JP2013098611A - Gradation conversion device, gradation conversion method and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To suppress reduction of an SN ratio by reducing color noise generated by gradation conversion.SOLUTION: Each of random noise addition parts 33-1 to 33-3 adds random noise to a pixel value of an RGB component signal in each pixel constituting an image. One-dimensional ΔΣ modulation parts 34-1 to 34-3 respectively apply one-dimensional ΔΣ modulation to pixel values outputted from the random noise addition parts 33-1 to 33-3. When the pixel values are white or black, a control part 31 does not operate the one-dimensional ΔΣ modulation parts 34-1 to 34-3, and in the other case, executes control so as to operate the one-dimensional ΔΣ modulation parts 34-1 to 34-3. The technique can be applied to a head mount display.

Description

  The present technology relates to a gradation conversion device, a gradation conversion method, and a program, and in particular, a gradation conversion device and a gradation capable of suppressing an S / N reduction caused by gradation conversion and realizing appropriate color development. The present invention relates to a conversion method and a program.

  For example, to display an image having an N-bit pixel value (hereinafter also referred to as an N-bit image) on a display device that displays an image having an M-bit pixel value smaller than N bits, It is necessary to perform conversion to an M-bit image, that is, gradation conversion for converting the gradation of the image.

  As a method of gradation conversion from an N-bit image to an M-bit image (gradation conversion method), for example, lower N−M bits of an N-bit pixel value are discarded to obtain an M-bit pixel value. There is a way.

For example, if the lower 4 bits of an 8-bit image are discarded and quantized to 4 bits, 256 (= 2 ⁸ ) gradations can be expressed with ⁸ bits, but 16 (= 2 with 4 bits). ⁴ ) Since only the gradation can be expressed, the gradation conversion in which the lower 4 bits of the 8-bit image are discarded causes banding in which the gradation change looks like a band.

  A gradation conversion method for preventing such banding from occurring and for representing the gradation of an image before gradation conversion in a pseudo manner in an image after gradation conversion, that is, for example, as described above, As a method of visually expressing 256 gradations in 16 gradations when a person views an image in a 16 gradation image obtained by gradation conversion of a tone image, a random dither method, There are systematic dither method and error diffusion method.

  Among them, in the error diffusion method, after adding random noise, the quantization error −Q (x, y) is fed back to the input side through a two-dimensional filter, thereby allowing a two-dimensional ΔΣ modulator to be used. There is known a method of configuring and performing two-dimensional ΔΣ modulation using this.

  According to such a two-dimensional ΔΣ modulator, the quantization error −Q (x, y) is high in the spatial frequency in both the horizontal direction (x direction) and the vertical direction (y direction). As a result, an image with better image quality can be obtained as an image after gradation conversion as compared with the random dither method or the systematic dither method (see Patent Document 1).

  However, in a method using a two-dimensional ΔΣ modulator and using this, a line memory is required, so that the apparatus scale is likely to increase, and the apparatus cost also increases.

  In view of this, there is a technique for providing a one-dimensional ΔΣ modulator instead of a two-dimensional ΔΣ modulator, thereby eliminating the need for a line memory, reducing the device scale, and reducing the device cost. It has been proposed (see Patent Document 2).

Japanese Patent No. 3959698 JP 2009-207113 A

  However, in the gradation conversion as described in Patent Document 2, since the same processing is performed regardless of the pixel value, for example, from the RGB (red, green, blue) signal component of the pixel value, black or white Although it is obvious that error diffusion is performed.

  For this reason, there is a possibility that the complete black or complete white color that is actually desired to be expressed cannot be expressed by gradation conversion.

  Also, if a specific pixel is not a black or white color pixel but a color pixel, the same random noise is added to each RGB component corresponding to that pixel and the tone conversion is performed. There was a possibility that the S / N generated by the conversion could not be sufficiently improved.

  The present technology has been made in view of such a situation, and in particular, by adjusting the type of random noise to be added and dithering according to the RGB component of the pixel value, and appropriately applying ΔΣ modulation. The S / N reduction caused by the gradation conversion is suppressed, and more appropriate color development can be realized.

  The gradation conversion device according to one aspect of the present technology is the gradation conversion device that converts the gradation of an image, the random noise adding unit that adds random noise to the pixel value of each pixel constituting the image, and the random conversion device. A one-dimensional ΔΣ modulation unit that performs one-dimensional ΔΣ modulation on the pixel value output from the noise addition unit, and controls operations of the random noise addition unit and the one-dimensional ΔΣ modulation unit based on the pixel value. And a control unit.

  The random noise addition unit includes a random noise addition switch that switches between an on state in which random noise is added to the pixel value and output, and an off state in which the pixel value is output without adding random noise. In the ΔΣ modulation unit, the pixel value output from the random noise addition unit is subjected to ΔΣ modulation and output, and the pixel value output from the random noise addition unit is ΔΣ modulation. A ΔΣ modulation switch that switches between an OFF state and an output without being applied can be included, and when the pixel value is white or black, the control unit includes the random noise addition switch, and the ΔΣ Any of the modulation switches can be controlled to be turned off.

  The random noise addition unit includes a random noise addition switch that switches between an on state in which random noise is added to the pixel value and output, and an off state in which the pixel value is output without adding random noise. In the ΔΣ modulation unit, the pixel value output from the random noise addition unit is subjected to ΔΣ modulation and output, and the pixel value output from the random noise addition unit is ΔΣ modulation. A ΔΣ modulation switch that switches between an off state and an output without being applied, and the control unit includes the random noise addition switch when the pixel value is neither white nor black, and Any of the ΔΣ modulation switches can be controlled to be turned on.

  The pixel value can include an RGB (red, green, and blue) signal component in each pixel, and further includes a signal conversion unit that converts the RGB signal component into a YCbCr (luminance color difference) signal. When the pixel is considered to be monochrome because the sum of the CbCr (color difference) signals among the YCbCr (luminance color difference) signals is smaller than a first predetermined value, the RGB in each pixel The random noise adding unit corresponding to each signal component can add the same random noise to each of the RGB signal components in each pixel.

  The pixel value can include an RGB (red, green, and blue) signal component in each pixel, and further includes a signal conversion unit that converts the RGB signal component into a YCbCr (luminance color difference) signal. If the sum of the CbCr (color difference) signals among the YCbCr (luminance color difference) signals is smaller than this, the pixel is considered to be monochrome and is larger than a first predetermined value, and the first When the pixel is considered to be similar to monochrome by being smaller than a second predetermined value that is larger than the predetermined value, the random noise adding unit corresponding to each RGB signal component in each pixel has the pixel It is possible to add similar random noise to the RGB signal components in FIG.

