JP2010262175A - Image processing device and image processing method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image processing device capable of improving the suppression effect of gradation irregularity caused by dithering when an image is expanded by backlight control or the like, and to provide an image processing method. <P>SOLUTION: A lightness histogram creation section 110 creates a lightness histogram with respect to a frame to be processed, and a maximum lightness maxL is obtained based on a lightness error number threshold ErrThld based on the lightness histogram and a panel characteristic, by a lightness error number counting section 120, a lightness error number threshold comparison section 130 and a BL brightness search section 140. In response thereto, a bit extraction section 210 shifts bits of an input image RGBin to an upper direction, a dithering section 220 dithers the lowest bit, and a color subtraction section 230 subtracts a color. The output of the color subtraction section 230 is gradation-converted by a gradation conversion section 240 by using a gradation conversion table based on the maximum lightness maxL. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to an image processing apparatus that corrects an image displayed on a display panel using liquid crystal or the like.

  As is well known, a conventional image processing device used in a display device using a display panel such as an LCD (Liquid Crystal Display) has a backlight luminance for displaying a bright scene when displaying a dark scene. By making the backlight darker and correcting the gradation higher, the visual brightness of the image is maintained, thereby improving the gradation and color reproducibility of the low gradation image (for example, Patent Document 1). For this reason, the gradation of the image as a whole shifts to the high gradation side, and as a result, the gradation step becomes coarse and gradation skip occurs.

  On the other hand, for small panels such as mobile phones, the input image is 8 bit gradation (24 bit), but the panel performance is inferior to 6 bit gradation, and the input size is reduced due to memory size and memory transfer load restrictions. By applying so-called dithering such as error diffusion and reducing the color to RGB666 (18bit), RGB565 (16bit), etc., and inputting it to the display panel, the gradation skip due to the color reduction is inconspicuous. However, when tone conversion is performed such that the upper 6 bits change in the gamma conversion circuit after dithering, the dithering effect is lost and tone skipping occurs.

JP2007-249085

The conventional image processing apparatus has a problem that the effect of suppressing gradation skip due to dithering cannot be sufficiently expected when the image is expanded by backlight control or the like.
The present invention has been made to solve the above-described problems, and an object thereof is to provide an image processing apparatus and an image processing method that have improved the effect of suppressing gradation skip due to dithering when an image is expanded by backlight control or the like. And

  In order to achieve the above object, according to the present invention, there is provided a shift amount determining means for determining a shift amount for bit-shifting each pixel data in a bit direction in accordance with an image expansion amount, and a shift amount according to the shift amount. The bit shift means for bit-shifting each pixel data in the upper direction, the dithering means for performing dithering based on the data bit-shifted by the bit shift means, and the color reduction for the output of the dithering means And a color reduction means for processing.

  As described above, according to the present invention, each bit-represented pixel data is bit-shifted in the upper direction in accordance with the image expansion amount, dithering is performed based on the bit-shifted data, and further color reduction processing is performed. Like to do.

  Therefore, according to the present invention, the dithering process is performed after the pixel data is bit-shifted upward by the number of bits corresponding to the amount of image expansion. An image processing apparatus and an image processing method with improved suppression effects can be provided.

1 is a circuit block diagram showing a configuration of an embodiment of an image processing apparatus according to the present invention. The figure which shows an example of the histogram which the lightness histogram generation part of the image processing apparatus shown in FIG. 1 produced | generated. FIG. 3 is a diagram illustrating an example of characteristics of a gradation conversion table created by a gradation conversion table creating unit of the image processing apparatus shown in FIG. 1. The figure for demonstrating a dithering process by the dithering part of the image processing apparatus shown in FIG. The figure for demonstrating the bit shift by the image processing apparatus shown in FIG.

Hereinafter, an embodiment of the present invention will be described with reference to the drawings.
FIG. 1 shows the configuration of an image processing apparatus according to an embodiment of the present invention. This image processing apparatus includes a frame unit processing unit 100 and a pixel unit processing unit 200. The frame unit processing unit 100 processes the input image RGBin (RGB888) in units of frames, while the pixel unit processing unit 200 performs processing in units of pixels constituting a frame. In the following description, when expressed as RGBxyz, it indicates that the data of each pixel of R, G, and B is composed of x bits, y bits, and z bits, respectively.

  The frame unit processing unit 100 includes a lightness histogram generation unit 110, a lightness error number counting unit 120, a lightness error number threshold comparison unit 130, a BL (backlight) luminance search unit 140, a backlight dimming unit 150, A gradation conversion table creation unit 160 and a gradation conversion table storage unit 170 are provided. The pixel unit processing unit 200 includes a bit extraction unit 210, a dithering unit 220, a color reduction unit 230, and a gradation conversion unit 240.

