation values of each pixel in the frame for each frame of the input image signal, and outputs the calculated average luminance to the microcomputer control unit 21. To do. Further, the image signal whose average luminance is calculated is output to the noise reduction processing unit 14. The noise reduction processing unit 14 performs noise removal processing on the input image signal and outputs the processed image signal to the gradation conversion processing unit 15.

  The gradation conversion processing unit 15 performs gradation conversion processing by so-called pixel demming on the input image signal and outputs it to the video signal processing unit 16. The gradation conversion processing unit 15 raises the gradation value in the low gradation region while maintaining the local contrast by two-step gradation conversion processing, and further enhances the contrast to improve the image quality.

  The video signal processing unit 16 performs predetermined signal processing relating to image quality correction such as contrast, color tint, brightness, and sharpness on the image signal output from the gradation conversion processing unit 15 to provide three colors / 4 colors. Output to the conversion unit 17.

In the three-color / four-color conversion unit 17, image signals corresponding to three colors R (red), G (green), and B (blue) correspond to four colors R, G, B, and Y (yellow). The image signal is converted into an image signal and output to the 4K drive unit 18. The 4K driving unit 18 performs image signal processing for doubling the resolution in the horizontal direction with respect to a 4K2K input image signal (for example, signal processing for displaying one pixel in RGB and one pixel in BYR to separate luminance), Image signal processing for doubling the resolution in the vertical direction is performed by dividing and driving the upper and lower sub-pixels with a multi-pixel structure.
Further, in the case where the input image signal is a low-resolution image signal such as an SD image or an FHD (Full High Definition) image, the 4K driving unit 18 increases the resolution by performing upscaling such as so-called super-resolution processing. Thus, image signal processing for doubling the resolution in the horizontal and vertical directions is performed.

The image signal output from the 4K driving unit 18 is input to the liquid crystal timing control unit 19. In the liquid crystal timing control unit 19, the input image signal is converted into a format that can be displayed on the image display unit 20, and is displayed on the image display unit 20.
For example, the liquid crystal timing control unit 19 generates a clock signal or the like for allocating an image signal to the pixels of the liquid crystal panel of the image display unit 20 and supplies the clock signal to the data driver and gate driver of the liquid crystal panel. Then, an image based on the image signal is displayed.
Here, the resolution in the vertical direction is improved by dividing each color pixel into two in the vertical direction and changing the voltage-transmittance characteristics of the divided sub-pixels. Further, the LED driver 22 controls the lighting of the LED backlight included in the liquid crystal panel of the image display unit 20 in accordance with the display control by the liquid crystal timing control unit 19.

  The microcomputer control unit 21 to which the average luminance of the image signal detected by the average luminance detection unit 13 is input selects a table (LUT (Look up Table)) stored in the microcomputer control unit according to the average luminance of the image signal. The gradation conversion processing unit 15 outputs a table used for gradation conversion. The microcomputer control unit 21 stores a plurality of tables with different correction coefficients when the gradation conversion processing unit 15 performs gradation conversion processing, and corrects the correction value according to the average luminance of the image detected by the average luminance detection unit 13. The table used for gradation conversion using is automatically selected.

(Tone conversion processing in the gradation conversion processor)
In the embodiment according to the present invention, the correction coefficient used when the gradation conversion processing unit 15 performs the gradation conversion process is changed based on the average luminance detected by the average luminance detection unit 13, thereby changing the correction coefficient. Thus, the optimum image quality processing is performed. The gradation conversion processing applied to the embodiment according to the present invention will be specifically described below.

  FIG. 2 is a diagram for explaining a functional configuration example of the gradation conversion processing unit 15. The gradation conversion processing unit 15 includes a first illumination light distribution calculation unit 101, a first correction coefficient calculation unit 102, a first gradation conversion processing unit 103, a second illumination light distribution calculation unit 104, and a second correction coefficient calculation unit 105. And a second gradation conversion processing unit 106.

The first illumination light distribution calculation unit 101 calculates the illumination light distribution Y1 (first illumination light distribution) of each pixel for each frame of the image signal based on the image signal input to the gradation conversion processing unit 15. To do.
The illumination light is an index indicating the light reflected by the subject, the illumination light is calculated for each pixel in the frame, and the calculated two-dimensional distribution of the illumination light is defined as the illumination light distribution Y1. Here, the illumination light is obtained for each pixel as a value obtained by averaging the brightness Y of the reference pixel and the pixels in a predetermined area (first area) around the reference pixel.

The brightness Y is, for example, a gradation value indicating a lightness (Value) component indicated in an HSV (Hue, Saturation, Value) color space. That is, the brightness Y is a kind of gradation value for each pixel. In the present embodiment, the brightness Y may be a value indicating the brightness gradation for each pixel, and is not limited thereto.
For example, the first illumination light distribution calculation unit 101 may calculate the brightness Y based on the signal values of red, green, and blue indicated by RGB (Red, Green, Blue) color spaces. For example, the brightness Y is calculated as Y = 0.299 × R + 0.587 × G + 0.114 × B. Alternatively, in order to simplify the calculation, the first illumination light distribution calculation unit 101 calculates the brightness Y as Y = (R + G + B) / 3, Y = 0.25 × R + 0.5 × G + 0.25 × B, and the like. May be. Alternatively, the first illumination light distribution calculation unit 101 determines the maximum value (brightness indicated in the HSV color space) among the red signal value R, the green signal value G, and the blue signal value B as the brightness Y. Also good.

The averaging of the brightness Y in the first illumination light distribution calculation unit 101 may be, for example, a simple average or a weighted average between pixels in a predetermined area (first area) of the input image signal. Good.
The predetermined area (first area) only needs to have a size that allows the brightness to be compared within the predetermined area when the user gazes at the image display unit 20. For example, the size of the area is the image display unit. It can be appropriately adjusted according to the size of 20 (display size) and the assumed viewing distance of the user. The first illumination light distribution calculation unit 101 outputs the calculated illumination light distribution Y1 to the first correction coefficient calculation unit 102.

The first correction coefficient calculation unit 102 calculates the first correction coefficient α for each pixel based on the illumination light distribution Y1 input from the first illumination light distribution calculation unit 101.
FIG. 3 shows an example of the first correction coefficient calculated by the first correction coefficient calculation unit 102. In FIG. 3, the horizontal axis represents the gradation value of the first illumination light distribution Y1, and the vertical axis represents the value of the first correction coefficient α.
The first correction coefficient α calculated by the first correction coefficient calculation unit 102 corresponds to an increase in the gradation value of the illumination light distribution Y1 when the gradation value of the illumination light distribution Y1 is equal to or greater than a predetermined threshold Yth. The correction coefficient decreases and becomes a constant value when the gradation value of the illumination light distribution Y1 is less than the threshold value Yth.

When the gradation value of the illumination light distribution Y1 is equal to or greater than the predetermined threshold Yth, it can be considered that the greater the gradation value of the illumination light distribution Y1, the greater the amount of illumination light irradiated to the subject.
In that case, the image of the subject is brightly displayed and sufficient visibility is obtained. In such a case, the first correction coefficient calculation unit 102 sets the first correction coefficient α to a small value. The first correction coefficient α is 1 at the highest gradation value of the illumination light distribution Y1.

