JP5135134B2 - Image processing apparatus and method - Google Patents

Description

  The present invention relates to an image processing apparatus and method for generating a high-quality image.

  In various models that handle image data, such as recent televisions, mobile phones, and digital cameras, it is common to perform image quality enhancement processing in order to improve the image data quality. Such tone correction processing may be referred to as gamma correction processing (γ correction processing).

  In the case of a television, a saturation enhancement process for enhancing the saturation is also performed on the image signal that has been subjected to the gradation correction process as described above.

  However, if the above saturation enhancement processing is further applied to the image data in which the saturation of the dark portion (low luminance portion) of the image is particularly increased by performing black extension by the gradation correction processing, the dark portion becomes too high in saturation, There is a problem that the image data includes noise.

  This problem also occurs in high-quality processing of mobile phone images, digital camera image processing, and the like.

Patent Document 1 discloses a technique for accurately determining a saturation modulation characteristic, that is, a characteristic that defines a saturation level gain with respect to a luminance level in order to reduce color noise in a low luminance region. Yes.
JP 2003-235055 A

  However, Patent Document 1 does not consider the frequency characteristics of saturation in the low luminance region. For this reason, there is a problem in that all colors above a specific frequency are cut off due to the saturation modulation characteristic, and the color tone of the texture in the low luminance region is lost more than necessary.

  Accordingly, the present invention has been made in view of the above problems, and an object of the present invention is to provide an image processing apparatus and method capable of generating a high-quality image with a natural tone and high gradation and contrast. And

  The present invention generates a smoothed image by separating the input image into a brightness component image indicating brightness and a color component image indicating color, and smoothing the brightness component image. A filter, a contrast calculation unit that calculates a contrast component image of the input image from the brightness component image and the smoothed image, and level conversion so that the contrast is corrected for the smoothed image, An image correction unit that generates a corrected smoothed image, a brightness correction unit that generates a corrected brightness component image by multiplying the corrected smoothed image and the contrast component image, the corrected smoothed image, and the brightness A coefficient calculation unit that calculates a correction coefficient for correcting the color component image based on the rate of change of the component image; and correcting the color component image using the correction coefficient, thereby correcting the color component image The A color correction unit for forming an image processing apparatus characterized by having a.

  According to the present invention, it is possible to generate a high-quality image with natural color tone and high gradation and contrast.

  Hereinafter, an image processing apparatus 1 according to an embodiment of the present invention will be described with reference to FIG.

(1) Configuration of Image Processing Device 1 FIG. 1 is a block diagram showing the configuration of the image processing device 1 of the present embodiment.

  The image processing apparatus 1 includes a separation unit 10, a filter 20, a contrast calculation unit 30, a contrast correction unit 40, an image correction unit 50, a brightness correction unit 60, a coefficient calculation unit 70, a color correction unit 80, and an output unit 90. .

  The image processing apparatus 1 can be realized by using, for example, a general-purpose computer apparatus as basic hardware. That is, the separation unit 10, the filter 20, the contrast calculation unit 30, the contrast correction unit 40, the image correction unit 50, the brightness correction unit 60, the coefficient calculation unit 70, the color correction unit 80, and the output unit 90 are the above computer devices. This can be realized by causing a processor mounted on the computer to execute a program. At this time, the image processing apparatus 1 may be realized by installing the above program in a computer device in advance, or may be stored in a storage medium such as a CD-ROM or distributed through the network. Then, this program may be realized by appropriately installing it in a computer device.

  The separation unit 10 separates the input image into a brightness component image Y and a color component image UV for each pixel.

  The filter 20 generates a smoothed image LPF_P1 by superimposing a smoothing filter on the brightness component image Y for each pixel.

  The contrast calculation unit 30 generates the contrast component image CP of the brightness component image Y by dividing the brightness component image Y by the smoothed image LPF_P1 for each pixel.

  The contrast correction unit 40 performs level conversion of the contrast component image CP for each pixel to generate a corrected contrast component image T_CP.

  The image correction unit 50 performs level conversion of the smoothed image LPF_P1 for each pixel to generate a corrected smoothed image T_LPF_P1.

  The brightness correction unit 60 multiplies the corrected contrast component image T_CP and the corrected smoothed image T_LPF_P1 for each pixel to generate a corrected brightness component image.

  The coefficient calculation unit 70 calculates a correction coefficient H for correcting the color component image UV from the corrected smoothed image T_LPF_P1 and the brightness component image Y for each pixel.

  The color correction unit 80 corrects the color component image UV using the correction coefficient H for each pixel, generates a corrected color component image, and outputs it from the output unit 90.

