JP2008217479A - Image arrangement method and device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To enable even a person who does not particularly have any technology to easily prepare a collage picture by arranging a plurality of images as he or she wants. <P>SOLUTION: This image arrangement method includes a process for selecting a template from a plurality of templates having a plurality of areas where images arranged, and where the attributes of the image arranged therein are designated for every area; a process for selecting the image of a person being a target from among the plurality of images stored in an image database; a process for extracting the selected image on which the person being the target is projected from the image database; and a process for automatically arranging the extracted image in every area according to the attributes designated for every area of the template. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to an image arrangement method and apparatus, and in particular, selects a user-designated image from an image database and automatically creates a collage image by arranging the selected image in a predetermined area in a template. The present invention relates to an image arrangement method and apparatus.

  In recent years, digital cameras have been remarkably widespread, and it has become possible to easily perform various editing by taking an image (digital image data) taken with a digital camera or the like into a computer.

  For example, an image area of an appropriate size is cut out from multiple images, and the cut image is enlarged, reduced, or rotated and pasted into a single image to create a new image. To be done. In addition, for example, a face image that is an image of a face portion of a person image is combined with a background image or clothes image, and a face is drawn out from the cut out face portion with a standing sign of a character that is often found in sightseeing spots. It is also easy to create a composite image such as a photograph taken.

In addition, a certain area is selected from a plurality of areas for displaying images, and an image to be displayed in the area is selected from a video (moving image) or a series of still images, and a specific part of the selected image is selected. There is known an image arrangement method in which an image is enlarged or slightly brightened to be edited and displayed in the area (see, for example, Patent Document 1).
Japanese Patent Laid-Open No. 2003-324683

  However, in the image arrangement method described in the above-mentioned patent document, there are many operations by an operator, and there is a problem that it takes time to create a composite image having an image arrangement that satisfies the operator. In addition, for example, when creating a so-called collage image in which a plurality of images of the same person are arranged and combined into a single image, the operation is complicated, and a general person without technology creates such a collage photo. It was difficult.

  The present invention has been made in view of such circumstances, and even an unskilled person can arrange a plurality of images as desired and can easily create a collage photo. It is an object to provide a method and apparatus.

  In order to achieve the object, the invention according to claim 1 is a plurality of templates each having an area in which an image is arranged, and a plurality of areas in which attributes of the image arranged therein are designated for each area. Selecting a template from among the plurality of images, selecting a target person image from a plurality of images stored in an image database, and displaying an image of the selected target person in the image An image placement method comprising: extracting from the database; and placing the extracted image in each area automatically in accordance with an attribute designated for each area of the template. provide.

  As a result, even a non-technical user can easily create a composite image (collage image) in which favorite images are arranged simply by designating a template and a target person image.

  According to a second aspect of the present invention, the attribute is any one or more of the number of faces, the face orientation, the face size, and the image orientation.

  As described above, by specifying the attribute of the image to be arranged in each area of the template in detail, it is possible to create a composite image that satisfies the user. Here, the orientation of the image means whether it is a vertical position or a horizontal position.

  According to a third aspect of the present invention, in the image arrangement method according to the first or second aspect, a plurality of images that match the attribute specified for each area are further extracted for one area. A step of designating a priority of an image to be arranged in the area among the plurality of extracted images.

  As described above, even when there are a plurality of candidate images to be arranged in each area, it is possible to arrange the images completely automatically by specifying the priority.

  Moreover, as shown in claim 4, the image placement method according to any one of claims 1 to 3, wherein a plurality of images that match an attribute designated for the area of the template are further extracted. In the case where one of the images is arranged in the area, there is a step of changing from the image arranged in the area to another image out of the plurality of extracted images.

  If the automatically generated composite image does not like the user, the image can be changed, so that a composite image satisfying the user can be generated.

  Moreover, as shown in claim 5, the image placement method according to any one of claims 1 to 4, further comprising a step of adding or changing an attribute specified in each area of the template. It is characterized by having.

  Thus, by enabling addition and change of attributes, it is possible to create a composite image that satisfies the user.

  Similarly, in order to achieve the object, the invention according to claim 6 is an area where an image is arranged, and a plurality of areas in which attributes of the image arranged there are designated for each area. A template selection unit that selects a template from a plurality of templates, a target selection unit that selects a target person image from a plurality of images stored in the image database, and the selected target person Image extracting means for extracting an image in which the image appears in the image database; and image placement means for automatically placing the extracted image in each area according to the attribute designated for each area of the template; An image arrangement apparatus characterized by comprising:

  As a result, even a non-technical user can easily create a composite image (collage image) in which favorite images are arranged simply by designating a template and a target person image.

  According to a seventh aspect of the present invention, the attribute is any one or more of the number of faces, the face orientation, the face size, and the image orientation.

  As described above, by specifying the attribute of the image to be arranged in each area of the template in detail, it is possible to create a composite image that satisfies the user.

  According to an eighth aspect of the present invention, in the image arrangement device according to the sixth or seventh aspect, a plurality of images that match the attribute specified for each area are further extracted for one area. In this case, there is provided a priority designating unit for designating a priority of an image to be arranged in the area among the plurality of extracted images.

  As described above, even when there are a plurality of candidate images to be arranged in each area, it is possible to arrange the images completely automatically by specifying the priority.

  Moreover, as shown in claim 9, the image placement device according to any one of claims 6 to 8, wherein a plurality of images that match an attribute specified for the area of the template are further extracted. And when one of the images is arranged in the area, an image changing means is provided for instructing a change from the image arranged in the area to the other extracted image. Features.