  In the gradation range equal to or greater than the threshold Yth, as the gradation value of the illumination light distribution Y1 decreases, the amount of illumination light applied to the subject becomes insufficient, and the subject image is displayed darkly. The visibility of the display image becomes insufficient. Accordingly, the first correction coefficient α is increased as the gradation value of the illumination light distribution Y1 becomes smaller at the gradation value equal to or greater than the threshold Yth. Thereby, the gradation value of a relatively dark area can be increased to improve visibility.

  In addition, the first correction coefficient calculation unit 102 calculates the first correction coefficient α having the largest and constant value in the entire gradation range in the region where the gradation value of the illumination light distribution Y1 is less than the predetermined threshold Yth. calculate. In a dark region less than the threshold Yth, visibility is improved by applying the maximum first correction coefficient α within the entire gradation range and performing gradation conversion. As an example, the threshold Yth can be set to 255 gradations when an image is expressed with 4096 gradations.

Since the illumination light intensity of the illumination light distribution Y1 is calculated by averaging the brightness Y of the target pixel and the surrounding pixels in a predetermined range, for two adjacent pixels, the gradation value of the illumination light distribution Y1 Are close to each other, and the first correction coefficient α is also close to each other.
That is, the relationship between the signal values of the two pixels after the gradation conversion is similar to the relationship of the signal values before the gradation conversion because the first correction coefficients α of the two adjacent pixels are close to each other. To be kept. Therefore, by determining the first correction coefficient α using the illumination light distribution Y1, the dark area is maintained while maintaining local contrast without degrading the image quality such as compressing the gradation of the bright region. The visibility of the area can be improved. The local contrast is a spatial variation in luminance within a predetermined range with a certain pixel as a reference.

The first correction coefficient calculation unit 102 outputs the calculated first correction coefficient α to the first gradation conversion processing unit 103 and the second correction coefficient calculation unit 105.
The first gradation conversion processing unit 103 applies the first correction coefficient to the image signal input to the gradation conversion processing unit 15, performs gradation conversion processing, and generates a first gradation conversion image signal. .
In the gradation conversion process, the relationship between the signal value (input gradation value) P0 for each pixel included in the input image signal and the converted signal value (output gradation value) P1 is P1 = P0 × α. To do. When the input image is a color image represented by signal values of each color such as red, green, and blue, the first gradation conversion processing unit 103 uses the same correction coefficient α for each color signal value. Multiply Thereby, gradation conversion can be performed without changing the ratio of signal values of each color, that is, the hue.

  In addition, the first gradation conversion processing unit 103 separates the signal value of the input image signal into a brightness component indicating brightness and a color component indicating color, and only the brightness component is multiplied by the correction coefficient α. Tone conversion may be performed. For example, the first gradation conversion processing unit 103 converts the signal value of each pixel indicated in the RGB color space into a signal value indicated in the HSV color space, and multiplies only the lightness component signal value by the correction coefficient α. It is also possible to perform tone conversion.

  FIG. 4 is a diagram illustrating an example of the state of pixel values in a local area when gradation conversion is performed using the first correction coefficient α, where the horizontal axis indicates the position in the image, and the vertical axis indicates the pixel value. Indicates. 4A shows the state of the pixel value of the input image, and FIG. 4B shows the state of the pixel value of the image output by the first gradation conversion processing unit 103. Since these are local regions, the illumination light distribution is substantially the same, that is, the correction coefficients are also substantially the same. Therefore, the change from FIG. 4A to FIG. 4B is a change in which the brightness of the region is improved while the fine pixel value variation is maintained.

The first gradation conversion processing unit 103 outputs the first gradation conversion image signal generated by performing the gradation conversion processing to the second illumination light distribution calculation unit 104 and the second gradation conversion processing unit 106.
The second illumination light distribution calculation unit 104 calculates an illumination light distribution Y2 (second illumination light distribution) for each pixel based on the first gradation conversion image signal output from the first gradation conversion processing unit 103. . Here, similarly to the first illumination light distribution calculation unit 101, the second illumination light distribution calculation unit 104 is a pixel in each pixel indicated by the first gradation conversion image signal and a predetermined region (second region) around the pixel. Average brightness between. The calculation method of the brightness Y in the second illumination light distribution calculation unit 104 can be the same as the calculation method of the brightness Y in the first illumination light distribution calculation unit 101.

  At this time, the size of the predetermined area (second area) of the pixels averaged by the second illumination light distribution calculation unit 104 is equal to the predetermined area (first area) of the pixels averaged by the first illumination light distribution calculation unit 101. ) Is smaller than the size of. That is, the second illumination light distribution Y2 is less affected by the peripheral pixels due to the gradation value averaging than the first illumination light distribution Y1, and the influence of the gradation value of the pixel of interest is greater than the first illumination light distribution. Become.

The second illumination light distribution calculation unit 104 outputs the calculated illumination light distribution Y2 to the second correction coefficient calculation unit 105. The second correction coefficient calculation unit 105 calculates the correction coefficient β for each pixel based on the illumination light distribution Y2 output from the second illumination light distribution calculation unit 104.
FIG. 5 is an example of the second correction coefficient calculated by the second correction coefficient calculation unit 105. In FIG. 5, the horizontal axis is the gradation value of the second illumination light distribution Y2, and the vertical axis is the value of the second correction coefficient β.
The second correction coefficient calculation unit 105 calculates a second correction coefficient β that monotonously increases with respect to the gradation value indicated by the illumination light distribution Y2. Here, the second correction coefficient calculation unit 105 sets the second correction coefficient β to 1 at the maximum value of the gradation value of the illumination light distribution Y2, and goes from the maximum gradation value toward the low gradation side. The second correction coefficient β is set so as to decrease.

The pixel region (second region) for which the second illumination light distribution calculation unit 104 calculates the average value is smaller than the pixel region (first region) for which the first illumination light distribution calculation unit 101 calculates the average value. Accordingly, the second correction coefficient β of the target pixel is strongly influenced by the gradation value of the target pixel, and the difference between the second correction coefficients β of adjacent pixels is increased.
As a result, in the output image obtained by performing gradation conversion using the second correction coefficient, the maximum value and the minimum value of the image signal before the second correction coefficient is applied are determined after the second correction coefficient is applied. Although both decrease, the difference between the maximum and minimum values increases and the local contrast is enhanced.

  FIG. 6 is a diagram illustrating an example of a state of pixel values in a local area when gradation conversion is performed using the second correction coefficient β, where the horizontal axis indicates the position in the image, and the vertical axis indicates the pixel value. Indicates. 6A shows the state of the pixel value of the image output from the first gradation conversion processing unit 103, and FIG. 6B shows the state of the pixel value of the image output from the second gradation conversion processing unit 106. State. Since the second region to be referred to when calculating the illumination light distribution Y2 is smaller than the first region, the variation of each pixel is increased and the local contrast is enhanced.