  The pixel value can include an RGB (red, green, and blue) signal component in each pixel, and further includes a signal conversion unit that converts the RGB signal component into a YCbCr (luminance color difference) signal. If the sum of the CbCr (luminance difference) signals among the YCbCr (luminance color difference) signals is smaller than the first predetermined value, the pixel is regarded as monochrome. Random noise addition corresponding to each RGB signal component at each pixel if the pixel is considered to be color by being less than a second predetermined value that is considered to be monochrome-like The unit may add different random noises to the RGB signal components in the pixels.

  A gradation conversion method according to an aspect of the present technology is a gradation conversion method of a gradation conversion apparatus that converts the gradation of an image, and includes random noise that adds random noise to the pixel value of each pixel constituting the image. A noise addition process is performed, and a one-dimensional ΔΣ modulation process is performed to apply a one-dimensional ΔΣ modulation to the pixel value output from the random noise addition unit. Based on the pixel value, the random noise addition unit, and the 1 Including a step of performing a control process for controlling the operation of the dimension ΔΣ modulator.

  A program according to one aspect of the present technology includes: a random noise adding unit that adds a random noise to a pixel value for each pixel constituting the image; A one-dimensional ΔΣ modulation unit that performs one-dimensional ΔΣ modulation on the pixel value output from the random noise addition unit, and controls the operations of the random noise addition unit and the one-dimensional ΔΣ modulation unit based on the pixel value. Function as a control unit.

  In one aspect of the present technology, random noise is added to the pixel value of each pixel constituting the image, and one-dimensional ΔΣ modulation is performed on the pixel value to which the random noise is added, and the pixel value is Based on this, the operation of adding the random noise and the operation of applying the one-dimensional ΔΣ modulation are controlled.

  The gradation conversion apparatus of the present technology may be an independent apparatus or a block that performs gradation conversion.

  According to one aspect of the present technology, it is possible to suppress an S / N reduction caused by gradation conversion and to realize appropriate color development.

It is a block diagram which shows the structural example of one Embodiment of the television receiver to which this technique is applied. It is a figure explaining the process of the gradation conversion part of FIG. It is a block diagram explaining the structural example of the gradation conversion part of FIG. It is a block diagram explaining the structural example of the random noise output part of FIG. 1, and a control part. It is a figure explaining operation | movement of the weight control part of FIG. FIG. 2 is a block diagram illustrating a configuration example of a random noise addition unit and a one-dimensional ΔΣ modulation unit in FIG. 1. It is a block diagram explaining the structural example of the one-dimensional filter of FIG. It is a flowchart explaining the gradation conversion process by the gradation conversion part of FIG. It is a figure explaining a raster scan. It is a figure explaining the image before the gradation conversion process by the gradation conversion part of FIG. It is a figure explaining the image after the gradation conversion process by the gradation conversion part of FIG. And FIG. 11 is a diagram illustrating a configuration example of a general-purpose personal computer.

[Example configuration of a television receiver]
FIG. 1 shows a configuration example of an embodiment of a television receiver to which the present technology is applied. In FIG. 1, a configuration example of a television receiver is shown. However, the present technology can be applied to a device other than a television receiver as long as it can be realized by the same configuration, for example, an HMD. (Head Mount Display) can also be applied.

  In FIG. 1, the television receiver 1 includes a tuner 11, a demultiplexer 12, a decoder 13, a noise reduction unit 14, a gradation conversion unit 15, a display control unit 16, and a display unit 17.

  For example, the tuner 11 receives a broadcast signal of digital broadcasting, demodulates a transport stream from the broadcast signal, and supplies the demodulated signal to the demultiplexer 12.

  The demultiplexer 12 separates necessary TS (Transport Stream) packets from the transport stream from the tuner 11 and supplies them to the decoder 13.

  The decoder 13 decodes MPEG (Moving Picture Expert Group) encoded data included in the TS packet from the demultiplexer 12 to obtain, for example, an 8-bit image (data), and sends it to the noise reduction unit 14. Supply.

  The noise reduction unit 14 performs noise reduction processing on the 8-bit image from the decoder 13 and supplies, for example, a 14-bit image obtained as a result to the gradation conversion unit 15.

  That is, according to the noise reduction processing by the noise reduction unit 14, an 8-bit image is expanded to a 14-bit image.

  The gradation converting unit 15 performs gradation conversion for converting the 14-bit image supplied from the noise reduction unit 14 into an image having the number of bits that can be displayed by the display unit 17.

  That is, the gradation conversion unit 15 acquires the number of bits of an image that can be displayed by the display unit 17 and other necessary information from the display control unit 16.

  When the number of bits of the image that can be displayed on the display unit 17 is, for example, 8 bits for each of the RGB signal components, the gradation conversion unit 15 converts the RGB from the noise reduction unit 14 as shown in FIG. Each 14-bit (total 42 bits) image of the signal component is converted to 8 bits.

  Then, the gradation conversion unit 15 supplies an image obtained by gradation-converting each 14-bit image of the RGB signal components from the noise reduction unit 14 to 8 bits, to the display control unit 16.

  The display control unit 16 controls the display unit 17 to display the image from the gradation conversion unit 15 on the display unit 17.

  The display unit 17 includes, for example, an LCD (Liquid Crystal Display), an organic EL (organic Electro Luminescence), and the like, and displays an image under the control of the display control unit 16.

[Configuration example of gradation conversion unit]
Next, a configuration example of the gradation conversion unit 15 in FIG. 1 will be described with reference to FIG.

  3 includes a control unit 31, a random noise output unit 32, random noise addition units 33-1 to 33-3, and one-dimensional ΔΣ modulation units 34-1 to 34-3. .

  The control unit 31 controls the operations of the random noise addition units 33-1 to 33-3 and the one-dimensional ΔΣ modulation units 34-1 to 34-3 according to the RGB signal components in each pixel of the input image. .

  In addition, the control unit 31 performs predetermined processing on the random noise NR, NG, NB corresponding to each image of the RGB signal components output by the random noise output unit 32, and adds random noise as the random noise NRW, NGW, NBW. Supply to the units 33-1 to 33-3.

  The random noise adding units 33-1 to 33-3 are controlled by the control unit 31 and add dither noise NRW, NGW, and NBW to the pixel values of the RGB signal components that are the input images as necessary. Are output to the one-dimensional ΔΣ modulators 34-1 to 34-3.

