JP4037336B2 - Image processing apparatus and method, and program - Google Patents

Description

  The present invention relates to an image processing apparatus and method capable of improving the compression rate particularly when compressing image data, and a program for causing a computer to execute the image processing method.

  Since image data has a large capacity, it is compressed so that it can be handled easily by reducing the capacity. In particular, in order to display an image of a relatively large size created using a personal computer or the like on a monitor of a portable terminal device such as a mobile phone, the data capacity is greatly increased by converting the image size and increasing the compression rate. It is small. However, when the compression rate is increased, the capacity is reduced, but the image quality is deteriorated.

  For this reason, various techniques have been proposed to suppress the deterioration of image quality while improving the compression rate. For example, the image data is subjected to filtering processing by a low-pass filter and then compressed after removing high-frequency components, thereby improving the compression ratio without significantly reducing the image quality (see Patent Document 1). Of obtaining the statistical property information of the image, obtaining a filter for reducing the signal component that has little influence on the image quality based on this information, and filtering the image data by this filter to prevent image quality deterioration during compression (patent Reference 2) has been proposed. In addition, the compression rate is improved by a compression algorithm on the premise that there is no data between lines for each of two fields constituting one frame at an interval of one line for an interlaced moving image. A method is known (see Patent Document 3).

  By the way, in the methods described in Patent Documents 1 and 2, when decompressing compressed image data, it is necessary to perform post-processing to restore the image quality of the original image to the decompressed image data. is there. For this reason, a display device that displays compressed image data, such as a mobile phone, for example, needs to have a dedicated compression / decompression algorithm for performing the post-processing. Therefore, in order to display image data compressed by the methods of Patent Documents 1 and 2 on an existing display device that has already been widely used, the existing compression / decompression algorithm must be replaced with a dedicated compression / decompression algorithm.

  However, it is not practical to replace the compression / decompression algorithm in a display device that has already been widely used, and there are many cases where it is not possible to replace the compression / decompression algorithm. The current situation is that data cannot be expanded to high image quality. The method of Patent Document 3 can be displayed on a display device equipped with a decompression algorithm corresponding to interlaced image data, but can be displayed on a display device not compatible with the interlaced method. Can not.

For this reason, the applicant proposes a method for correcting all pixel values to a constant value, for example, for each horizontal line on the image so that the compression rate can be improved without a dedicated compression / decompression algorithm. (See Patent Document 4). According to this method, the information of the corrected line is lost, but the compression rate is improved because all the pixel values of the predetermined number of lines are constant.
JP-A-10-257484 JP 2002-209209 A JP-A-8-317386 Japanese Patent Application No. 2003-137170

  However, in the method described in Patent Document 4, for example, when the constant value is black, the image represented by the corrected image data is compared with the image represented by the uncorrected image data. Color and brightness will be reduced. Conversely, when the constant value is white, the color is reduced before and after correction, and the brightness is emphasized. For this reason, depending on the method described in Patent Document 4, the image quality is changed as compared with the image before correction.

  The present invention has been made in view of the above circumstances, and an object thereof is to improve the compression rate while maintaining the image quality even without a dedicated compression / decompression algorithm.

An image processing apparatus according to the present invention includes input receiving means for receiving input of original image data representing an original image in which pixels are arranged in a two-dimensional shape,
An image processing apparatus comprising correction means for correcting all pixel values of a predetermined number of lines of the original image to a constant value so that a compression rate is improved when the original image data is compressed by a predetermined compression algorithm. Because
The image processing apparatus further includes image quality changing means for performing processing for changing the image quality of the original image according to the fixed value with respect to the original image data before being corrected by the correcting means.

  The “predetermined compression algorithm” takes into account the redundancy of the original image represented by the original image data in either one of the vertical and horizontal lines without pre-processing such as special filtering processing. This is an algorithm for performing compression processing, and means an algorithm for generating image data in a file format that can be handled by a general-purpose personal computer, portable terminal device, or other device that displays images. Specific file formats include gif, png, and the like.

  As the “constant value”, the maximum value, the minimum value, the average luminance of all the pixels, a value lower than the average luminance, or the like can be used.