  In the example of FIG. 2, after the tone conversion is performed on the input image signal using the first correction coefficient by the first tone conversion processing unit 103, the second correction is further performed by the second tone conversion processing unit 106. The tone conversion is performed using the coefficients, but the first correction coefficient and the second correction coefficient are calculated based on the input image signal, and the first correction coefficient and the second correction coefficient are combined. Three correction coefficients may be generated, and gradation conversion may be performed by applying the third correction coefficient to the input image signal.

(Gradation conversion processing based on the average luminance of the image signal)
In the present embodiment, the first correction coefficient calculation unit 102 changes the value of the first correction coefficient in the gradation conversion processing unit 15 according to the average luminance for each frame of the image signal detected by the average luminance detection unit 13. The first correction coefficient based on the average luminance detected by the average luminance detection unit 13 is instructed from the microcomputer control unit 21 to the gradation conversion processing unit 15 based on a table selected by the microcomputer control unit 21 based on the average luminance.
Specifically, the first correction coefficient calculation unit 102 relatively lowers the value of the first correction coefficient that is less than the threshold Yth for an image signal having a relatively low average luminance. At this time, the first correction coefficient calculation unit 102 divides the average luminance level detected by the average luminance detection unit 13 into a plurality of levels, and the lower the average luminance, the lower the first correction coefficient less than the threshold Yth. Lower the value.

  FIG. 7 is a diagram illustrating a setting example of the first correction coefficient to be converted based on the average luminance. Here, the first correction coefficient α is determined according to the three steps of “strong”, “medium”, and “weak” in the gradation conversion processing using the first correction coefficient. FIG. 7A shows an example in which the gradation conversion process using the first correction coefficient is “strong”, and the first correction coefficient α less than the threshold Yth is set to the highest level. Above the threshold Yth, the first correction coefficient α decreases as the gradation value of the illumination light distribution Y1 increases, and the first correction coefficient α becomes 1 at the highest gradation of the illumination light distribution Y1. When the gradation conversion process is “strong”, the amount of enhancement (enhancement) of the low gradation area of the image signal is the largest.

FIG. 7B shows a case where the gradation conversion process using the first correction coefficient is “medium”, and the first correction coefficient less than the threshold Yth is set smaller than the case of “strong”. Also in this case, the first correction coefficient is 1 at the highest gradation of the illumination light distribution Y1.
FIG. 7C shows a case where the gradation conversion process using the first correction coefficient is “weak”, and the first correction coefficient less than the threshold Yth is set to be smaller than that in the “medium” case. Also in this case, the first correction coefficient is 1 at the highest gradation of the illumination light distribution Y1. When the gradation conversion processing by the first correction coefficient is “weak”, the amount of lifting (enhancement) of the low gradation region of the image signal is the smallest.
When the first correction coefficient α changes stepwise, the second correction coefficient β also changes accordingly. Basically, when the first correction coefficient α increases, the second correction coefficient β decreases to compensate for it. Here, the second correction coefficient β is appropriately determined according to the first correction coefficient α.

FIG. 8 is a diagram showing the relationship between the average luminance executed in the present embodiment and the intensity of the gradation conversion process using the first correction coefficient.
The first correction coefficient calculation unit 102 classifies the average luminance that can be taken by the image signal into three stages of “low”, “medium”, and “high” in advance, and the first correction coefficient when the average luminance is “low”. The gradation conversion process is set to “weak”, the gradation conversion process using the first correction coefficient is set to “medium” when the average luminance is “medium”, and the first correction coefficient is used when the average luminance is “high”. Set the key conversion process to “strong”.

When the average luminance of the image signal is low, there are many low-tone noise components. Therefore, increasing the first correction coefficient makes the noise components stand out. Therefore, when the average luminance of the image signal is low, the gradation conversion process using the first correction coefficient is set to “weak” so that noise is not noticeable.
In addition, when the average luminance of the image signal is high, the noise component in the low gradation region of the image signal is also small. Therefore, the gradation conversion process using the first correction coefficient is set to “strong” and the effect of the gradation conversion process is improved. Increase image quality to the maximum. Further, when the average luminance of the image signal is medium, the gradation conversion processing by the first correction coefficient is set to “medium” in order to obtain an intermediate effect between the case where the average luminance is low and the case where the average luminance is high. To do.
Thus, by changing the first correction coefficient during the gradation conversion process according to the average luminance of the image signal, the optimum gradation conversion process can be performed according to the state of the noise component in the image, The image quality can be appropriately improved according to the characteristics of the image signal.

(Embodiment 2)
FIG. 9 is a diagram illustrating a configuration example of an image processing apparatus according to the second embodiment of the present invention. In the present embodiment, in addition to the configuration of the first embodiment, a peak luminance detection unit 23 that detects the peak luminance for each frame of the input image signal is provided. The peak luminance detected by the peak luminance detection unit 23 is output to the microcomputer control unit 21 together with the average luminance detected by the average luminance detection unit 13. Further, the image signal whose peak luminance is detected is output to the noise reduction processing unit 14.
The other configuration of FIG. 9 is the same as that of the embodiment shown in FIG. 1, and the repeated description thereof will be omitted except for the portion characterizing this embodiment. Further, the configuration of the gradation conversion processing unit 15 is the same as the configuration of FIG. 2 of the first embodiment.

In the present embodiment, the first correction coefficient calculation unit 102 changes the value of the first correction coefficient α less than the threshold according to the average luminance detected by the average luminance detection unit 13 and the peak luminance detected by the peak luminance detection unit 23. Let The average luminance detected by the average luminance detector 13 and the first correction value based on the peak luminance detector 23 are instructed from the microcomputer controller 21 to the gradation conversion processor 15 based on the table selected by the microcomputer controller 21. The
Specifically, the first correction coefficient calculation unit 102 determines that the average luminance is relative except when the average luminance detected by the average luminance detection unit 13 is lower than a predetermined level and the peak luminance level is equal to or higher than the predetermined level. For an image signal that is very low, the value of the first correction coefficient that is less than the threshold value is relatively low. Further, when the average luminance is lower than the predetermined level and the peak luminance level is equal to or higher than the predetermined level, the value of the first correction coefficient that is less than the threshold is relatively increased.

A specific example of the above is shown in FIG. FIG. 10 is a diagram illustrating a relationship between the average luminance and the peak luminance executed in the present embodiment and the intensity of the gradation conversion process using the first correction coefficient.
In the first correction coefficient calculation unit 102, the average luminance that can be taken by the image signal is divided into three stages of “low”, “medium”, and “high” in advance. The peak luminance is divided into two stages of “low” and “high” in advance. When the peak luminance is high, for example, when the gradation value of an image signal expressed by 4096 gradations is 3900 or more, it can have a high peak luminance ground value.
In the gradation conversion process, it is assumed that first correction coefficients corresponding to three levels of “strong”, “medium”, and “weak” are determined in advance. It is assumed that the three-stage first correction coefficients are the same as those in the example of FIG.

