JP2006350621A - Image processor and image processing program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To prevent an image of a person from being unnaturally colored in color conversion of a still image, so as to provide unity with the background color of a template. <P>SOLUTION: When image data of still images is read from a storage medium, a background color of template is designated by a user. The designated background color is compared with the color of each still image, and a color conversion quantity is calculated. A subject is detected from each still image, and a skin color area of the person is specified. Color conversion is prohibited in the skin color area, and color conversion processing based on color conversion is applied to a non-skin color area except the skin color area. The color-converted still images are synthesized with the template to form album images. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to an image processing apparatus and an image processing program for performing suitable image editing based on image color information.

  A still image of a subject photographed using a digital still camera equipped with a CCD image sensor is stored as digital image data in a storage medium such as a memory card. When viewing a captured image, for example, if a digital still camera is connected to a home television, the captured image can be displayed on the television screen. Further, the image data stored in the memory card can be transferred to a personal computer, and the photographed image can be displayed on the monitor of the personal computer. When viewing a plurality of photographed images, it is convenient to use a slide show function that displays designated photographed images in a predetermined order. The slide show function is incorporated as part of the image playback function of a digital still camera or a personal computer, and displays a plurality of images by switching them for a predetermined time, for example, every 5 seconds (see, for example, Patent Document 1).

  There are two types of slideshows: a simple one that displays captured images one by one in order, and a slideshow that displays them as an album image in which the captured images are combined with a decorative template image. In the template image, a layout area into which the captured image is inserted is set, and a decorative background area is set outside the layout area. For example, 2 to 4 shot images can be combined with one template image, and the shot images inserted in each layout area are output on the screen as album images (for example, Patent Document 2). reference).

When creating an album image, assigning the shot image to the template image in the order of shooting, etc., mixes images with different subject types, etc., so there is no sense of uniformity in the shooting theme and color tone, and the appearance It may be a bad album image. For this reason, album images that use important commemorative photos such as weddings and entrance ceremonies are selected by selecting images that fit into one template image and performing color tone conversion etc. of the captured image according to the color of the template. Editing work to adjust the color balance of the captured image.
Japanese Patent Laid-Open No. 2005-039359 Japanese Patent No. 2982149

  However, if the color tone of a photographed image in which a person is photographed as a subject is converted, for example, the color of the skin such as the face and arms is unnaturally converted, which deteriorates the appearance of the album image.

  The present invention has been made in consideration of the above-described problems, and provides an image processing apparatus and an image processing program that prevent deterioration of the appearance of an album image when performing tone conversion in accordance with the color of a template. With the goal.

  In order to achieve the above object, an image processing apparatus of the present invention converts a color tone of a still image composed of image data, detects a human subject from the still image, and detects a detected human skin color region. And a color conversion area determining means for determining an area excluding the skin color area from the still image as a color conversion area to be subjected to color tone conversion. It is prohibited to change the color tone of the subject.

  In addition, using a template image in which a layout region for inserting a still image and a background region outside the layout region is set, an image composition unit that creates a composite image by composing a still image with the template image; When a color of the background area is designated, the image processing apparatus includes a color conversion amount calculation unit that calculates a color conversion amount based on a result of comparing the color of the still image and the color of the background area, Tone conversion is performed to bring the color of the still image closer to the color of the template image.

  An image processing program of the present invention is for executing a process of converting the color tone of a still image composed of image data, detects a human subject from the still image, and extracts a detected human skin color region A general-purpose computer, wherein the computer functions as a skin color area extracting means for performing color conversion area determination means for determining an area excluding the skin color area from the still image as a color conversion area to be subjected to color tone conversion. Is used to realize an image processing function.

  According to the present invention, when performing color tone conversion, a human subject is detected, a human skin color region is extracted, and a skin color region is determined from a color conversion target region in which a range for performing still image color conversion is determined. Since the color conversion target area is corrected by excluding the above, color conversion of the person's skin color area is not performed, and the person's skin color can be prevented from being converted to an unnatural color.

  In addition, since the color conversion amount is determined based on the background color of the template image into which the still image is inserted, and color conversion is performed on the region other than the skin color region, a sense of unity can be given to the color tone of the template image and the still image. It is possible to create a good-looking album image.

  In FIG. 1, an album image creation apparatus 10 to which the present invention is applied includes a CPU 13 that functions as a control unit 12 by executing an album creation program PG1 stored in a ROM 11. An image data reading unit 14, an image display unit 15, and an operation panel 16 are connected to the CPU 13. The image data reading unit 14 includes a connector for reading image data from a memory card and a disk drive for reading image data from an optical disk such as a CD. The image display unit 15 includes a liquid crystal display, and displays a still image read as image data, an album image created using the still image, and the like. The operation panel 16 detects a user input operation required when creating an album image and sends an operation signal to the CPU 13.