  The one-dimensional ΔΣ modulation units 34-1 to 34-3 are controlled by the control unit 31, and for example, a 14-bit pixel value F (x, y) of the supplied RGB signal component is converted into ΔΣ as necessary. Modulation is performed, for example, conversion into an 8-bit pixel value Out (x, y).

  The one-dimensional ΔΣ modulators 34-1 to 34-3 each output a pixel value Out (x, y) modulated to 8 bits.

  If there is no particular need to distinguish the random noise adding units 33-1 to 33-3 and the one-dimensional ΔΣ modulation units 34-1 to 34-3, the random noise adding unit 33 and the one-dimensional ΔΣ modulation are simply used. The same shall apply to the other components.

[Configuration example of control unit and random noise output unit]
Next, configuration examples of the control unit 31 and the random noise output unit 32 will be described with reference to FIG.

  The random noise output unit 32 includes random noise generation units 51-1 to 51-3 and HPF (High Pass Filter) 52-1 to 52-3.

  Each of the random noise generators 51-1 to 51-3 generates random noise and outputs it to the HPFs 52-1 to 52-3, respectively.

  Here, the random noise generators 51-1 to 51-3 all have the same configuration. For example, when a function for generating random noise is used, each of the random noise generators 51-1 to 51-3 is randomized using different initial values. Generate noise.

  Therefore, all the random noises generated by the random noise generators 51-1 to 51-3 are different random noises.

  The random noise generators 51-1 to 51-3 may be configured to generate different random noises at the time of generation. Therefore, as long as different random noises can be generated, the random noise generation methods may be different from each other. A method other than a method using a function may be used.

  The HPFs 52-1 to 52-3 have the same configuration, and high-frequency random noises NR, NG, and NB composed of high-frequency components among the random noises generated by the random noise generators 51-1 to 51-3. Are extracted and supplied to the control unit 31.

  The control unit 31 in FIG. 4 includes a black and white detection unit 71, a YCbCr conversion unit 72, a weight control unit 73, and a random noise control unit 74.

  The black-and-white detection unit 71 determines whether the pixel value IN (x, y) to be processed of each input image of the RGB signal components is white or black, and uses the determination result WB as a random noise addition unit 33-1 to 33-3 and the one-dimensional ΔΣ modulators 34-1 to 34-3.

  The YCbCr conversion unit 72 uses Y (luminance), Cb (blue difference), and Cr (red difference) using the pixel values of the RGB signal components (hereinafter, the color differences are collectively referred to as color differences Cb and Cr). The color differences Cb and Cr are supplied to the weight control unit 73.

  The weight control unit 73 sets the weight w based on the sum (Cb + Cr) of the color differences Cb and Cr supplied from the YCbCr conversion unit 72 and supplies the weight w to the random noise control unit 74.

  More specifically, the weight control unit 73 sets the weight w based on a function set as shown in FIG. 5, for example, corresponding to the sum value of the color differences Cb and Cr.

  In FIG. 5, for example, when the sum of the color differences Cb and Cr is 0 to A, the weight control unit 73 sets the weight w to 0, and when the sum of the color differences Cb and Cr is larger than B, the weight w is set. Set to 1.

  Then, when the sum of the color differences Cb and Cr is A to B, the weight control unit 73 sets the weight w to ((Cb + Cr) −A) / (B−A).

  The values A and B are set to be approximately A≈B, and A is a value close to 0.

  That is, this indicates a condition when the input image is a monochrome image or an image close to monochrome, and the sum of the color differences Cb and Cr is small. In such a case, the weight w is 0 or 0 to 1 Is set to the value of

  Conversely, when the input image is a color image and the sum of the color differences Cb and Cr is large, the weight w is set to 1.

  The random noise control unit 74 includes multipliers 91-1 to 91-3 and adders 92-1 and 92-2.

  The multiplier 91-1 multiplies the R (red) component high-frequency random noise NR by the weight (1-w) and supplies the result to the adders 92-1 and 92-2.

  The multipliers 91-2 and 91-3 multiply the high frequency random noises NG and NB of the G (green) and B (blue) components by the weight w, respectively, and supply them to the adders 92-2 and 92-3, respectively. .

  The adder 92-1 includes a high frequency random noise NR of the R component multiplied by the weight (1-w) from the multiplier 91-1, and a G component multiplied by the weight w from the multiplier 91-2. Are added together with the high frequency random noise NG, and the addition result is output as a G component random noise NGW.

  The adder 92-2 has the R component high frequency random noise NR multiplied by the weight (1-w) from the multiplier 91-1, and the B component multiplied by the weight w from the multiplier 91-3. The high frequency random noise NB is added, and the addition result is output as B component random noise NBW.

  That is, the R component random noise NRW is the R component high frequency random noise NR supplied from the random noise output unit 32 as it is.

  The G component random noise NGW is generated by mixing the R component high frequency random noise NR supplied from the random noise output unit 32 and the G component high frequency random noise NG with a weight w.

  Further, the B component random noise NBW is generated by mixing the R component high frequency random noise NR supplied from the random noise output unit 32 and the B component high frequency random noise NB with a weight w.

  That is, when the pixel value of the input image is monochrome, the weight w is set to 0.

  For this reason, the random noises NRW, NGW, and NBW of the RGB signal components are all the R component high frequency random noise NR, and are the same random noise.

  Further, when the pixel value of the input image is close to a monochrome image, the weight w is set to any value from 0 to 1.

  For this reason, the RGB component random noises NRW, NGW, and NBW are different from each other, but are close to the R component high frequency random noise NR, and are therefore similar to each other.

  Further, when the input image is a color image, the weight w is set to 1.

  For this reason, the RGB component random noises NRW, NGW, and NBW are the high-frequency random noises NR, NG, and NB that are independent of the RGB signal components, respectively, and are different random noises.

[Configuration Example of Random Noise Adder and One-Dimensional ΔΣ Modulator]
Next, configuration examples of the random noise addition unit 33 and the one-dimensional ΔΣ modulation unit 34 will be described with reference to FIG.

  FIG. 6 shows the input and output of signals in the R component random noise adding unit 33-1 and the one-dimensional ΔΣ modulation unit 34-1; Are omitted, and the RGB components are not particularly distinguished.

  The random noise adding unit 33 includes an adder 111 and a switch 112. The adder 111 adds the pixel value IN (x, y) of the input image of the RGB signal components and the random noises NRW, NGW, and NBW of the RGB components respectively supplied from the control unit 31.