  As shown in FIG. 10, when the average luminance is “high”, the gradation conversion process using the first correction coefficient is set to “strong” regardless of the level of the peak luminance. When the average luminance of the image signal is high, the noise component in the low gradation area is also small, so the gradation conversion process using the first correction coefficient is set to “strong” to maximize the effect of the gradation conversion process. To improve image quality. When the average luminance of the image signal is medium, the gradation conversion process using the first correction coefficient is performed in order to obtain an intermediate effect between when the average luminance of the peak luminance is low and when the average luminance is high. "Middle".

When the average luminance is low, the gradation conversion processing is different depending on whether the peak luminance is “high” or “low”. That is, when the average luminance is “low” and the peak luminance is “low”, there are many noise components in the low gradation region, so that the noise components become conspicuous when the first correction coefficient is increased. In addition, since the peak luminance is not so high at this time, priority is given to suppressing the first correction coefficient and making the noise component inconspicuous rather than making the peak luminance bright.
On the other hand, when the average luminance is “low” and the peak luminance is “high”, for example, the image is dark like the starry sky, but has the brightness of stars. When the correction coefficient is reduced, the brightness of an image having a peak luminance (for example, a star) is also suppressed. Therefore, when the average luminance is low and the peak luminance is high, the gradation conversion process is set to “strong” by the first correction coefficient to emphasize the brightness of the peak luminance.

  As described above, by changing the first correction coefficient at the time of gradation conversion processing according to the average luminance and peak luminance of the image signal, an optimum level is selected according to the state of noise components and the peak state in the image. Tone conversion processing can be performed, and high image quality can be appropriately achieved according to the characteristics of the image signal.

(Embodiment 3)
FIG. 11 is a diagram illustrating a configuration example of an image processing apparatus according to the third embodiment of the present invention. In the present embodiment, instead of the average luminance detection unit 13 of the first embodiment, a luminance histogram detection unit 24 that detects a luminance histogram obtained by integrating the number of pixels for each luminance value is provided for each frame of the input image signal. Yes. The luminance histogram detected by the luminance histogram detection unit 24 is output to the microcomputer control unit 21. Further, the image signal from which the luminance histogram is detected is output to the noise reduction processing unit 14.
Other configurations in FIG. 11 are the same as those in the embodiment shown in FIGS. 1 and 2, and thus repeated description thereof is omitted except for portions characterizing the present embodiment.

In the present embodiment, the first correction coefficient calculation unit 102 changes the value of the first correction coefficient less than the threshold according to the luminance histogram detected by the luminance histogram detection unit 24. The luminance histogram detected by the luminance histogram detection unit 24 is instructed from the microcomputer control unit 21 to the gradation conversion processing unit 15 based on the table selected by the microcomputer control unit 21.
FIG. 12 shows an example of the luminance histogram. The luminance histogram is obtained by integrating the number of pixels for each luminance value of each pixel of the image signal as described above. In the present embodiment, the area of a predetermined low gradation area and the area of a predetermined high gradation area are detected based on the luminance histogram, and the value of the first correction coefficient is converted based on the detected area.

As a specific example of changing the value of the first correction coefficient according to the luminance histogram, the first correction coefficient calculation unit 102 includes a predetermined low gradation region of the luminance histogram detected by the luminance histogram detection unit 24, a predetermined Based on the area of the high gradation area,
(1) In the predetermined range in which the area of the predetermined low gradation region is the smallest, the value of the first correction coefficient less than the threshold value is set to be relatively largest,
(2) In the predetermined range where the area of the predetermined low gradation region is an intermediate level, when there is a predetermined high gradation region, the value of the first correction coefficient less than the threshold is set to the intermediate level relatively, When there is no high gradation area, the value of the first correction coefficient that is less than the threshold is set to the lowest value,
(3) In the predetermined range in which the area of the predetermined low gradation region is the largest, when there is a predetermined high gradation region, the value of the first correction coefficient less than the threshold value is relatively increased and the predetermined high gradation is When there is no area, the value of the first correction coefficient that is less than the threshold value is set to the lowest.

A specific example of the above is shown in FIG. FIG. 13 is a diagram illustrating the area of the low gradation region and the area of the high gradation region of the luminance histogram executed in the present embodiment, and the relationship between the peak luminance and the intensity of the gradation conversion process using the first correction coefficient.
Here, predetermined low gradation regions, intermediate gradation regions, and high gradation regions are determined for the luminance values that can be taken by the image signal detected by the luminance histogram. As an example, in an image signal represented by 0 to 4095 gradations, the low gradation area is an area of 447 gradations or less, the intermediate gradation area is an area of 448 to 3422 gradations, and the high gradation area is 3423 gradations. The above area.

Then, the range of the size of the area is divided into three stages of “large”, “medium”, and “small” in advance. As an example, if the area of each gradation region is 75% or more, the area is “large”, if it is in the range of 75 to 10%, the area is “medium”, and if it is less than 10%, the area is “ It can be determined to be “small”. In addition, if there is a pixel in the low gradation region or the high gradation region, it is “present”, and if it is not, “not present”.
In the gradation conversion process, it is assumed that first correction coefficients corresponding to three levels of “strong”, “medium”, and “weak” are determined in advance. It is assumed that the three stages of the first correction coefficients are the same as those in the example of FIG.

As shown in FIG. 13, when the area of the low gradation region is “small”, the gradation conversion processing by the first correction coefficient is set to “strong” regardless of the area of the high gradation region. When the area of the low gradation region of the image signal is small, the noise component in the low gradation region is also small. Therefore, the gradation conversion processing by the first correction coefficient is set to “strong”, and the effect of the gradation conversion processing is improved. Increase image quality to the maximum.
Further, when the area of the low gradation region is “medium”, the gradation conversion process is different depending on whether the high gradation region is “present” or “not present”. Here, when the area of the low gradation region is “medium” and the high gradation region is “present”, an intermediate effect between the case where the area of the low gradation region is small and the case where it is large Therefore, the gradation conversion process using the first correction coefficient is set to “medium”.
On the other hand, when the area of the low gradation region is “medium” and the high gradation region is “none”, the first correction coefficient is used to suppress the enhancement of the noise component on the low gradation side. The gradation conversion processing by is set to “weak” so that noise components are not noticeable.

  When the area of the low gradation region is “large”, the gradation conversion process is different depending on whether the high gradation region is “present” or “not present”. Here, when the area of the low gradation region is “large” and the high gradation region is “present”, the gradation conversion processing by the first correction coefficient is set to “strong”. When the area of the low gradation area is large and there is a high gradation area, the entire screen is dark, but the image has a high gradation part in it. In this case, if the first correction coefficient is reduced, In addition, the brightness of the high gradation part is also suppressed. Therefore, when the low gradation area is large and there is a high gradation area, the gradation conversion process is set to “strong” by the first correction coefficient to emphasize the brightness of the high gradation area.