  The ROM 11 stores an album creation program PG1, template data TD1 used for album creation, and the like. The RAM 17 stores image data PD1 read from the image data reading unit 14, album image data AD1 created by the CPU 13, and the like. The CPU 13 executes the album creation program PG1, and as a person detection unit 20, a skin color area extraction unit 21, a representative color detection unit 22, a color comparison calculation unit 23, a color conversion unit 24, and an album image composition unit 25 Function.

  In FIG. 2, the person detection unit 20 divides a still image into a plurality of regions, and estimates a skin color pixel estimated to be human skin from the plurality of divided regions A1 and a pupil surrounded by the skin color pixels. A divided area A1 including pixels of the color to be selected is selected. Further, based on the selected divided area A1, the pupil detection area A2 for one person is specified from the position of a pair of pupils. In FIG. 3, the skin color area extraction unit 21 extracts a face outline based on skin color pixels distributed around the pupil specified from the pupil detection area A <b> 2, and specifies a face area. Furthermore, based on the skin color of the face, the contours of human hands, feet, etc. composed of pixels of similar colors are extracted. As a result, the skin color area of the person is extracted.

  The representative color detection unit 22 counts the number of pixels belonging to a specific gradation value for each color component of R, G, and B based on the luminance information for each pixel of the image data PD1 and the color information of R, G, and B. A frequency distribution, that is, a histogram for each color component is created, and a representative color having a representative value of the histogram as a color component value is determined. The representative value of the histogram includes an average value, a median value, a mode value, etc., and any one of them is used. Furthermore, a value obtained by multiplying any one value by a correction parameter corresponding to the standard deviation of the frequency distribution or the like may be used as the representative value.

  The color difference calculator 23 compares the representative color of the still image with the color designated as the background color of the template image, and calculates a color difference indicating the difference between the representative color and the background color. The color conversion unit 24 obtains a color conversion amount from the calculated color comparison difference, and performs color conversion of the still image based on the color conversion amount. The color conversion amount is a value that determines how much the R, G, and B color components of the still image are increased or decreased. In addition, the color conversion unit 24 determines a color conversion target region indicating a region to be subjected to color conversion. When determining the color conversion target area, if no person is present in the still image, the entire still image is determined as the color conversion target area. When a person is present in the still image, an area excluding the skin color area R3 from the entire still image is determined as the color conversion target area.

  The album image composition unit 25 composes a still image with the template image T1 to create an album image. In FIG. 4, the template image T1 is preset with a plurality of layout areas LA1, LA2, LA3 for arranging still images and a background area BA1 outside thereof. The background area BA1 is an area that becomes a background of a still image inserted in each layout area, and the background color is designated by the user as a template color.

  Next, the operation of the album image creating apparatus 10 will be described. In FIG. 5, when a memory card or an optical disk is set in the image data reading unit 14, the control unit 12 detects this and reads out the stored still image data PD1. The read image data PD1 is copied to the RAM 17. For example, the control unit 12 creates thumbnail images of the read still images and displays a list of a plurality of still images as thumbnail images on the screen of the image display unit 15.

  The user selects a still image to be used for creating an album image with reference to the thumbnail images displayed in a list. When selection of the still image is completed or use of all the still images is selected, the representative color detecting unit 22 represents the representative value obtained by creating a histogram of each still image as a color component. Detect color. When the representative color of each still image is detected, the background color of the template image T1 is designated by the user. When specifying the background color, for example, a favorite color can be selected from a 256 color palette displayed on the image display unit 15.

  As shown in FIG. 6, a representative color (R, G, B) = (120, 160, 50) is detected from the still image P1, and (R, G, B) = (70, 115) from the still image P2. , 45) and the representative color (R, G, B) = (90, 175, 100) is detected from the still image P3. Further, it is assumed that the user selects a bright yellow-green color of (R, G, B) = (120, 250, 50) as the background color of the template image T1.

  The color difference calculation unit 23 compares the background color designated by the user with the representative color of each still image, and obtains the color difference. In the case of the still image P1, the color comparison difference is (ΔR, ΔG, ΔB) = (0, −90, 0). In the case of the still image P2, the color comparison difference is (ΔR, ΔG, ΔB) = (− 50, −135, −5). In the case of the still image P3, the color difference is (ΔR, ΔG, ΔB) = (− 3, +25, +35). When the color difference of each still image is obtained, the color conversion amount is determined based on this.