  “Every predetermined number” means every arbitrary line such as every 1 line or every 2 lines. Note that the line interval to be corrected to a constant value may be determined in consideration of the position of the line that can improve the compression rate.

  The constant value may be a representative value of all pixel values of the image.

  As the representative value, an average value, an intermediate value, an average value obtained by weighting the central portion, or the like can be used.

  “Change the image quality according to a certain value” means that the original image whose quality has been changed is corrected so that the corrected image obtained by correcting the original image and the image quality of the original image are substantially the same. This means changing the image quality.

  In the image processing apparatus according to the present invention, the image quality changing means may be a means for performing processing for changing the image quality of the original image by changing the lightness and / or saturation of the original image.

In the image processing apparatus according to the present invention, the image quality is improved by comparing the original image with the corrected image represented by the corrected image data obtained by correcting the original image data by correction means. Provided with a parameter determining means for determining a parameter for changing,
The correction unit may be a unit that performs a process of changing the image quality according to the parameter determined by the parameter determination unit.

  The image processing apparatus according to the present invention may further include a compression unit that compresses the corrected image data obtained by the correction unit using the predetermined compression algorithm.

The image processing method according to the present invention accepts input of original image data representing an original image in which pixels are arranged two-dimensionally,
An image processing method for correcting all pixel values of a predetermined number of lines of the original image to a constant value so that a compression rate is improved when the original image data is compressed by a predetermined compression algorithm,
The original image data before the correction is subjected to a process of changing the image quality of the original image according to the certain value.

  The image processing method according to the present invention may be provided as a program for causing a computer to execute the image processing method.

  According to the present invention, since all the pixel values of a predetermined number of lines of the original image are corrected to a constant value so that the compression rate is improved when the original image data is compressed by a predetermined compression algorithm, The compression ratio can be improved for the line whose value has been changed. Further, the compressed original image data can be decompressed without using a special decompression algorithm. When all the pixel values of a given number of lines are corrected to a constant value, the information of that line is lost and the image quality is changed before and after the correction, but it is represented by the original image data according to the fixed value. Since the processing for changing the image quality of the original image is performed on the original image data before correction, it is possible to prevent the image quality of the corrected image represented by the corrected image data from being changed. Therefore, the compression rate can be improved while preventing the image quality from changing.

  In addition, if the image quality of the original image is changed by changing the brightness and / or saturation of the original image, the change of the brightness and / or saturation of the corrected image represented by the corrected image data is prevented. can do.

  In addition, the parameter for changing the image quality is determined by comparing the original image with the corrected image represented by the corrected image data obtained by correcting the original image data. Since the parameters can be determined so that there is no change, it is possible to reliably prevent a change in image quality.

  Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a schematic block diagram showing the configuration of an image compression system including an image processing apparatus according to the first embodiment of the present invention. As shown in FIG. 1, the image compression system 1 according to the first embodiment of the present invention inputs the original image data S0 by reading the image data S0 from a digital camera, a network database, a hard disk, various media, and the like. An input unit 2 to receive, and an image quality changing unit for changing the image quality of the original image represented by the original image data S0 (hereinafter, the same reference symbol S0 is used for the original image) to obtain the image quality changed image data S1 in which the image quality has been changed. 3 and corrected image data by correcting all the pixel values of a predetermined number of lines of the image quality changed image represented by the image quality changed image data S1 (hereinafter, the same code S1 is used for the image quality changed image) to a constant value. A correction unit 4 that obtains S2 and a compression unit 5 that compresses the corrected image data S2 to obtain compressed image data S3 are provided.

  The input unit 2 includes an interface corresponding to the digital camera for reading the original image data S0 and a database on the network.

  The correction unit 4 corrects all the pixel values of the horizontal lines in the image quality-changed image S1 so as to be a constant value for each line (hereinafter, the reference image S2 is also used for the corrected image). The corrected image data S2 representing is acquired. That is, as shown in FIG. 2, the horizontal lines of the image quality-change image S1 are corrected so as to have a constant value (shaded portion) for each line.