  In addition, when the area of the low gradation region is “large” and the high gradation region is “none”, the gradation conversion process using the first correction coefficient is set to “weak”. When the area of the low gradation area is large and there is no high gradation area, it is a dark image without the high gradation part. In this case, the first correction coefficient is suppressed so that the noise component is not noticeable.

  As described above, by changing the first correction coefficient at the time of gradation conversion processing according to the area of the high gradation region and the low gradation region of the image signal or the presence / absence of the noise component in the image and the brightness An optimum gradation conversion process can be performed according to the state, and an image quality can be appropriately improved according to the characteristics of the image signal.

(Embodiment 4)
FIG. 14 is a diagram illustrating a configuration example of an image processing apparatus according to the fourth embodiment of the present invention. In the present embodiment, an edge histogram detection unit 25 is provided instead of the average luminance detection unit having the configuration of the first embodiment.
For each frame of the image signal, the edge histogram detection unit 25 obtains a luminance difference between adjacent pixels that is an absolute value of a luminance difference between adjacent pixels for each pixel constituting the frame, and a pixel for each luminance difference between adjacent pixels. An edge histogram indicating the number is created, and the amount of change between the frames of the edge histogram is detected. As the amount of change, for example, the amount of change in the edge histogram can be obtained by integrating the difference between the same luminance differences between adjacent pixels with respect to the luminance difference between all adjacent pixels.

The change amount of the edge histogram detected by the edge histogram detection unit 25 is output to the microcomputer control unit 21. Further, the image signal in which the change amount of the edge histogram is detected is output to the noise reduction processing unit 14. The microcomputer control unit 21 controls the noise removal strength in the noise reduction processing unit 14 and the magnitude of the first correction coefficient in the gradation conversion processing unit 15 based on the amount of change in the edge histogram.
Other configurations in FIG. 14 are the same as those in the embodiment shown in FIGS. 1 and 2, and repeated description thereof is omitted except for portions characterizing this embodiment.

In the present embodiment, the first correction coefficient calculation unit 102 changes the value of the first correction coefficient less than the threshold according to the amount of change in the edge histogram detected by the edge histogram detection unit 25, and the noise reduction processing unit 14 performs noise reduction. Vary the intensity.
The intensity of the noise reduction in the noise reduction processing unit 14 and the gradation conversion processing by the first correction coefficient in the gradation conversion processing unit 15 are based on the change amount of the edge histogram detected by the edge histogram detection unit 25, and the microcomputer control unit 21 Controlled by

  Specifically, the first correction coefficient calculation unit 102 relatively increases the value of the first correction coefficient less than the threshold for an image signal in which the change amount of the edge histogram detected by the edge histogram detection unit 25 is relatively small. The noise reduction processing unit 14 relatively weakens the noise removal strength for the image signal in which the change amount of the edge histogram detected by the edge histogram detection unit 25 is relatively small.

An example of the above is shown in FIG. FIG. 15 is a diagram illustrating the relationship between the change amount of the edge histogram executed in the present embodiment, and the conversion process using the noise reduction intensity and the first correction coefficient.
Here, the level of the amount of change in the edge histogram is defined in advance as to whether it is “large” or “small”. When the amount of change in the edge histogram is large, for example, when there is noise that causes flickering for each frame, if there is noise, the edge histogram detected from the frame varies greatly between frames, and the edge histogram The amount of change increases. The amount of change is classified in advance at a predetermined level. When the amount is greater than or equal to the predetermined level, the amount of change is “large”. When the amount is less than the predetermined level, the amount of change is “small”.

The noise reduction intensity is a value that defines the degree of reduction of the noise component of the image signal. The intensity of noise reduction represents the intensity of influence of surrounding pixels, for example. As the intensity increases, the output value of the target pixel approaches the pixel values of the surrounding pixels. In the present embodiment, the noise reduction method is not particularly limited.
Here, the intensity of noise reduction is determined in advance by dividing it into “strong” and “weak” stages. In the gradation conversion process, it is assumed that first correction coefficients corresponding to three levels of “strong”, “medium”, and “weak” are determined in advance. It is assumed that the three stages of the first correction coefficients are the same as those in the example of FIG.

As shown in FIG. 15, when the amount of change in the edge histogram is “large”, the noise reduction strength is set to “strong”, and the gradation conversion processing by the first correction coefficient is set to “weak”. When the change amount of the edge histogram is large, it is considered that the image has noise. Therefore, the noise reduction strength is increased to enhance the noise removal effect. At this time, the gradation conversion process using the first correction coefficient is weakened to suppress the enhancement of noise in the low gradation region.
On the other hand, when the amount of change in the edge histogram is “small”, the noise reduction strength is set to “weak” and the tone conversion processing by the first correction coefficient is set to “strong”. When the amount of change in the edge histogram is small, it is considered that the image is low in noise. Therefore, the intensity of noise reduction is reduced to suppress degradation in resolution due to noise reduction. In addition, since the image is low in noise, the tone conversion process using the first correction coefficient is strengthened to maximize the effect of the tone conversion process and improve the image quality.

  In this way, by changing the first correction coefficient at the time of gradation conversion processing and the intensity of noise reduction in accordance with the amount of change in the edge histogram of the image signal, the optimum level according to the state of the noise component in the image. Tone conversion processing can be performed, and high image quality can be appropriately achieved according to the characteristics of the image signal.

(Embodiment 5)
FIG. 16 is a diagram illustrating a configuration example of an image processing apparatus according to the fourth embodiment of the present invention. In the present embodiment, in addition to the average luminance detection unit 13 having the configuration of the first embodiment, a motion detection determination unit 26 that detects the motion of an image and determines the degree of the motion is provided.
The motion detection determination unit 26 detects the motion of the image signal, determines the degree of motion, and outputs the determination result to the microcomputer control unit 21. A known technique can be applied as appropriate to the determination of the degree of movement. For example, a motion vector for each block between frames can be calculated using a block matching method or the like, and the degree of motion of the video can be classified according to the magnitude of the motion vector. Alternatively, the accumulation of pixel values in a frame can be compared between frames, and the state of image movement can be determined based on the difference. At this time, an accumulation difference of pixel values may be calculated for each block in the frame.
Here, it is determined whether the image signal is a completely still image or a moving image, and in the case of a moving image, the degree of movement can be further divided into a plurality of stages. it can.

The degree of motion of the image determined by the motion detection determination unit 26 is output to the microcomputer control unit 21 together with the average luminance detected by the average luminance detection unit 13. Further, the image signal whose motion is detected is output to the noise reduction processing unit 14. The microcomputer control unit 21 controls the noise reduction intensity in the noise reduction processing unit 14 and the gradation change processing in the gradation conversion processing unit 15 based on the determination result of the average luminance and the degree of motion.
The other configuration of FIG. 16 is the same as that of the embodiment shown in FIGS. 1 and 2, and the repeated description thereof is omitted except for the portion characterizing this embodiment.