  Hereinafter, the separation unit 10, the filter 20, the contrast calculation unit 30, the contrast correction unit 40, the image correction unit 50, the brightness correction unit 60, the coefficient calculation unit 70, the color correction unit 80, and the output unit 90 will be described in detail.

(2) Separation unit 10
First, the separation unit 10 will be described.

  The separation unit 10 separates the brightness component image Y and the color component image UV for each pixel of the input image.

  For example, when the input image is input in the YUV format, the Y signal is a brightness component image, and the U signal and the V signal are tint component images. Similarly, in the case of YCbCr, the Y signal is a brightness component image, and the Cb signal and the Cr signal are tint component images. In the case of input in RGB, component separation is performed by converting into YUV, YCbCr or the like by generally known signal conversion.

In the case of RGB input, for example, conversion is performed based on the following equation (1) to separate the brightness component image and the color component image.

  The same applies to other input formats.

Here, for the sake of explanation, the origin is set at the upper left corner of the image to be processed, and the x axis is set in the horizontal direction and the y axis is set in the vertical direction. Thereafter, the value of the brightness component image of the pixel at coordinates (x, y) is P ₀ (x, y), and the values of the color component images are U ₀ (x, y) and V ₀ (x, y), respectively. It will be expressed as

(3) Filter 20
Next, the filter 20 will be described.

  The filter 20 generates a smoothed image LPF_P1 by superimposing a smoothing filter on the brightness component image Y for each pixel.

  The smoothing filter used in the filter 20 is, for example, a Gaussian filter, an average value filter, an edge-preserving nonlinear low-pass filter, an epsilon filter, or the like.

  That is, the filter used in the filter 20 may be any filter that can extract at least a low frequency component. If necessary, it may be designed to pass a specific frequency. The above epsilon filter is an example.

A value at the coordinates (x, y) of the smoothed image LPF_P1 generated using these filters is represented as P _{LPF_P1} (x, y).

(4) Contrast calculation unit 30
Next, the contrast calculation unit 30 will be described.

The contrast calculation unit 30 generates the contrast component image CP by dividing the brightness component image Y by the smoothed image LPF_P1 at each pixel. That is, the value P _CP (x, y) at the coordinates (x, y) of the contrast component image CP can be obtained by the following equation (2).

(5) Contrast correction unit 40
Next, the contrast correction unit 40 will be described.

In the contrast correction unit 40, the contrast of the contrast component image CP is converted for each pixel, thereby expanding or compressing the contrast of the image. There are various methods for converting the contrast component image. For example, the following method (3) is used.

  When the value of γ in equation (3) is greater than 1, the contrast component is expanded. Incidentally, when γ becomes smaller than 1, the contrast component is compressed.

(6) Image correction unit 50
Next, the image correction unit 50 will be described.

  The image correction unit 50 performs level conversion so as to expand the contrast of the smoothed image LPF_P1 for each pixel, and generates a corrected smoothed image T_LPF_P1.

  Various methods can be considered as a method for correcting the contrast of the smoothed image LPF_P1, and a generally known method may be used as the method. For example, histogram equalization, power level conversion, etc.

In the conversion using the cumulative histogram, the following equation (4) is obtained.

  In Equation (4), HE (k) is obtained by normalizing the cumulative histogram to [minimum output value, maximum output value]. According to equation (4), the greater the value of α, the greater the contrast. Incidentally, when α becomes negative, compression is performed.

(7) Brightness correction unit 60
Next, the brightness correction unit 60 will be described.

  The brightness correction unit 60 multiplies the corrected contrast component image T_CP and the corrected smoothed image T_LPF_P1 for each pixel to generate a corrected brightness component image.

In the brightness correction unit 60, the corrected contrast component image T_CP and the corrected smoothed image T_LPF_P1 are multiplied by each pixel to generate a corrected brightness component image. The value P _OUT (x, y) at the coordinates (x, y) of the corrected brightness component image can be obtained by the following equation (5).

(8) Coefficient calculation unit 70
Next, the coefficient calculation unit 70 will be described.

The coefficient calculation unit 70 calculates a correction coefficient H for correcting the color component image UV from the corrected smoothed image T_LPF_P1 and the brightness component image Y. The correction coefficient H is calculated for each pixel. The correction coefficient H is determined for each pixel according to the change rate of the brightness component image Y and the corrected smoothed image T_LPF_P1. The correction coefficient H (x, y) at the coordinates (x, y) is obtained by the following equation (6).

This correction coefficient H (x, y) indicates the change rate of the low frequency component of the final output P _OUT (x, y) of the brightness component image and the input brightness image component Y. By using the correction coefficient H (x, y) for correcting the color component, it is possible to suppress color noise riding on the high frequency and to suppress the color tone of the low frequency component.