  The image quality changing unit 3 includes a feature amount extracting unit 31 that extracts a feature amount of the original image S0 from the original image S0, a feature amount correcting unit 32 that corrects the feature amount extracted by the feature amount extracting unit 31, and a corrected feature. And an inverse conversion unit 33 that inversely converts the amount to obtain the image quality-changed image data S1.

  The feature amount extraction unit 31 converts each pixel value of the original image S0 composed of RGB color data by the following conversion formula (I) to characterize H (hue), S (saturation), and V (lightness). Get as a quantity. When the value of RGB color data is 8 bits and takes a value of 0 to 255, it is divided into 255 and normalized so as to have a value of 0 to 1 and then converted to HSV.

(1) V = MAX {R, G, B}
(2) When V = 0 S = 0, H = Undefined (3) When other than (2) (not V = 0) (3-1) v = MIN {R, G, B} v: R, G , B (3-2) S = (V−v) / V
(3-3) If R = V, H0 = (GB) / (V−v)
If G = V, H0 = 2 + (BR) / (V-v)
If B = V, H0 = 4 + (RG) / (V-v)
(3-4) H = H0 × 60
(3-5) If H <0, H = (H0 × 60) +360 (I)
H takes a value of 0 to 359, and S and V take a value of 0 to 1.

  The feature amount correction unit 32 corrects the saturation S and the lightness V ′ extracted by the feature amount extraction unit 31 according to the fixed value of the line changed by the correction unit 4 to obtain the corrected saturation S ′ and the correction lightness V ′. obtain. Here, when the constant value of the line is a dark color such as black, the corrected image S2 has reduced saturation and lightness compared to the original image S0. On the contrary, when the fixed value of the line is a bright color such as white, the corrected image S2 is reduced in saturation as compared with the original image S0, but the brightness is enhanced.

For this reason, when the constant value of the line is a dark color such as black, the feature amount correcting unit 32 sets a constant value for each of the saturation S and the brightness V so as to emphasize the saturation S and the brightness V. Multiplying k1, k2 (k1, k2> 1) to obtain a corrected saturation S ′ and a corrected brightness V ′. That is,
S ′ = S × k1 (II)
V ′ = V × k2 (III)
The corrected saturation S ′ and the corrected brightness V ′ are obtained by the calculation of

On the contrary, when the constant value of the line is a bright color such as white, the saturation S and the lightness V are reduced so that the saturation S is enhanced and the lightness V is reduced. , K4 (<1) to obtain corrected saturation S ′ and corrected brightness V ′. That is,
S ′ = S × k3 (IV)
V ′ = V × k4 (V)
The corrected saturation S ′ and the corrected brightness V ′ are obtained by the calculation of

  The inverse conversion unit 33 converts the hue H, the corrected saturation S ′, and the corrected brightness V ′ according to the following conversion formula (VI) to obtain image quality-changed image data S1 composed of RGB three-color data. Since the RGB color data obtained by the conversion formula (VI) takes a value of 0 to 1, it is assumed that image quality-changed image data S1 composed of RGB values is obtained by multiplication by 255.

(1) If S = 0, R = G = B = V ′
(2) When not S = 0 (2-1) H0 = H / 60
(2-2) f = Floor (H0) f: Decimal part of H (2-3) i = Int (H0) i: Integer part of H (2-4) p1 = V ′ × (1-S ′)
p2 = V ′ × (1−S ′ × f)
p3 = V ′ × (1−S × (1−f))
(2-5) When i = 0: R = V ′, G = p3, B = p1
When i = 1: R = p2, G = V ′, B = p1
When i = 2: R = p1, G = V ′, B = p3
When i = 3: R = p1, G = p2, B = V ′
When i = 4: R = p3, G = p1, B = V ′
When i = 5: R = V ′, G = p1, B = p2 (VI)
The compression unit 5 compresses the corrected image data S2 obtained by the correction unit 4 using a predetermined compression algorithm. The predetermined algorithm is an algorithm that performs compression processing in consideration of redundancy in a horizontal line in the corrected image S2 without performing preprocessing such as special filtering processing on the corrected image data S2. . Here, since the line with the corrected pixel value has a constant value, the amount of information can be greatly reduced by compression.