In the present embodiment, the first correction coefficient calculation unit 102 changes the value of the first correction coefficient less than the threshold according to the degree of motion determined by the motion detection determination unit 26, and the noise reduction processing unit 14 performs noise reduction intensity. To change.
The intensity of noise reduction in the noise reduction processing unit 14 and the gradation conversion processing in the gradation conversion processing unit 15 are controlled by the microcomputer control unit 21 based on the degree of image motion determined by the motion detection determination unit 26.

Specifically, the first correction coefficient calculation unit 102 calculates the first correction coefficient less than the threshold for an image signal with a relatively large degree of motion determined by the motion detection determination unit 26 at the same average luminance level. While the value is relatively lowered, the noise reduction processing unit 14 increases the noise reduction strength by the two-dimensional noise reduction function and weakens the noise reduction strength by the three-dimensional noise reduction function.
In addition, when the degree of motion is the same level, the first correction coefficient calculation unit 102 sets the value of the first correction coefficient that is less than the threshold for an image signal having a relatively low average luminance detected by the average luminance detection unit 13. Make it relatively low.

A specific example of the above is shown in FIG. FIG. 17 is a diagram illustrating the relationship between the average luminance and the degree of image motion executed in the present embodiment, and the tone conversion process using the noise reduction intensity and the first correction coefficient.
Here, first, whether the degree of motion is “still image”, a slow motion video (referred to as “video (slow)”), or a fast motion video (referred to simply as “video”). It prescribes. Whether a moving image is fast moving or slow moving is determined by comparing the result of motion detection with a predetermined level.
As for the average luminance detected by the average luminance detector 13, the average luminance that can be taken by the image signal is divided into “low”, “medium”, and “high” levels in advance.

The noise reduction processing unit 14 can execute both functions of known two-dimensional noise reduction (2D NR) and three-dimensional noise reduction (3D NR). In the two-dimensional noise reduction, a noise component is detected using a line correlation within one frame, and a spatial filter process is performed to remove noise from the output image. In the three-dimensional noise reduction, noise is detected and removed using frame correlation in the time axis direction. The intensity of noise reduction is a value that defines the degree of reduction of the noise component of the image signal.
Here, the intensity of noise reduction is determined in advance for each of the two-dimensional noise reduction and the three-dimensional noise reduction, divided into “strong”, “medium”, and “weak” stages.

In the gradation conversion process, first correction coefficients corresponding to three levels of “strong”, “medium”, and “weak” are determined in advance. It is assumed that the three stages of the first correction coefficients are the same as those in the example of FIG. Furthermore, “strong 1”, “strong 2”, and “strong 3” are set in “strong”. Similarly for “medium” and “weak”, “medium 1”, “medium 2”, “medium 3”, “weak 1”, “weak 2”, and “weak 3” are set, respectively.
“Strong 1”, “Strong 2”, and “Strong 3” are obtained by adding the first correction coefficient corresponding to “Strong” in FIG. Yes, the first correction coefficient less than the threshold Yth has the smallest “strong 1” and the largest “strong 3”. The same applies to “medium” and “weak”.

  As shown in FIG. 17, when the average luminance is the same level and the image is a moving image with fast motion, the strength of the two-dimensional noise reduction is increased and the strength of the three-dimensional noise reduction is decreased. For a slow moving video, the intensity of 2D noise reduction and 3D reduction is moderate. Further, in the case of a still image, the intensity of two-dimensional noise reduction is reduced and the intensity of three-dimensional noise reduction is increased.

  In order to maximize the effect of the gradation conversion processing unit 15, it is preferable that a noise component is not included in the image signal input to the gradation conversion processing unit 15. Therefore, it is preferable to remove noise in front of the gradation conversion processing unit 16, but to sufficiently remove noise, three-dimensional noise reduction having a high noise removal function is used. However, when the function is performed with the intensity of three-dimensional noise reduction being increased, there are problems such as blurring of the image and afterimages in a moving image. On the other hand, if two-dimensional noise reduction is used, the image blurring feeling and the afterimage feeling due to noise reduction are reduced, but the noise removal function is lowered.

  Therefore, in an image with a large motion, if the intensity of the three-dimensional noise reduction is increased, image degradation occurs. Therefore, the intensity of the three-dimensional noise reduction is decreased and the intensity of the two-dimensional noise reduction is increased. In still images, noise components can be sufficiently removed by three-dimensional noise reduction, so that the strength of three-dimensional noise reduction is increased and the strength of two-dimensional noise reduction is decreased. The slow moving image has an intermediate intensity between the fast moving moving image and the still image.

In the gradation change process using the first correction coefficient, when the average luminance level is the same, a fast moving video is set to “weak”, a slow moving video is set to “medium”, and a still image is set to “strong”.
As described above, in the case of a fast moving video, the 3D noise reduction cannot be applied strongly, and the noise component may not be sufficiently removed. Therefore, the gradation conversion process using the first correction coefficient is set to “weak”. Therefore, make noise components inconspicuous.
In addition, since the motion of slow motion can be stronger than the motion of fast motion, the intensity of three-dimensional noise reduction is set to “medium” by the first correction coefficient to enhance the image quality improvement effect.

In still images, the intensity of three-dimensional noise reduction can be increased and noise components can be removed. Therefore, the gradation conversion process using the first correction coefficient is set to “strong”, and the effect of the gradation conversion process is improved. To make the most of it.
As described above, by dynamically switching between the noise reduction intensity and the tone conversion process according to the degree of movement, it is possible to remove noise according to the state of the image and improve the image quality by the tone conversion process. .

Then, the degree of gradation conversion processing by the first correction coefficient is further varied according to the average luminance level. When the degree of movement is the same, the value of the first correction coefficient less than the threshold is lowered as the average luminance is lower.
That is, in the case of a fast-moving moving image, when the average brightness is low, the gradation conversion processing is “weak 1” with the first correction coefficient less than the threshold being the smallest among the “weak”, and the average brightness is medium. When the average luminance is the highest, the first correction coefficient less than the threshold is “weak 2”. When the average brightness is the highest, the first correction coefficient less than the threshold is “weak” and the first correction value less than the threshold is the largest “weak 3”. "

Similarly, for a slow-moving video, as the average brightness increases from the lower one, it becomes “medium 1”, “medium 2”, “medium 3”, and the first correction coefficient below the threshold value is the same medium. However, it is the smallest when the average luminance is low and the largest when the average luminance is high.
Similarly, in the case of still images, as the average luminance increases from the lowest, the values become “strong 1”, “strong 2”, and “strong 3”, and the first correction coefficient less than the threshold is of the same strong level. However, it is the smallest when the average luminance is low and the largest when the average luminance is high.

When the average luminance of the image signal is low, there are many low-tone noise components. Therefore, increasing the first correction coefficient makes the noise components stand out. Therefore, if the degree of motion is the same level, the first correction coefficient is lowered so that the noise is not noticeable as the average luminance of the image signal is lower.
As described above, by changing the first correction coefficient at the time of gradation conversion processing and the intensity of noise reduction according to the degree of movement of the image signal and the average luminance of the image signal, the state of the noise component in the image is changed. Accordingly, optimum gradation conversion processing can be performed, and high image quality can be appropriately achieved according to the characteristics of the image signal.