(9) Color correction unit 80
Next, the color correction unit 80 will be described.

The color correction unit 80 corrects the color component image using the correction coefficient H (x, y) to generate a corrected color component image. The pixel values of the corrected color component image at the coordinates (x, y) are U _out (x, y) and V _out (x, y).

U _out (x, y) and V _out (x, y) are obtained by the following equation (7).

When U _out and V _out corrected by the equation (7) exceed the set values, the output is set according to the set values.

(10) Output unit 90
Finally, the output unit 90 varies depending on the required form of image output. For example, when an output format such as YUV is required, the output unit 90 may output the output as P _OUT , U _out (x, y), and V _out (x, y).

In addition, when an output format such as RGB is required, the output unit 90 performs an inverse conversion of the conversion performed by the separation unit 10 and outputs the result. In the case of the conversion according to the above embodiment, if the output is R′G′B ′, for example, the following equation (8) is obtained.

(11) Modification Examples The present invention is not limited to the above-described embodiment, and various modifications can be made without departing from the gist of the present invention.

  For example, although the contrast correction unit 40 is provided in the image processing apparatus 1 of the above embodiment, the brightness correction unit 60 multiplies the contrast component image CP and the corrected smoothed image T_LPF_P1 to correct the corrected brightness. A component image may be generated. Even in this case, it is possible to generate a high-quality image with a natural color tone and high gradation and contrast.

1 is a block diagram illustrating a configuration of an image processing apparatus according to an embodiment of the present invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Image processing apparatus 10 Separation part 20 Filter 30 Contrast calculation part 40 Contrast correction part 50 Image correction part 60 Brightness correction part 70 Coefficient calculation part 80 Tint correction part 90 Output part

Claims (8)

A separation unit that separates the input image into a brightness component image indicating brightness and a color component image indicating color;
A filter that smoothes the brightness component image to generate a smoothed image;
A contrast calculation unit that calculates a contrast component image of the input image from the brightness component image and the smoothed image;
An image correction unit that performs level conversion on the smoothed image so that contrast is corrected, and generates a corrected smoothed image;
A brightness correction unit that multiplies the corrected smoothed image and the contrast component image to generate a corrected brightness component image;
A coefficient calculation unit for calculating a correction coefficient for correcting the color component image based on the change rate of the corrected smoothed image and the brightness component image;
A color correction unit that corrects the color component image using the correction coefficient to generate a corrected color component image;
An image processing apparatus comprising: The contrast component image further includes a contrast correction unit that performs level conversion so that the contrast is corrected, generates a corrected contrast component image, and outputs the corrected contrast component image to the brightness correction unit.
The image processing apparatus according to claim 1. An output unit for multiplying and outputting the corrected brightness component image and the corrected color component image;
The image processing apparatus according to claim 1. The filter is a smoothing filter that extracts a low-frequency component.
The image processing apparatus according to claim 1. The filter is a Gaussian filter, an average value filter, an edge-preserving nonlinear low-pass filter, or an epsilon filter.
The image processing apparatus according to claim 1. The separation unit, the filter, the contrast calculation unit, the image correction unit, the brightness correction unit, the coefficient calculation unit, and the color correction unit perform processing for each pixel of the image.
The image processing apparatus according to claim 1. A separation step of separating the input image into a brightness component image indicating brightness and a color component image indicating color;
A filter that smoothes the brightness component image to generate a smoothed image;
A contrast calculating step for calculating a contrast component image of the input image from the brightness component image and the smoothed image;
An image correction step for converting the level of the smoothed image so that the contrast is corrected and generating a corrected smoothed image;
A brightness correction step of multiplying the corrected smoothed image and the contrast component image to generate a corrected brightness component image;
A coefficient calculating step of calculating a correction coefficient for correcting the color component image based on the change rate of the corrected smoothed image and the brightness component image;
A color correction step of correcting the color component image using the correction coefficient to generate a corrected color component image;
An image processing method comprising: A separation function for separating an input image into a brightness component image indicating brightness and a color component image indicating color;
A filter that smoothes the brightness component image to generate a smoothed image;
A contrast calculation function for calculating a contrast component image of the input image from the brightness component image and the smoothed image;
An image correction function that performs level conversion on the smoothed image so that contrast is corrected, and generates a corrected smoothed image;
A brightness correction function for multiplying the corrected smoothed image and the contrast component image to generate a corrected brightness component image;
A coefficient calculation function for calculating a correction coefficient for correcting the color component image based on the change rate of the corrected smoothed image and the brightness component image;
A color correction function that corrects the color component image using the correction coefficient to generate a corrected color component image;
An image processing program for realizing the above in a computer.

Images