  Next, processing performed in the first embodiment will be described. FIG. 3 is a flowchart showing processing performed in the first embodiment. First, when the input unit 2 receives the input of the original image data S0 (step S1), the image quality changing unit 3 changes the image quality of the original image data S0 to obtain the image quality changed image data S1 (step S2). Next, the correcting unit 4 corrects the horizontal lines in the image quality changed image S1 to a constant value for each line to obtain corrected image data S2 (step S3). Then, the compressed image data S2 is compressed by the compression unit 5 to obtain the compressed image data S3 (step S4), and the process ends.

  As described above, in the first embodiment, since all pixel values are corrected to a constant value for each line, the compression rate is improved. On the other hand, when all the pixel values of a predetermined number of lines are corrected to a constant value, the information of the line is lost and the image quality is changed before and after the correction. Since the process for changing the image quality is performed on the original image data S0 before correction, it is possible to prevent the image quality of the corrected image S2 from being changed. Therefore, the compression rate can be improved while preventing the image quality from changing.

  Here, according to the experiment of the present applicant, the data capacity when the corrected image data S2 obtained by the first embodiment is compressed is about 60% of the data capacity when the original image data S0 before correction is compressed. It was confirmed that

  Further, in the original image S0 as shown in FIG. 4A, when the line for each line is corrected so as to have a constant black value, the corrected image S2 has a saturation as shown in FIG. 4B. And brightness is reduced. On the other hand, the corrected image S2 obtained by correcting the image quality changed image S1 whose image quality has been changed so as to emphasize the saturation and lightness of the image S0 as in the first embodiment is shown in FIG. It can be seen that the saturation and lightness can be prevented from changing as compared with the original image S0.

  In the first embodiment, the fixed value is a black or white achromatic color, but may be a chromatic color. Even in this case, the image quality changed image S1 may be obtained by changing the image quality of the original image S0 by correcting the saturation and lightness of the original image S0 in accordance with a certain value.

  In the first embodiment, both brightness and saturation are corrected. However, only one of brightness and saturation may be corrected.

  Next, a second embodiment of the present invention will be described. FIG. 5 is a schematic block diagram showing a configuration of an image compression system 1 ′ having an image processing apparatus according to the second embodiment of the present invention. In the second embodiment, the same components as those in the first embodiment are denoted by the same reference numerals, and detailed description thereof is omitted here. The second embodiment is different from the first embodiment in that an image processing unit 7 and a parameter determination unit 8 are provided.

Image processing section 7 obtains the modified image data S4 by modifying all pixels of a predetermined number per line of original image S0 represented by the same way the original image data S0 and correction unit 4 to a constant value.

  The parameter determination unit 8 calculates a ratio C0 (= VM0 / VM4) of the lightness average value VM4 of the corrected image S4 represented by the corrected image data S4 and the lightness average value VM0 of the original image S0 as the lightness V. Determine the parameters to change. Note that the brightness of the original image S0 and the corrected image S4 may be the maximum RGB value of the original image S0 and the corrected image S4, as shown in the above conversion equation.

In the second embodiment, the feature amount correcting unit 32 of the image quality changing unit 3 sets a constant value k1 (k1> 1) and a ratio C0 for each of the saturation S and the brightness V as shown in the following equations. Multiplication is performed to obtain corrected saturation S ′ and corrected brightness V ′. That is,
S ′ = S × k1 (VII)
V ′ = V × C0 (VIII)
The corrected saturation S ′ and the corrected brightness V ′ are obtained by the calculation of

Next, processing performed in the second embodiment will be described. FIG. 6 is a flowchart showing processing performed in the second embodiment. First, when the input unit 2 receives input of the original image data S0 (step S11), the image processing unit 7 corrects the horizontal line in the original image S0 to a constant value for each line, and the corrected image data S4. Then, the parameter determination unit 8 calculates a ratio C0 between the average brightness value VM4 of the corrected image S4 represented by the corrected image data S4 and the average brightness value VM0 of the original image S0. Then, this is determined as a parameter for changing the lightness V (step S13).