  The technical features (components) described in each of the above embodiments can be combined with each other, and new technical features can be formed by combining them.

DESCRIPTION OF SYMBOLS 1 ... Tuner, 2 ... Video input terminal, 3 ... YPbPr terminal, 4 ... PC input terminal, 5 ... HDMI input terminal, 8 ... Analog HD input switching SW, 9 ... Digital input switching SW, 10 ... YC separation part, 11 ... AD conversion unit, 12 ... color space conversion unit, 13 ... average luminance detection unit, 14 ... noise reduction processing unit, 15 ... gradation conversion processing unit, 16 ... gradation conversion processing unit, 16 ... video signal processing unit, 17 ... 3 color / 4 color conversion unit, 18 ... 4K drive unit, 19 ... liquid crystal timing control unit, 20 ... image display unit, 21 ... microcomputer control unit, 22 ... LED driver, 23 ... peak luminance detection unit, 24 ... luminance histogram detection , 25 ... edge histogram detection unit, 26 ... motion detection determination unit, 100 ... image display device, 101 ... first illumination light distribution calculation unit, 102 ... first correction coefficient calculation unit, 10 ... first gradation conversion processing unit, 104 ... second illumination light distribution calculation unit, 105 ... second correction coefficient calculating unit, 106 ... second gradation conversion processing unit.

Claims (16)