  Then, the adder 111 supplies each addition result of the RGB signal components to the switch 112.

  Based on the determination result WB from the monochrome detection unit 71 of the control unit 31, the switch 112 adds an image obtained by adding the random noises NRW, NGW, NBW of the RGB components supplied from the adder 111, and an adder The input image supplied by bypassing 111 is switched and supplied to the one-dimensional ΔΣ modulator 34 as a pixel value F (x, y).

  Therefore, when the pixel value IN (x, y) of the input image supplied by bypassing the adder 111 by the switch 112 is output, the random noise NRW, NGW, NBW is output without being added. .

  That is, the pixel value IN (x, y) of the input image is output as it is as the pixel value F (x, y).

  Hereinafter, the state in which the switch 112 outputs the pixel value F (x, y) to which the random noises NRW, NGW, and NBW are added is referred to as an on state.

  Further, the switch 112 bypasses the adder 111 and outputs a pixel value IN (x, y) to which the random noise NRW, NGW, NBW is not added as it is as the pixel value F (x, y) as an off state. Shall be called.

  The one-dimensional ΔΣ modulation unit 34 includes an adder 131, a switch 132, a quantization unit 133, a subtracter 134, and a one-dimensional filter 135.

  The adder 131 calculates the pixel value F (x, y) of the image supplied from the random noise adding unit 33 and the value obtained by using the quantization error for the previous several pixels by the one-dimensional filter 135. The sum is supplied to the switch 132.

  Based on the determination result WB from the monochrome detection unit 71 of the control unit 31, the switch 132 is a result of filtering using the quantization error for the previous several pixels by the one-dimensional filter 135 supplied from the adder 131. Are switched to the input image supplied by bypassing the adder 131 and supplied to the quantization unit 133 and the subtracter 134.

  In the following description, the switch 132 quantizes an image obtained by adding the result of filtering using the quantization error for the previous several pixels by the one-dimensional filter 135 supplied from the adder 131 to obtain a pixel value. A state output as Out (x, y) is referred to as an on state.

  Further, the switch 132 turns off the state in which the pixel value F (x, y) of the input image supplied is quantized as it is, bypassing the adder 131, and output as the pixel value Out (x, y). It shall be called a state.

  The quantization unit 133 quantizes and outputs the image supplied from the switch 132 to the number of bits to be supplied to the display control unit 16 in the subsequent stage, and supplies the quantization result to the subtracter 134.

  The subtractor 134 supplies a quantization error −Q (x, y) obtained by subtracting the quantization result from the image supplied from the switch 132 to the one-dimensional filter 135.

  The one-dimensional filter 135 performs a one-dimensional filtering process on the quantization error −Q (x, y) supplied from the subtracter 134 and supplies the result to the adder 131.

  That is, when the switch 132 is turned on, ΔΣ modulation is performed by the subtractor 134, the one-dimensional filter 135, and the adder 131.

  Therefore, when an image in which the adder 131 is bypassed by the switch 132 being turned off is output to the quantization unit 133, the image is quantized in a state where ΔΣ modulation is not performed.

[Configuration example of one-dimensional filter]
Next, a configuration example of the one-dimensional filter 135 in FIG. 6 will be described with reference to FIG.

  In FIG. 7, the one-dimensional filter 135 includes delay units 151-1 to 151-5, multiplication units 152-1 to 152-5, and an adder 153, and a 5-tap FIR (Finite Impulse Response) filter. It is composed.

  That is, the stored value of the preceding delay unit 151- (i-1) is input to the delay unit 151-i (i = 1, 2, 3, 4, 5).

  The delay unit 151-i temporarily stores the input to the delay unit 151-i, delays the input by a time corresponding to one pixel, and outputs it to the subsequent delay unit 151- (i + 1) and the multiplication unit 152-i. To do.

  Note that the quantization error −Q (x, y) from the subtractor 134 (FIG. 6) is supplied to the first delay unit 151-1, and the delay unit 151-1 receives the quantization error −Q ( Delay by storing x, y).

  Further, the delay unit 151-5 at the rearmost stage outputs the delayed input only to the multiplication unit 152-5.

  The multiplication unit 152-i multiplies the output of the delay unit 151-i and the filter coefficient a (i), and supplies the multiplication value obtained as a result to the addition unit 153.

  The adder 153 adds the multiplication values from the multipliers 152-1 to 152-5, and uses the resulting addition value as a result of filtering the quantization error −Q (x, y) as an adder 131. (FIG. 6).

  As described above, the one-dimensional filter 135 requires the delay unit 151-i that stores the quantization error of several pixels (5 pixels in FIG. 7) on one horizontal line. It is not necessary to provide the necessary line memory.

  Therefore, according to the one-dimensional ΔΣ modulator 34 including such a one-dimensional filter 135, the apparatus can be reduced in size and cost compared to the two-dimensional ΔΣ modulator. .

[Tone conversion processing]
Next, tone conversion processing will be described with reference to the flowchart of FIG.

  In step S1, the control unit 31 acquires the pixel value IN (x, y) of the input image.

  That is, when the pixel value IN (x, y) of the target image is supplied from the noise reduction unit 44 (FIG. 1), the control unit 31 acquires this.

  For example, as illustrated in FIG. 9, the noise reduction unit 44 supplies the pixel value IN (x, y) of the pixel (x, y) of the target image to the gradation conversion unit 45 in the raster scan order. .

  Therefore, in the gradation conversion unit 45, the pixel value IN (x, y) of the pixel (x, y) of the target image is subjected to gradation conversion processing in the raster scan order.

  Note that the pixel value IN (x, y) here includes RGB components.

  In step S2, the black and white detection unit 71 determines whether the pixel to be processed is white or black based on whether the pixel value is the minimum value or the maximum value for each of the RGB components. Determine whether.

  Then, the black and white detection unit 71 supplies the determination result WB to the switch 112 (FIG. 6) of the random noise addition unit 33 and the switch 132 (FIG. 6) of the one-dimensional ΔΣ modulation unit 34.

  In step S3, the random noise generating units 51-1 to 51-3 of the random noise output unit 22 generate random noise corresponding to the RGB signal components and supply the random noise to the HPFs 52-1 to 52-3.

  In step S4, the HPFs 52-1 to 52-3 extract only the high frequency components from the random noises supplied from the random noise generating units 51-1 to 51-3, and the high frequency random noises NR, NG, It supplies to the control part 31 as NB.