The lightness histogram generation unit 110 creates a lightness histogram histoL for each frame based on the input image RGBin as illustrated in FIG. Specifically, the brightness histogram histoL is generated by counting the brightness of the pixels constituting the frame for each brightness level L based on the following equation (1).

Here, the lightness L used for each pixel is the maximum value among R, G, and B of each pixel of RGBin as in the following equation (2).

The lightness error number counting unit 120 calculates the following equation (3) based on the search gradation threshold L-Thld given from the BL luminance search unit 140 described later and the lightness histogram histoL obtained by the lightness histogram generation unit 110. Therefore, a frequency equal to or higher than the search gradation threshold L-Thld is obtained, and this is output as the brightness error number ErrL.

  The lightness error number threshold comparison unit 130 compares the lightness error number ErrL obtained by the lightness error number counter 120 with a preset lightness error number threshold ErrThld, and determines whether the lightness error number ErrL is greater than the lightness error number threshold ErrThld. Determine whether. Here, when the lightness error number ErrL is larger than the lightness error number threshold value ErrThld, “1” is output as the determination result. On the other hand, when the lightness error number ErrL is less than or equal to the lightness error number threshold value ErrThld, “0” is output.

  Here, the lightness error number threshold ErrThld is a value set according to the balance between the image quality of the bright part and the image quality of the dark part. For example, when it is set to a small value, the gradation of the bright part can be maintained. On the other hand, when it is set to a large value, low power consumption and improvement of the gradation and color reproducibility of the dark part can be expected. For this reason, the lightness error number threshold ErrThld is set in advance depending on which of image quality and power consumption is important. Further, the brightness error indicates a state where gradation cannot be maintained.

  The BL luminance search unit 140 sets a search gradation threshold L-Thld for the lightness error number counting unit 120. The initial value of the search gradation threshold L-Thld is 255, and the brightness error number counter 120 searches the brightness error number counter 120 until the determination result output by the brightness error number threshold comparator 130 is “1”. The process of setting the gradation threshold L-Thld is repeated. Eventually, when the determination result becomes “1”, the BL luminance search unit 140 outputs the search gradation threshold L-Thld at that time as the maximum brightness maxL to the gradation conversion table creation unit 160.

Then, the BL luminance search unit 140 determines the luminance value BL of the backlight of the display based on the maximum brightness maxL, and notifies the backlight dimming unit 150 of the luminance value BL. The luminance value BL is obtained according to the following equation (4). In Expression (4), BLmax is the maximum value of the backlight luminance of the display. In addition, γ is a characteristic of the display panel, and is generally 2.2 according to the characteristics of the cathode ray tube.

  Note that the gradation of the bright portion can be maintained as the lightness error number threshold ErrThld set in accordance with the balance between the image quality of the bright portion and the dark portion of the frame is smaller. However, lowering the backlight brightness can be expected to reduce power consumption, improve gradation and color reproducibility in dark areas, and noise is included in the image, so the brightness error number threshold ErrThld is not zero. For example, it is set to about 10% of the total number of pixels.

The backlight dimming unit 150 controls the luminance of the backlight of the display based on the luminance value BL notified from the BL luminance searching unit 140.
The gradation conversion table creation unit 160 obtains the bit shift number shiftL of the input image RGBin (RGB888) based on the maximum brightness maxL in order to maintain the brightness of the display image before and after the change of the backlight luminance, Create a key conversion table LUT. The gradation conversion table LUT is recorded in the gradation conversion table storage unit 170.

Specifically, the gradation conversion table creation unit 160 uses the reciprocal of the maximum brightness maxL as an image expansion amount, and separates this expansion amount into a bit shift number shiftL and an expansion gain gainL. As shown in the following equation, the bit shift number shiftL is an expansion amount in bit units (double unit), and the expansion gain gainL indicates an expansion amount less than double. Note that INT (x) is the maximum integer value that does not exceed x. The expansion amount is a value indicating how many times the pixel value is increased.

  For example, assuming that the maximum brightness maxL is 100, the gradation conversion table creation unit 160 obtains the bit shift number shiftL = 1 and the expansion gain gainL = 1.275. In this case, since the expansion gain gainL is 1.275, the gradation conversion table creation unit 160 creates a gradation conversion table LUT as shown in FIG. That is, when the expansion gain gainL = 1.275, a gradation conversion table LUT is created that has a maximum value of 255 when the RGB value of the input image is 200 or more.