Subsequently, using the parameters determined by the parameter determination unit 8, the image quality changing unit 3 changes the image quality of the original image S0 data to obtain the image quality changed image data S1 (step S14). Next, the correcting unit 4 corrects the horizontal lines in the image quality changed image S1 to a constant value for each line to obtain corrected image data S2 (step S15). Then, the compressed image data S2 is compressed by the compression unit 5 to obtain the compressed image data S3 (step S16), and the process ends.

  Thus, in the second embodiment, the parameter for changing the image quality is determined by comparing the corrected image S4 represented by the corrected image data S4 with the original image S0 before correction. Thus, the parameter can be determined so that the change in the image quality before and after the correction is eliminated, so that the change in the image quality can be surely prevented.

  In the first and second embodiments, the pixel value is corrected for each line. However, even if the pixel value is corrected at an arbitrary line interval, such as every two lines or every three lines. Good.

  Also, by analyzing the original image S0 and considering lines that can improve the compression rate, the position of the line to be corrected to a constant value, for example, a line having a constant value is an odd line or an even line, etc. You may make it decide.

  In the first and second embodiments described above, the feature amount of the original image S0 is extracted by converting the RGB original image data S0 into the HSV space. However, the HLS space other than the HSV space and the YCC luminance are extracted. The feature amount of the original image S0 may be extracted by converting the color difference space.

1 is a schematic block diagram showing a configuration of an image compression system including an image processing apparatus according to a first embodiment of the present invention. The figure for demonstrating correction of the pixel value in 1st Embodiment. The flowchart which shows the process performed in 1st Embodiment. The figure which shows the example of the image before and after correction of a pixel value Schematic block diagram showing a configuration of an image compression system including an image processing apparatus according to a second embodiment of the present invention. The flowchart which shows the process performed in 2nd Embodiment.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1,1 'Image compression system 2 Input part 3 Image quality change part 4 Correction part 5 Compression part 7 Image processing part 8 Parameter determination part 31 Feature-value extraction part 32 Feature-value correction part 33 Inverse conversion part

Claims (6)

Input receiving means for receiving input of original image data representing an original image in which pixels are arranged two-dimensionally;
An image processing apparatus comprising correction means for correcting all pixel values of a predetermined number of lines of the original image to a constant value so that a compression rate is improved when the original image data is compressed by a predetermined compression algorithm. Because
An image processing apparatus, further comprising: an image quality changing unit that performs a process of changing the image quality of the original image according to the fixed value with respect to the original image data before being corrected by the correcting unit.   The image processing apparatus according to claim 1, wherein the image quality changing unit is a unit that performs a process of changing the image quality of the original image by changing lightness and / or saturation of the original image. An image processing means for obtaining corrected image data by performing a process on the original image data to correct all the pixel values of the predetermined number of lines of the original image to the constant value performed by the correcting means;
By comparing the corrected image and the original image represented by the corrected image data, and a parameter determination means for determining a parameter for changing the image quality,
The image processing apparatus according to claim 1, wherein the image quality changing unit is a unit that performs a process of changing the image quality according to the parameter determined by the parameter determining unit.   4. The image processing apparatus according to claim 1, further comprising a compression unit that compresses the corrected image data obtained by the correction unit using the predetermined compression algorithm. Accepts input of original image data representing an original image in which pixels are arranged in a two-dimensional manner,
An image processing method for correcting all pixel values of a predetermined number of lines of the original image to a constant value so that a compression rate is improved when the original image data is compressed by a predetermined compression algorithm,
An image processing method, comprising: subjecting the original image data before correction to a process of changing an image quality of the original image according to the certain value. A procedure for receiving input of original image data representing an original image in which pixels are arranged two-dimensionally;
An image processing method comprising: a step of correcting all pixel values of a predetermined number of lines of the original image to a constant value so that a compression rate is improved when the original image data is compressed by a predetermined compression algorithm. A program for executing
The program further comprising a procedure for performing processing for changing the image quality of the original image according to the fixed value on the original image data before the correction.

Images