  In step S 5, the YCbCr conversion unit 72 of the control unit 31 converts the supplied RGB signal component information into luminance Y and color differences Cb, Cr, of which the color differences Cb, Cr are sent to the weight control unit 73. Supply.

  In step S6, the weight control unit 73 determines whether or not the pixel value to be processed is monochrome based on the sum of the color differences Cb and Cr (Cb + Cr).

  More specifically, the weight control unit 73 determines whether the sum (Cb + Cr) of the color differences Cb and Cr is 0 to A in FIG.

  In step S6, for example, when the sum (Cb + Cr) of the color differences Cb and Cr is 0 to A in FIG. 5 and is sufficiently small and the pixel value is considered to be monochrome, the process proceeds to step S7. .

  In step S <b> 7, the weight control unit 73 sets the weight w of the pixel value regarded as monochrome to 0 and supplies it to the random noise control unit 74.

  In step S6, for example, if the sum (Cb + Cr) of the color differences Cb and Cr is not 0 to A in FIG. 5, the process proceeds to step S8.

  In step S8, the weight control unit 73 determines whether or not the pixel to be processed is color based on the sum of the color differences Cb and Cr (Cb + Cr).

  More specifically, the weight control unit 73 determines whether or not the sum of the color differences Cb and Cr (Cb + Cr) is larger than B in FIG.

  In step S8, for example, when the sum (Cb + Cr) of the color differences Cb and Cr is sufficiently larger than B in FIG. 5, the process proceeds to step S9.

  In step S <b> 9, the weight control unit 73 sets the weight w of the pixel value regarded as color to 1 and supplies it to the random noise control unit 74.

  Furthermore, in step S8, for example, when the sum of color differences Cb and Cr (Cb + Cr) is not larger than B in FIG. 5, that is, when the sum of color differences Cb and Cr (Cb + Cr) is A to B in FIG. The process proceeds to step S10.

  In step S10, the weight control unit 73 sets the weight w to ((Cb + Cr) −A) / (B−A) based on the color differences Cb and Cr.

  That is, based on the color differences Cb and Cr, the weight control unit 73 sets the weight w to 0 when the pixel value is monochrome and does not include the color difference, and includes the color difference when the pixel value is color. If w is set to 1 and is in the middle other than that, the weight w is set to ((Cb + Cr) -A) / (BA).

  In step S11, the random noise control unit 74 performs processing on the high frequency random noise NR, NG, NB based on the weight w, and outputs the result as random noise NRW, NGW, NBW.

  That is, the high frequency random noise NR is output as it is as the random noise NRW of the R signal component.

  Further, the high frequency random noise NR (1-w) of the R signal component multiplied by the weight (1-w) by the multiplier 91-1 and the high frequency of the G signal component multiplied by the weight w by the multiplier 91-2. Random noise NGw is added by the adder 92-1.

  As a result, the addition result is output as random noise NGW (= high-frequency random noise NR (1-w) + high-frequency random noise NGw) of the G signal component.

  Further, the high-frequency random noise NR (1-w) of the R signal component multiplied by the weight (1-w) by the multiplier 91-1 and the high-frequency of the B signal component multiplied by the weight w by the multiplier 91-3. The random noise NBw is added by the adder 92-2.

  As a result, the addition result is output as random noise NBW (= high frequency random noise NR (1-w) + high frequency random noise NBw) of the B signal component.

  As a result, when the pixel to be processed is monochrome and the weight w = 0 is set, the random noises NRW, NGW, and NBW of the RGB signal components output from the control unit 31 are all the R component high frequency random noises. Output as the same as NR.

  Further, when the pixel to be processed is close to monochrome and is set with a weight w≈0, the random noises NRW, NGW, and NBW of the RGB signal components output from the control unit 31 are all high-frequency components of the R component. Output as similar to random noise NR.

  Further, when the pixel to be processed is color and set with a weight w = 1, the random noise NRW, NGW, and NBW of the RGB signal components output from the control unit 31 are the high-frequency random noise NR of the RGB components, respectively. , NG, NB, which are output as different random noises.

  Therefore, when the pixel value to be processed is close to monochrome, as described above, the weight w may be close to 0, and thus is limited to ((Cb + Cr) -A) / (BA). Instead, a fixed value close to 0 may be set.

  In step S12, the switch 112 and the switch 132 each determine whether the determination result WB from the black and white detection unit 71 is either a white image or a black image.

  In step S12, when the determination result WB supplied from the black and white detection unit 71 is neither a white pixel nor a black pixel, the process proceeds to step S13.

  In step S13, in any of the random noise adding units 33-1 to 33-3, the switch 112 causes the signal output from the adder 111 to be output from the random noise adding units 33-1 to 33-3. Switch the state to the on state.

  In step S14, in any of the one-dimensional ΔΣ modulation units 34-1 to 34-3, the switch 132 switches the state to the on state so that the signal output from the adder 131 is output to the quantization unit 133.

  In step S15, the random noise adding units 33-1 to 33-3 add the random noise NRW, NGW, and NBW supplied from the control unit 31 to the RGB signal component of IN (x, y) that is the input image. Dithering is performed by addition, and the result is output as a pixel value F (x, y).

  More specifically, the adder 111 adds random noises NRW, NGW, and NBW to each of the RGB signal components of IN (x, y) that is an input image, and supplies the result to the switch 112.

  Then, the pixel value F (x, y) of the RGB signal component obtained by adding the random noises NRW, NGW, and NBW supplied from the adder 111 based on the state set by the process of step S13. ) To the one-dimensional ΔΣ modulator 34.

  In step S <b> 16, the one-dimensional ΔΣ modulation units 34-1 to 34-3 perform one-dimensional conversion on the RGB signal component pixel values F (x, y) supplied from the random noise addition units 33-1 to 33-3, respectively. ΔΣ modulation processing is performed and output as pixel values Out (x, y) of RGB signal components, respectively.

  More specifically, the adder 131 adds the filtering processing result by the one-dimensional filter 135 to the pixel value F (x, y) of the RGB signal component.

  The switch 132 supplies the processing result of the adder 131 to the quantization unit 133 and the subtracter 134 based on the state set by the process of step S14.

  The quantization unit 133 quantizes the number of bits of the pixel value supplied from the switch 132 to the number of bits that can be displayed on the display unit 17 and outputs the result as the pixel value Out (x, y), and also supplies it to the subtracter 134. To do.