  The color conversion unit 24 determines the color conversion amount of each still image from the color comparison difference. The tone correction value is, for example, −10 gradation when the color range difference is +20 to +40, −20 gradation when the color range difference is +41 or more, and +10 gradation when the color range difference is −20 to −40. The case where the difference is −41 or less is predetermined as +20 gradations. In any case, the value is determined so that the representative color of each still image approaches the background color of the template image T1. The color conversion amount for the still image P1 is +20 gradations for the gradation level of the G component. The tone correction value of the still image P2 has an R component of −20 gradation and a G component of −20 gradation. The tone correction value of the still image P3 has −10 gradations for the G component and −10 gradations for the B component.

  When the color conversion amount is determined, the color conversion processing of each still image is performed based on this. When color conversion processing is performed, the person detection unit 20 detects a person appearing in a still image and identifies the pupil detection region R2. Further, a skin color region R3 including a face region and a limb region is specified from the pupil detection region R2. The color conversion unit 24 corrects the color conversion target area to an area excluding the skin color area R3 from the entire still image. That is, the color conversion target region is limited to a non-skin color region different from the skin color region R3, and the color conversion unit 24 performs color conversion on this non-skin color region.

  When the color conversion processing for all the still images is completed, the album image composition unit 25 reads the template data TD1 from the RAM 17 and composes the color-converted still image and the template image T1 with the specified background color. In the template image T1, a plurality of types having different numbers or shapes of layout areas LA1, LA2, LA3 are prepared. The album image synthesizing unit 25 randomly selects one template image T1 from a plurality of types of template images T1, and synthesizes a still image with the template image T1. The album image composition unit 25 performs resolution adjustment and trimming of still images in accordance with the sizes and shapes of the layout areas LA1, LA2, and LA3.

  The image display unit 15 displays an album image in which the color of the template image T1 and the color of the still image are harmonized. The user confirms the album image and selects whether to save it. The created album image data is stored in the RAM 17 as album image data AD1. The album image data AD1 stored in the RAM 17 can be displayed as an album image on the image display unit 15 by operating the operation panel 16. Note that the album image creating apparatus 10 is preferably configured so that the album image data AD1 can be copied to a removable storage medium or the like, or an interface for transferring the album image data AD1 to a personal computer or the like may be provided.

  As described above, the album image creating apparatus 10 has been described. However, the present invention is not limited to the above-described embodiment, and appropriate modifications can be made. For example, an album image can be created using a digital still camera incorporating the image processing apparatus of the present invention or a personal computer incorporating the album creation program PG1.

  When creating an album image, a color image obtained by scanning a photographic film and an image obtained by scanning a color original can be used in addition to an image taken by a digital still camera. In addition, the present invention is not limited to use for the purpose of creating an album image in which a plurality of still images are combined into one template image. When performing a slide show that displays still images one by one, the color tones of still images are increased. Color conversion for giving a sense of unity may be performed. In addition, the present invention is not limited to an album image in which a plurality of still images are combined with a template image. For example, there is no template, and an album image composed of a combined image in which a plurality of still images are arranged in one screen is displayed. It may be created. Further, the created album image is not limited to being displayed as a slide show on a screen of a monitor or the like, and may be output as a single printed photograph.

It is a block diagram which shows the schematic structure of an album image production apparatus. It is a conceptual diagram which shows a pupil detection area | region. It is a conceptual diagram which shows a skin color area | region. It is explanatory drawing of a template image. It is a flowchart which shows the flow of a process of an album image creation apparatus. It is explanatory drawing which shows the flow of the procedure at the time of performing the color tone correction of a still image.

Explanation of symbols

10 Album Image Creation Device 12 Control Unit 13 CPU
20 Human detection unit 21 Skin color region extraction unit 22 Representative color detection unit 23 Color comparison calculation unit 24 Color conversion unit 25 Album image composition unit LA1, LA2, LA3 Layout region BA1 Background region PG1 Album creation program PD1 Image data AD1 Album image data T1 Template image

Claims (3)

In an image processing apparatus that converts the color tone of a still image composed of image data,
A skin color area extracting means for detecting a human subject from the still image and extracting the detected human skin color area;
An image processing apparatus comprising: a color conversion area determination unit that determines an area excluding the skin color area from the still image as a color conversion area to be subjected to color tone conversion. An image composition unit for creating a composite image by combining the still image with the template image using a template image in which a layout region for fitting the still image and a background region outside the layout region is set;
A color conversion amount calculation unit configured to calculate a color conversion amount based on a result of comparing the color of the still image and the color of the background region when the color of the background region is designated; The image processing apparatus according to claim 1. In an image processing program for executing processing for converting the color tone of a still image composed of image data,
A skin color area extracting means for detecting a human subject from the still image and extracting the detected human skin color area;
An image processing program that causes a computer to function as color conversion area determination means that determines an area obtained by excluding the skin color area from the still image as a color conversion area to be subjected to color tone conversion.

Images