  The bit extraction unit 210 bit-shifts the RGB value of each pixel of the input image RGBin (RGB888) to the left by the bit shift number shiftL (lower bits in the upper bit direction) to obtain a shifted image shiftRGB. For example, when the bit shift number shiftL = 1 (extension gain is 2), the input image RGBin (RGB888) is shifted to the left by 1 bit, and the input image RGBin (RGB888) becomes the shifted image shiftRGB (RGB777).

  The dithering unit 220 performs dithering processing on the R, G, and B pixel data of the output (shifted image shiftRGB) of the bit extraction unit 210. For example, the dithering unit 220 processes frames in the raster scan order. A pixel to be configured is selected, and a dithering process is performed for each pixel. Here, a case where error diffusion is adopted as an example of the dithering processing method will be described.

  The dithering unit 220 includes a spread value adding unit 221, a spread bit extracting unit 222, and a spread value holding line buffer 223. The diffusion value holding line buffer 223 includes a buffer that holds a diffusion value for each pixel constituting the frame.

  The diffusion value adding unit 221 reads the diffusion value obtained by the dithering process before the pixel to be processed from the buffer corresponding to the pixel to be processed among the buffers of the diffusion value holding line buffer 223, and performs processing To the pixel value (the value of the corresponding pixel in the shifted image shiftRGB). Thereby, ditherRGB (RGB888) is obtained.

  In the initial stage of dithering (raster scan), since the diffusion value is not obtained, the initial value is used. The spread value holding line buffer 223 is initialized to an initial value 0 when a frame to be processed is switched.

  The diffusion bit extraction unit 222 extracts, from the output of the diffusion value addition unit 221 (ditherRGB), the lower bits that are reduced by the color reduction processing performed by the subsequent color reduction unit (color reduction memory) 230 for the pixel to be processed. Corresponding diffusion values are accumulated and recorded in the buffer of the diffusion value holding line buffer 223 corresponding to pixels around the pixel to be processed.

  For example, as shown in FIG. 4, when the value of the pixel to be processed is 31 (= 28 + 3), 3 corresponding to the lower 2 bits is extracted, and this value corresponds to the surrounding pixels. It is distributed to the buffer to be recorded and recorded cumulatively. The distribution pattern is not limited to the example of FIG. 4, and various patterns are conceivable. The value stored in the buffer in this way is used by the diffusion value adding unit 221 as a diffusion value.

  The color reduction unit 230 includes a color reduction memory that stores the upper 6 bits of ditherRGB (RGB888) output from the diffusion value addition unit 221. Thereby, the pixel value read from the color reduction unit 230 becomes 6 bits. Hereinafter, it is referred to as RGBout (RGB666) before conversion.

The gradation conversion unit 240 converts the pixel value of RGBout (RGB666) before conversion based on the following equation (6) using the gradation conversion table LUT stored in the gradation conversion table storage unit 170.

  As described above, the image processing apparatus configured as described above shifts the bits of the input image RGBin in the upper direction according to the maximum brightness maxL determined based on the brightness histogram of the input image RGBin and the characteristics of the display panel, Dithering is performed on the least significant bit, and then color reduction processing is performed.

  Therefore, according to the image processing apparatus having the above configuration, the bits of the input image RGBin are shifted upward in accordance with the maximum brightness maxL. Therefore, in the conventional processing, as shown in FIG. The lower bits are not discarded and are used in the dithering process. For this reason, it is possible to improve the effect of suppressing gradation skip in dithering processing accompanying image expansion by backlight control or the like.

  In the image processing apparatus having the above configuration, the gradation conversion table creating unit 160 creates a gradation conversion table corresponding to the decompression amount for the decompression amount less than the bit unit that is not reflected by the shift in bit units. Based on this table, the gradation conversion unit 240 performs gradation conversion on the color-reduction processing output (RGBout (RGB666) before conversion). Therefore, when the expansion amount is 2.5 times, the shift in bit units is the bit shift number ShiftL = 1, and the table converts to a gradation of 1.25 times.

  Note that the present invention is not limited to the above-described embodiment as it is, and can be embodied by modifying the constituent elements without departing from the scope of the invention in the implementation stage. In addition, various inventions can be formed by appropriately combining a plurality of constituent elements disclosed in the embodiment. Further, for example, a configuration in which some components are deleted from all the components shown in the embodiment is also conceivable. Furthermore, you may combine suitably the component described in different embodiment.

  For example, in the above embodiment, the gradation conversion table creating unit 160 obtains the reciprocal of the maximum brightness maxL as the image expansion amount, but the present invention is not limited to this. For example, the user may set how many times the pixel value of the image is increased (that is, the expansion amount) through a user interface (not shown), or an average value of pixel brightness is obtained and based on this. Thus, the extension amount may be determined.