  The subtractor 134 subtracts the output from the quantization unit 133 from the input to the quantization unit 133 to obtain a quantization error −Q (x, y) and supplies the quantization error −Q (x, y) to the one-dimensional filter 135.

  Here, the one-dimensional filter 135 performs a filtering process on the quantization error −Q (x, y) and supplies the filtering process result to the adder 131.

  In the one-dimensional ΔΣ modulation unit 34, the quantization error −Q (x, y) is fed back to the input side (adder 131) via the one-dimensional filter 135, whereby one-dimensional ΔΣ modulation is performed. Done.

  Accordingly, in the one-dimensional ΔΣ modulation unit 34, one-dimensional ΔΣ modulation is performed on the pixel value F (x, y) from the random noise adding unit 33, and the pixel value is obtained as a result of the one-dimensional ΔΣ modulation. Out (x, y) is output.

  In the one-dimensional ΔΣ modulation unit 34, the quantization error −Q (x, y) is a quantization error with respect to the pixel value F (x, y), but the pixel value F (x, y) is ΔΣ-modulated. In determining the pixel value Out (x, y) obtained in this way, the quantization error −Q (x, y) for the pixel value F (x, y) is not used, and the pixel value F (x , y), the quantization error for the pixel value before (pixel value processed earlier) is used.

  That is, in the adder 131, for example, as shown in FIG. 7, the pixel values F (x−1, y) and F (x−) of 5 pixels processed immediately before the pixel value F (x, y). 2, y), F (x−3, y), F (x−4, y), and F (x−5, y), respectively. It is added to the pixel value F (x, y).

  In step S <b> 17, the control unit 31 determines whether or not the next pixel is present. If there is, the process returns to step S <b> 1.

  On the other hand, if it is determined in step S17 that the next pixel does not exist, the process ends.

  If it is determined in step S12 that the pixel value is a white pixel or a black pixel based on the determination result WB, the process proceeds to step S18.

  In step S18, in any of the random noise adding units 33-1 to 33-3, the switch 112 outputs a signal output by bypassing the adder 111 from the random noise adding units 33-1 to 33-3. Switch the state to the off state.

  In step S19, in any of the one-dimensional ΔΣ modulation units 34-1 to 34-3, the switch 132 is turned off so that the signal output by bypassing the adder 131 is output to the quantization unit 133. Switch to.

  In step S20, the random noise adding units 33-1 to 33-3 directly use the RGB signal component of IN (x, y), which is the input image, as the pixel value F (x, y) without dithering. Output.

  More specifically, based on the state set by the process of step S18, the switch 112 bypasses the adder 111 and uses the RGB signal component as it is as the pixel value F (x, y) as a one-dimensional one. Each is output to the ΔΣ modulator 34.

  In step S21, the one-dimensional ΔΣ modulators 34-1 to 34-3 are one-dimensionally applied to the pixel values F (x, y) of the RGB signal components supplied from the random noise adders 33-1 to 33-3, respectively. Without being subjected to ΔΣ modulation processing, the display unit 17 is quantized to an appropriate number of bits and output as a pixel value Out (x, y).

  In more detail, based on the state set by the process of step S19, the switch 132 bypasses the adder 131 and inputs the pixel value F (x, y) as it is to the quantization unit 133 and the subtractor. 134.

  The quantization unit 133 quantizes the number of bits of the pixel value supplied from the switch 132 to the number of bits that can be displayed on the display unit 17 and outputs the result as the pixel value Out (x, y).

  The quantization unit 133 also outputs the quantized result to the subtracter 134, but here, the processing result of the one-dimensional filter 135 is not used, and can be substantially ignored. Then, the process proceeds to step S17.

  That is, for example, when the above-described gradation conversion processing is performed so that the 8-bit image shown in the upper part of FIG. 10 becomes a 4-bit image, the series of processing in steps S1 to S17, for example, FIG. (Image after gradation conversion processing) as shown in the upper part of FIG.

  The lower part of FIG. 11 shows the pixel values on the horizontal line with the 4-bit image after gradation conversion corresponding to the pixel values on the horizontal line with the 8-bit image shown in the lower part of FIG. ing.

  As shown in FIG. 10, 256 gradations can be expressed with 8 bits, whereas only 16 gradations can be expressed with 4 bits as shown in FIG. 11.

  However, in the 4-bit image after the gradation conversion processing by the gradation converting unit 15, the pixel value has a certain quantization value q, and the quantization value q + 1 that is 1 larger than the quantization value q. (Or a dense / dense area with a pixel that is a quantized value q−1 that is smaller by 1), that is, an area where the ratio of pixels with a quantized value q is large, or a quantized value q There are areas with many +1 pixels (areas where the ratio of pixels with the quantized value q + 1 is low, or areas where the pixel value is the quantized value q is small), and the pixel values of the coarse and dense areas However, it seems to change smoothly due to the human visual integration effect.

  As a result, whether the 4-bit image after gradation conversion by the gradation conversion unit 15 is an 8-bit target image before gradation conversion, although only 16 gradations can be expressed with 4 bits. As described above, 256 gradations can be expressed in a pseudo manner.

  Further, in the processing of steps S1 to S17, the random noise of the RGB signal components is switched by setting the weight w according to the sum (Cb + Cr) of the color differences Cb and Cr of the pixel values of the input image. Generated.

  For this reason, when the pixel value of the input image is monochrome, after adding the same random noise to each pixel value of the RGB signal component (after dithering with the same random noise), ΔΣ modulation is performed. As a result, color noise can be prevented.

  In addition, when the pixel value of the input image is close to monochrome, after adding random noise similar to each pixel value of the RGB signal component (after dithering with similar random noise), ΔΣ modulation is performed. As a result, the occurrence of color noise can be suppressed.

  In addition, if the pixel value of the input image is color, after adding different random noise to each pixel value of the RGB signal component (after dithering with different random noise), ΔΣ modulation is applied. As a result, the gradation performance can be improved, and the S / N reduction caused by the gradation conversion can be suppressed.

  As an example of conversion of the gradation conversion unit 15 described above, an example in which a 14-bit input image is converted to 8 bits has been described as shown in FIG. 2, but the number of bits of the input image and the bits after conversion are described. The same applies to other numbers of bits.

  Therefore, in the example described with reference to FIGS. 10 and 11, the case where an 8-bit input image is converted into 4 bits has been described, but the processing has been described with reference to the flowchart of FIG. 9. This is realized by a process similar to the gradation conversion process.