For each pixel of the input image signal, the brightness of each pixel in the first predetermined area around the pixel is averaged to calculate a first illumination light intensity that serves as an index indicating the intensity of reflected light from the subject, A first illumination light distribution calculating unit for calculating a distribution of the first illumination light intensity of the input image signal;
For each pixel of the input image signal, a constant level is maintained in a region where the gradation value of the first illumination light intensity is less than a predetermined threshold value, and a region where the gradation value of the first illumination light intensity is greater than or equal to the threshold value Then, a first correction coefficient calculation unit that calculates a first correction coefficient that decreases as the gradation value of the first illumination light increases;
Second illumination light that calculates the second illumination light intensity distribution by averaging the brightness of each pixel in the second predetermined area smaller than the first predetermined area from the image signal corrected by the first correction coefficient. A distribution calculator;
A second correction coefficient calculation unit that calculates a second correction coefficient that decreases as the gradation value of the second illumination light intensity decreases;
A gradation conversion processing unit that performs gradation conversion by applying the first correction coefficient and the second correction coefficient to an input image signal,
An average luminance detection unit that detects the average luminance for each frame of the input video signal,
The image processing apparatus according to claim 1, wherein the first correction coefficient calculation unit changes a value of the first correction coefficient less than the threshold according to the average luminance detected by the average luminance detection unit. The image processing apparatus according to claim 1.
The image processing apparatus according to claim 1, wherein the first correction coefficient calculation unit relatively lowers the value of the first correction coefficient that is less than the threshold for an image signal having a relatively low average luminance. The image processing apparatus according to claim 2,
The first correction coefficient calculation unit divides the average luminance level detected by the average luminance detection unit into a plurality of stages, and the lower the average luminance, the lower the value of the first correction coefficient less than the threshold value. An image processing apparatus. For each pixel of the input image signal, the brightness of each pixel in the first predetermined area around the pixel is averaged to calculate a first illumination light intensity that serves as an index indicating the intensity of reflected light from the subject, A first illumination light distribution calculating unit for calculating a distribution of the first illumination light intensity of the input image signal;
For each pixel of the input image signal, a constant level is maintained in a region where the gradation value of the first illumination light intensity is less than a predetermined threshold value, and a region where the gradation value of the first illumination light intensity is greater than or equal to the threshold value Then, a first correction coefficient calculation unit that calculates a first correction coefficient that decreases as the gradation value of the first illumination light increases;
Second illumination light that calculates the second illumination light intensity distribution by averaging the brightness of each pixel in the second predetermined area smaller than the first predetermined area from the image signal corrected by the first correction coefficient. A distribution calculator;
A second correction coefficient calculation unit that calculates a second correction coefficient that decreases as the gradation value of the second illumination light intensity decreases;
A gradation conversion processing unit that performs gradation conversion by applying the first correction coefficient and the second correction coefficient to an input image signal,
An average luminance detector that detects the average luminance of each frame of the input image signal;
A peak luminance detection unit for detecting the peak luminance for each frame of the input image signal,
The first correction coefficient calculation unit changes the value of the first correction coefficient less than the threshold according to the average luminance detected by the average luminance detection unit and the peak luminance detected by the peak detection unit. An image processing apparatus. The image processing apparatus according to claim 4.
The first correction coefficient calculation unit, except that the average luminance detected by the average luminance detection unit is lower than a predetermined level, and the peak luminance level detected by the peak luminance detection unit is equal to or higher than a predetermined level, An image processing apparatus characterized in that, for an image signal having a relatively low average luminance, the value of the first correction coefficient that is less than the threshold value is relatively low. The image processing apparatus according to claim 5.
An image processing apparatus, wherein when the average luminance is lower than a predetermined level and the level of the peak luminance is equal to or higher than a predetermined level, the value of the second correction coefficient less than the threshold is relatively increased. The image processing apparatus according to claim 6.
The first correction coefficient calculation unit divides the average luminance level detected by the average luminance detection unit into a plurality of stages, and for each of the divided average luminance levels, the peak luminance is reduced at the lowest average luminance level. It is divided into a high level above a predetermined level and a low level below a predetermined level.
Except for the case where the average luminance level is the lowest and the peak luminance level is equal to or higher than a predetermined level, the lower the average luminance, the lower the value of the first correction coefficient that is less than the threshold value, and the average luminance. When the level of the first luminance is the lowest and the level of the peak luminance is equal to or higher than a predetermined level, the value of the first correction coefficient that is less than the threshold is set to be as high as that of the highest average luminance. apparatus. For each pixel of the input image signal, the brightness of each pixel in the first predetermined area around the pixel is averaged to calculate a first illumination light intensity that serves as an index indicating the intensity of reflected light from the subject, A first illumination light distribution calculating unit for calculating a distribution of the first illumination light intensity of the input image signal;
For each pixel of the input image signal, a constant level is maintained in a region where the gradation value of the first illumination light intensity is less than a predetermined threshold value, and a region where the gradation value of the first illumination light intensity is greater than or equal to the threshold value Then, a first correction coefficient calculation unit that calculates a first correction coefficient that decreases as the gradation value of the first illumination light increases;
Second illumination light that calculates the second illumination light intensity distribution by averaging the brightness of each pixel in the second predetermined area smaller than the first predetermined area from the image signal corrected by the first correction coefficient. A distribution calculator;
A second correction coefficient calculation unit that calculates a second correction coefficient that decreases as the gradation value of the second illumination light intensity decreases;
A gradation conversion processing unit that performs gradation conversion by applying the first correction coefficient and the second correction coefficient to an input image signal,
A luminance histogram detection unit that detects a luminance histogram obtained by integrating the number of pixels for each luminance value for each frame of the input image signal;
The image processing apparatus according to claim 1, wherein the first correction coefficient calculation unit changes a value of the first correction coefficient that is less than the threshold value in accordance with the luminance histogram detected by the luminance histogram detection unit. The image processing apparatus according to claim 8.
The first correction coefficient calculation unit is based on a predetermined low gradation region and a predetermined high gradation region of the luminance histogram detected by the luminance histogram detection unit,
In the predetermined range where the predetermined low gradation region is the smallest, the value of the first correction coefficient less than the threshold value is relatively maximized,
When the predetermined low gradation region is in the predetermined range of the intermediate level, and there is a predetermined high gradation region, the value of the first correction coefficient less than the threshold is set to the intermediate level, and the predetermined high gradation region When there is no, the value of the first correction coefficient less than the threshold is relatively lowest,
In the predetermined range where the predetermined low gradation region is the highest, when there is a predetermined high gradation region, the value of the first correction coefficient less than the threshold is relatively maximized and there is no predetermined high gradation region Sometimes, the image processing apparatus is characterized in that the value of the first correction coefficient less than the threshold value is relatively lowest. For each pixel of the input image signal, the brightness of each pixel in the first predetermined area around the pixel is averaged to calculate a first illumination light intensity that serves as an index indicating the intensity of reflected light from the subject, A first illumination light distribution calculating unit for calculating a distribution of the first illumination light intensity of the input image signal;
For each pixel of the input image signal, a constant level is maintained in a region where the gradation value of the first illumination light intensity is less than a predetermined threshold value, and a region where the gradation value of the first illumination light intensity is greater than or equal to the threshold value Then, a first correction coefficient calculation unit that calculates a first correction coefficient that decreases as the gradation value of the first illumination light increases;
Second illumination light that calculates the second illumination light intensity distribution by averaging the brightness of each pixel in the second predetermined area smaller than the first predetermined area from the image signal corrected by the first correction coefficient. A distribution calculator;
A second correction coefficient calculation unit that calculates a second correction coefficient that decreases as the gradation value of the second illumination light intensity decreases;
A gradation conversion processing unit that performs gradation conversion by applying the first correction coefficient and the second correction coefficient to an input image signal;
A noise reduction processing unit that performs noise removal processing on an image signal input to the gradation conversion processing unit,
For each frame of the input image signal, a luminance difference between adjacent pixels, which is an absolute value of a luminance difference between adjacent pixels, is calculated for each pixel constituting the frame, and the number of pixels is integrated for each luminance difference between adjacent pixels. An edge histogram detection unit that generates an edge histogram and detects a change amount between frames of the generated edge histogram;
The first correction coefficient calculation unit changes the value of the first correction coefficient less than the threshold according to the amount of change in the edge histogram detected by the edge histogram detection unit,
The image processing apparatus, wherein the noise reduction processing unit changes a noise removal intensity according to an edge histogram detected by the edge histogram detection unit. The image processing apparatus according to claim 10.
The first correction coefficient calculation unit relatively increases the value of the first correction coefficient less than the threshold in an image signal in which the change amount of the edge histogram detected by the edge histogram detection unit is relatively small,
The image processing apparatus according to claim 1, wherein the noise reduction processing unit relatively weakens the intensity of the noise removal processing for an image signal in which a change amount of the edge histogram detected by the edge histogram detection unit is relatively small. The image processing apparatus according to claim 11.
Dividing the amount of change of the edge histogram detected by the edge histogram detector into a plurality of stages,
The first correction coefficient calculation unit increases the value of the first correction coefficient less than the threshold as the change amount of the edge histogram detected by the edge histogram detection unit is smaller,
The image processing apparatus according to claim 1, wherein the noise reduction processing unit weakens the noise removal processing as the level of change in the edge histogram detected by the edge histogram detection unit is smaller. For each pixel of the input image signal, the brightness of each pixel in the first predetermined area around the pixel is averaged to calculate a first illumination light intensity that serves as an index indicating the intensity of reflected light from the subject, A first illumination light distribution calculating unit for calculating a distribution of the first illumination light intensity of the input image signal;
For each pixel of the input image signal, a constant level is maintained in a region where the gradation value of the first illumination light intensity is less than a predetermined threshold value, and a region where the gradation value of the first illumination light intensity is greater than or equal to the threshold value Then, a first correction coefficient calculation unit that calculates a first correction coefficient that decreases as the gradation value of the first illumination light increases;
Second illumination light that calculates the second illumination light intensity distribution by averaging the brightness of each pixel in the second predetermined area smaller than the first predetermined area from the image signal corrected by the first correction coefficient. A distribution calculator;
A second correction coefficient calculation unit that calculates a second correction coefficient that decreases as the gradation value of the second illumination light intensity decreases;
A gradation conversion processing unit that performs gradation conversion by applying the first correction coefficient and the second correction coefficient to an input image signal,
An average luminance detector for detecting the average luminance of each frame of the input video signal;
A motion detection determination unit that detects the motion of the input image signal and determines the degree of motion;
A noise reduction processing unit that performs noise removal processing of the input image signal,
The noise reduction processing unit has a two-dimensional noise reduction function and a three-dimensional noise reduction function,
The first correction coefficient calculation unit changes the value of the first correction coefficient less than the threshold according to the average luminance detected by the average luminance detection unit and the degree of movement determined by the motion detection determination unit,
The image processing apparatus, wherein the noise reduction processing unit changes the intensity of noise removal processing according to the average luminance detected by the average luminance detection unit and the degree of motion determined by the motion detection determination unit. The image processing apparatus according to claim 13.
The first correction coefficient calculation unit relatively lowers the value of the first correction coefficient less than the threshold in an image signal having a relatively large degree of motion determined by the motion detection determination unit, and
The noise reduction processing unit increases the noise reduction intensity by the two-dimensional noise reduction function and decreases the noise reduction intensity by the three-dimensional noise reduction function,
When the degree of motion is the same level, an image signal having a relatively low average brightness detected by the average brightness detection unit has a relatively low value of the first correction coefficient less than the threshold value. An image processing apparatus. The image processing apparatus according to claim 14.
Classifying the average luminance level detected by the average luminance into a plurality of stages, and classifying the level of the degree of motion determined by the motion detection determination unit into a plurality of stages;
The first correction coefficient calculation unit calculates a value of the first correction coefficient that is less than the threshold for an image signal having a relatively large degree of motion determined by the motion detection determination unit at the same average luminance stage. The noise reduction processing unit increases the noise reduction strength by the two-dimensional noise reduction function and weakens the noise reduction strength by the three-dimensional noise reduction function.
At the stage of the same degree of movement, the first correction coefficient calculation unit lowers the value of the first correction coefficient that is less than the threshold value for an image signal whose average luminance detected by the average luminance detection unit is relatively low. An image processing apparatus.   An image display apparatus comprising: the image processing apparatus according to claim 1; and an image display unit that displays an image based on an image signal subjected to gradation conversion by the image processing apparatus.

Images