  In the above embodiment, the maximum brightness maxL is obtained and the backlight brightness is determined based on this and the maximum value of the backlight brightness of the display. Instead, as described above, The backlight luminance may be determined based on the extension amount and the maximum value of the backlight luminance of the display. The present invention can also be applied when the backlight luminance is a fixed value.

  DESCRIPTION OF SYMBOLS 100 ... Frame unit processing part, 110 ... Lightness histogram generation part, 120 ... Lightness error number counting part, 130 ... Lightness error number threshold comparison part, 140 ... BL brightness | luminance search part, 150 ... Backlight dimming part, 160 ... Tone Conversion table creation unit, 170 ... gradation conversion table storage unit, 200 ... pixel unit processing unit, 210 ... bit extraction unit, 220 ... dithering unit, 221 ... diffusion value addition unit, 222 ... diffusion bit extraction unit, 223 ... diffusion Value holding line buffer, 230... Color reduction section, 240... Gradation conversion section.

Claims (10)

A shift amount determining means for determining a shift amount for bit-shifting each data of the pixel represented by the bit in the upper direction according to the expansion amount of the image
Bit shift means for bit-shifting each data of the pixel in the upper direction according to the shift amount;
Dithering means for performing dithering based on the data bit-shifted by the bit shift means;
An image processing apparatus comprising: a color reduction unit that performs a color reduction process on the output of the dithering unit. Backlight luminance determining means for determining the backlight luminance based on the image expansion amount and the maximum luminance value of the display panel;
A shift amount determining means for determining a shift amount for bit-shifting each data of the pixel represented by the bit in an upper direction according to the expansion amount;
Bit shift means for bit-shifting each data of the pixel in the upper direction according to the shift amount;
Dithering means for performing dithering based on the data bit-shifted by the bit shift means;
An image processing apparatus comprising: a color reduction unit that performs a color reduction process on the output of the dithering unit.   The image processing apparatus according to claim 1, further comprising a gradation conversion unit that converts a gradation of an output of the color reduction unit according to the expansion amount.   The gradation conversion means converts the gradation of the output of the color reduction means according to the expansion amount that is less than the shift amount determined by the shift amount determination means. Item 3. The image processing apparatus according to Item 2. Brightness determination means for determining a maximum value of brightness based on a histogram of brightness of pixels constituting the input image;
Backlight luminance determining means for determining backlight luminance based on the maximum value of brightness and the maximum luminance value of the display panel;
A shift amount determining means for determining a shift amount for bit-shifting each data of the pixel expressed in bits according to the maximum value of the brightness;
Bit shift means for bit-shifting each data of the pixel in the upper direction according to the shift amount;
Dithering means for performing dithering based on the data bit-shifted by the bit shift means;
An image processing apparatus comprising: a color reduction unit that performs a color reduction process on the output of the dithering unit.   The image processing apparatus according to claim 5, further comprising a gradation conversion unit that converts a gradation of an output of the color reduction unit according to the maximum value of the brightness. The shift amount determination means obtains an expansion amount according to the maximum value of the brightness, determines the shift amount according to the expansion amount,
The gradation conversion means converts the gradation of the output of the color reduction means in accordance with the expansion amount that is less than the shift amount determined by the shift amount determination means. Image processing apparatus. A shift amount determining step for determining a shift amount for bit-shifting each data of the pixel represented by the bit in an upper direction according to the expansion amount of the image
In accordance with the shift amount, a bit shift step of bit-shifting each data of the pixel in the upper direction,
A dithering step for performing dithering based on the data shifted in the bit shift step;
An image processing method comprising: a color reduction process for performing color reduction processing on the output of the dithering means. A backlight luminance determination step for determining the backlight luminance based on the amount of image expansion and the maximum luminance value of the display panel;
A shift amount determination step for determining a shift amount for bit-shifting each data of the pixel represented by the bit in the upper direction according to the expansion amount;
In accordance with the shift amount, a bit shift step of bit-shifting each data of the pixel in the upper direction,
A dithering step for performing dithering based on the data shifted in the bit shift step;
An image processing method comprising: a color reduction process for performing a color reduction process on the output of the dithering process. A brightness determination step for determining a maximum value of brightness based on a histogram of brightness of pixels constituting the input image;
A backlight luminance determining step for determining a backlight luminance based on the maximum value of the brightness and the maximum luminance value of the display panel;
A shift amount determining step of determining a shift amount for bit-shifting each data of the pixel expressed in bits according to the maximum value of the brightness;
In accordance with the shift amount, a bit shift step of bit-shifting each data of the pixel in the upper direction,
A dithering step for performing dithering based on the data shifted in the bit shift step;
An image processing method comprising: a color reduction process for performing a color reduction process on the output of the dithering process.

Images