  On the other hand, according to the processing of steps S1 to S12, S18 to S21, and S17, when the pixel value of the input image is a white pixel and a black pixel, random noise is not added to any of the RGB signal components (dither is reduced). Not applied) and one-dimensional ΔΣ modulation is not applied.

  As a result, when the pixel value of the input image is a white pixel and a black pixel, error diffusion is not performed, so that the black pixel and the white pixel are appropriately gradation-converted as the minimum pixel value and the maximum pixel value as they are. Therefore, appropriate black pixels and white pixels can be obtained.

  In addition, in the random noise control unit 74 in FIG. 4, the weight w is used on the basis of the R component high frequency random noise NR and is mixed with the G and B component high frequency random noises NG and NB. , An example is described in which each is switched to the same, similar, or different state.

  However, the configuration of the random noise control unit 74 may be other configurations as long as each can be switched to the same, similar, or different state, for example, a high-frequency random of G or B component You may make it the structure on the basis of noise.

  As described above, according to the present technology, the type of random noise to be added is adjusted according to the RGB component of the pixel value, and dithering is performed, and ΔΣ modulation is appropriately performed, so that the S / N generated by the gradation conversion is obtained. It is possible to suppress the reduction of N and realize more appropriate color development.

  The series of processes described above can be executed by hardware or can be executed by software. When a series of processing is executed by software, a program constituting the software is installed in the computer. Here, the computer includes, for example, a general-purpose personal computer capable of executing various functions by installing various programs by installing a computer incorporated in dedicated hardware.

  FIG. 12 is a block diagram illustrating a hardware configuration example of a computer that executes the above-described series of processing by a program.

  In a computer, a central processing unit (CPU) 1001, a read only memory (ROM) 1002, and a random access memory (RAM) 1003 are connected to each other by a bus 1004.

  An input / output interface 1005 is further connected to the bus 1004. An input unit 1006, an output unit 1007, a storage unit 1008, a communication unit 1009, and a drive 1010 are connected to the input / output interface 1005.

  The input unit 1006 includes a keyboard, a mouse, a microphone, and the like. The output unit 1007 includes a display, a speaker, and the like. The storage unit 1008 includes a hard disk, a nonvolatile memory, and the like. The communication unit 1009 includes a network interface. The drive 1010 drives a removable medium 1011 such as a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory.

  In the computer configured as described above, the CPU 1001 loads the program stored in the storage unit 1008 to the RAM 1003 via the input / output interface 1005 and the bus 1004 and executes the program, for example. Is performed.

  The program executed by the computer (CPU 1001) can be provided by being recorded on the removable medium 1011 as a package medium, for example. The program can be provided via a wired or wireless transmission medium such as a local area network, the Internet, or digital satellite broadcasting.

  In the computer, the program can be installed in the storage unit 1008 via the input / output interface 1005 by attaching the removable medium 1011 to the drive 1010. Further, the program can be received by the communication unit 1009 via a wired or wireless transmission medium and installed in the storage unit 1008. In addition, the program can be installed in advance in the ROM 1002 or the storage unit 1008.

  The program executed by the computer may be a program that is processed in time series in the order described in this specification, or in parallel or at a necessary timing such as when a call is made. It may be a program for processing.

  In this specification, the system means a set of a plurality of components (devices, modules (parts), etc.), and it does not matter whether all the components are in the same housing. Accordingly, a plurality of devices housed in separate housings and connected via a network and a single device housing a plurality of modules in one housing are all systems. .

  The embodiments of the present technology are not limited to the above-described embodiments, and various modifications can be made without departing from the gist of the present technology.

  For example, the present technology can take a configuration of cloud computing in which one function is shared by a plurality of devices via a network and is jointly processed.

  In addition, each step described in the above flowchart can be executed by being shared by a plurality of apparatuses in addition to being executed by one apparatus.

  Further, when a plurality of processes are included in one step, the plurality of processes included in the one step can be executed by being shared by a plurality of apparatuses in addition to being executed by one apparatus.

In addition, this technique can also take the following structures.
(1) In a gradation conversion device for converting the gradation of an image,
A random noise adding unit that adds random noise to the pixel value of each pixel constituting the image;
A one-dimensional ΔΣ modulation unit that performs one-dimensional ΔΣ modulation on the pixel value output from the random noise addition unit;
A gradation conversion device including: a random noise adding unit and a control unit that controls operations of the one-dimensional ΔΣ modulation unit based on the pixel value.
(2) The random noise addition unit includes a random noise addition switch that switches between an on state in which random noise is added to the pixel value and output, and an off state in which the pixel value is output without adding random noise. ,
The delta-sigma modulation unit performs an on state in which the pixel value output from the random noise addition unit is subjected to delta-sigma modulation and outputs, and an off state in which the pixel value output from the random noise addition unit is not subjected to delta-sigma modulation. Including a ΔΣ modulation switch that switches between states,
The gradation conversion device according to (1), wherein when the pixel value is white or black, the control unit controls both the random noise addition switch and the ΔΣ modulation switch to be off.
(3) The random noise addition unit includes a random noise addition switch that switches between an on state in which random noise is added to the pixel value for output and an off state in which the pixel value is output without adding random noise. ,
The delta-sigma modulation unit performs an on state in which the pixel value output from the random noise addition unit is subjected to delta-sigma modulation and outputs, and an off state in which the pixel value output from the random noise addition unit is not subjected to delta-sigma modulation. Including a ΔΣ modulation switch that switches between states,
The control unit controls the random noise addition switch and the ΔΣ modulation switch to be turned on when the pixel value is neither white nor black. The gradation conversion device according to (1) or (2) .
(4) The pixel value includes an RGB (red green blue) signal component in each pixel,
A signal converter that converts the RGB signal component into a YCbCr (luminance color difference) signal;
Among the YCbCr (luminance color difference) signals, when the pixel is considered to be monochrome because the sum of the CbCr (color difference) signals is smaller than a first predetermined value, The corresponding random noise adding unit adds the same random noise to each of the RGB signal components in each pixel. (1) to (3).
(5) The pixel value includes an RGB (red green blue) signal component in each pixel,
A signal converter that converts the RGB signal component into a YCbCr (luminance color difference) signal;
If the sum of the CbCr (color difference) signals out of the YCbCr (luminance color difference) signals is smaller than this, the pixel is considered to be monochrome and is larger than a first predetermined value, and more than the first predetermined value. Is smaller than the second predetermined value, the random noise adding unit corresponding to each RGB signal component in each pixel, when the pixel is considered to be similar to monochrome, each RGB signal in each pixel The gradation conversion device according to any one of (1) to (3), wherein similar random noises are added to the components.
(6) The pixel value includes an RGB (red green blue) signal component in each pixel,
A signal converter that converts the RGB signal component into a YCbCr (luminance color difference) signal;
Of the YCbCr (luminance color difference) signals, if the sum of the CbCr (color difference) signals is smaller than this, the pixel is considered to be monochrome and is larger than a first predetermined value, and smaller than this. When the pixel is considered to be color by being larger than a second predetermined value that is considered to be similar to monochrome, the random noise adding unit corresponding to each RGB signal component in each pixel is The gradation conversion device according to any one of (1) to (3), wherein different random noises are added to the RGB signal components in each pixel.
(7) A gradation conversion method of a gradation conversion apparatus for converting the gradation of an image,
Random noise addition processing for adding random noise to the pixel value of each pixel constituting the image,
A one-dimensional ΔΣ modulation process for applying a one-dimensional ΔΣ modulation to the pixel value output from the random noise adding unit,
A gradation conversion method including a step of performing control processing for controlling operations of the random noise addition unit and the one-dimensional ΔΣ modulation unit based on the pixel value.
(8) A computer that controls a gradation conversion device that converts the gradation of an image;
A random noise adding unit that adds random noise to the pixel value of each pixel constituting the image;
A one-dimensional ΔΣ modulation unit that performs one-dimensional ΔΣ modulation on the pixel value output from the random noise addition unit;
A program that functions as a control unit that controls operations of the random noise addition unit and the one-dimensional ΔΣ modulation unit based on the pixel value.

  DESCRIPTION OF SYMBOLS 1 Television receiver, 11 Tuner, 12 Demultiplexer, 13 Decoder, 14 Noise reduction part, 15 Gradation conversion part, 16 Display control part, 17 Display part, 31 Control part, 32 Random noise output part, 33, 33- 1 to 33-3 random noise addition unit, 34, 34-1 to 34-3 one-dimensional ΔΣ modulation unit, 51, 51-1 to 51-3 random noise generation unit, 52, 52-1 to 52-3 HPF, 71 monochrome detection unit, 72 YCbCr conversion unit, 73 weight control unit, 74 random noise control unit, 91-1 to 91-3 multiplier, 92-1, 92-2 adder, 111 adder, 112 switch, 131 addition , 132 switch, 133 quantizer, 134 subtractor, 135 one-dimensional filter

Claims (8)

In a gradation conversion device for converting the gradation of an image,
A random noise adding unit that adds random noise to the pixel value of each pixel constituting the image;
A one-dimensional ΔΣ modulation unit that performs one-dimensional ΔΣ modulation on the pixel value output from the random noise addition unit;
A gradation conversion device including: a random noise adding unit and a control unit that controls operations of the one-dimensional ΔΣ modulation unit based on the pixel value. The random noise addition unit includes a random noise addition switch that switches between an on state for adding and outputting random noise to the pixel value and an off state for outputting without adding random noise to the pixel value,
The delta-sigma modulation unit performs an on state in which the pixel value output from the random noise addition unit is subjected to delta-sigma modulation and outputs, and an off state in which the pixel value output from the random noise addition unit is not subjected to delta-sigma modulation. Including a ΔΣ modulation switch that switches between states,
The gradation converter according to claim 1, wherein when the pixel value is white or black, the control unit controls both the random noise addition switch and the ΔΣ modulation switch to be off. The random noise addition unit includes a random noise addition switch that switches between an on state for adding and outputting random noise to the pixel value and an off state for outputting without adding random noise to the pixel value,
The delta-sigma modulation unit performs an on state in which the pixel value output from the random noise addition unit is subjected to delta-sigma modulation and outputs, and an off state in which the pixel value output from the random noise addition unit is not subjected to delta-sigma modulation. Including a ΔΣ modulation switch that switches between states,
The gradation conversion apparatus according to claim 1, wherein the control unit controls both the random noise addition switch and the ΔΣ modulation switch to be on when the pixel value is neither white nor black. The pixel value includes an RGB (red green blue) signal component in each pixel,
A signal converter that converts the RGB signal component into a YCbCr (luminance color difference) signal;
Among the YCbCr (luminance color difference) signals, when the pixel is considered to be monochrome because the sum of the CbCr (color difference) signals is smaller than a first predetermined value, The gradation conversion device according to claim 3, wherein the corresponding random noise addition unit adds the same random noise to each of the RGB signal components in each pixel. The pixel value includes an RGB (red green blue) signal component in each pixel,
A signal converter that converts the RGB signal component into a YCbCr (luminance color difference) signal;
If the sum of the CbCr (color difference) signals out of the YCbCr (luminance color difference) signals is smaller than this, the pixel is considered to be monochrome and is larger than a first predetermined value, and more than the first predetermined value. Is smaller than the second predetermined value, the random noise adding unit corresponding to each RGB signal component in each pixel, when the pixel is considered to be similar to monochrome, each RGB signal in each pixel The gradation conversion apparatus according to claim 3, wherein similar random noises are added to the components. The pixel value includes an RGB (red green blue) signal component in each pixel,
A signal converter that converts the RGB signal component into a YCbCr (luminance color difference) signal;
Of the YCbCr (luminance color difference) signals, if the sum of the CbCr (color difference) signals is smaller than this, the pixel is considered to be monochrome and is larger than a first predetermined value, and smaller than this. When the pixel is considered to be color by being larger than a second predetermined value that is considered to be similar to monochrome, the random noise adding unit corresponding to each RGB signal component in each pixel is The gradation conversion apparatus according to claim 3, wherein different random noises are added to the RGB signal components in each pixel. A gradation conversion method of a gradation conversion apparatus for converting the gradation of an image,
Random noise addition processing for adding random noise to the pixel value of each pixel constituting the image,
A one-dimensional ΔΣ modulation process for applying a one-dimensional ΔΣ modulation to the pixel value output from the random noise adding unit,
A gradation conversion method including a step of performing control processing for controlling operations of the random noise addition unit and the one-dimensional ΔΣ modulation unit based on the pixel value. A computer that controls a gradation conversion device that converts the gradation of an image;
A random noise adding unit that adds random noise to the pixel value of each pixel constituting the image;
A one-dimensional ΔΣ modulation unit that performs one-dimensional ΔΣ modulation on the pixel value output from the random noise addition unit;
A program that functions as a control unit that controls operations of the random noise addition unit and the one-dimensional ΔΣ modulation unit based on the pixel value.

Images