  By making it possible to change an image when the user does not like the automatically generated composite image in this way, it is possible to create a composite image that satisfies the user.

  The image layout device according to any one of claims 6 to 9, further comprising an attribute instruction for adding or changing an attribute specified in each area of the template. Means are provided.

  Thus, by enabling addition and change of attributes, it is possible to create a composite image that satisfies the user.

  As described above, according to the present invention, even a non-technical user can easily combine desired images by simply designating a template and a target person image (collage image). Can be created.

  Hereinafter, an image arrangement method and apparatus according to the present invention will be described in detail with reference to the accompanying drawings.

  FIG. 1 is a system configuration diagram showing an outline of an embodiment of an image arrangement apparatus according to the present invention.

  As shown in FIG. 1, the image arrangement apparatus 1 according to the present embodiment includes a computer 10 having a template storage memory 12, an image database 14, an image extraction unit 16, an image arrangement unit 18, and the like, and various inputs and instructions to the computer 10. The input unit 20 such as a keyboard and a mouse that performs the functions of the template selection unit 22, the attribute designation unit 23, the target selection unit 24, the priority designation unit 26, and the image change instruction unit 28, and the display unit such as a liquid crystal display 30 is mainly provided.

  The template storage memory 12 stores a plurality of templates. This template has a plurality of areas in which images (photos) are arranged (pasted), and a composite image (collage image) is created by arranging appropriate images in each area. Yes.

  FIG. 2 shows an example of a template. The template 40 shown in FIG. 2 has five rectangular areas 42 having substantially the same size as regions for arranging images.

  In each area of the template, attribute information (attribute) instructing what kind of image is arranged is designated in advance. The attribute information of the image is not particularly limited. For example, the attributes obtained by image analysis include the number of faces (number of people in the image), the orientation of the face (left / right / up / down or in the image plane). Rotation), the size of the face, etc., or the distinction between landscapes and people, landscapes, especially blue sky, clouds (cloudy), indoors, underwater, etc.

  Attributes that are artificially specified by the user by tagging include event names such as athletic meet, travel, birthday, and sports, other portraits, full body photos, sunset (sunset), user favorite ranks, or person names And so on.

  Examples of attributes that are instructed to the template as attributes when placing these images include the number of people in the image, face orientation, photo orientation (vertical or horizontal position), and trimming information (face up Etc.), event names such as athletic meet, favorite rank, etc.

  Such attribute information is designated in advance for each area of the template and attached to the template. In the template 40 shown in FIG. 2, the face orientation and the number of persons are designated as attributes of each area 42.

  The attributes are not limited to those specified in advance in the template as described above. The user selects a template, displays the selected template on the display means 30, and views each display area 30 while viewing the display means 30. May be added to or changed. The attribute is added or changed by the user through the attribute instruction means 23. For example, the user displays a template 40 as shown in FIG. 2 on the display means 30, designates an area 42 whose attribute information is to be changed, and inputs an attribute to be added or changed via the attribute instruction means 23. To do.

  The image database 14 mainly stores images taken by the user so far as digital image data. The user stores images in the image database 14 in advance. The image placement apparatus 1 analyzes the image in the image database 14 to acquire various attributes such as the face position, the face direction, the number of people in the image, and classifies the image for each attribute. .

  When creating a collage photo, the user first selects one image in the image database 14 via the target selection unit 24, for example, his own photo. The user calls a target image from the image database 14 by inputting an appropriate selection condition via the target selection means 24, displays it on the display means 30, and selects the target image while viewing it.

  It is not always necessary to select the target from the image database 14 as described above. The user may separately input a target image using a medium or the like.

  The image extraction unit 16 is an image in which a target person is captured from the image database 14 based on a template selected by the user using the template selection unit 22 and a target (person) selected using the target selection unit 24. Thus, an image suitable for the attribute designated for each area of the template is extracted.

  In order to extract an image that meets such conditions, the image extraction means 16 performs extremely advanced functions such as face detection, face direction calculation, determination of the same person (similarity calculation), and classification of shooting scenes as necessary. Image analysis processing is performed. Existing techniques are applied to this image analysis processing.

  The image arrangement unit 18 arranges the image extracted by the image extraction unit 16 in each area according to the attribute of each area. At this time, if a plurality of images that match the attribute specified for each area are extracted for each area, the image with the highest priority is first arranged according to the preset priority. This priority may be set in advance, or may be set by the user via the priority specifying means 26 each time. The priority is not particularly limited. For example, when there is an image of the same face that matches the attribute, priority is given to a face with a larger face, or a new one at the time of shooting. It is conceivable to specify such as placing them on the top of each other or preferentially placing the ones with bright lightness.

  In this way, a composite image in which an image is arranged in each area of the template (a template in which an image is arranged in each area) is displayed on the display unit 30. If the user looks at this display and does not like it, the user changes the image he / she does not like through the image change instruction means 28.

  When there are a plurality of extracted images (candidate images) that match the attribute of the area, other candidate images for the area are displayed by giving an instruction from the input means 20 as the image change instruction means 28. This display method is not particularly limited, and various methods are conceivable. The thumbnail images may be displayed together at the corners of the display screen of the display means 30, or the candidate images may be switched and displayed one by one by clicking on the area in the template. The user can replace the image in the area by designating a favorite image from the candidate images.

  A printer is preferably connected to the computer 10 in order to print out the composite image (collage image) created in this way.

  Next, a method for creating a composite image (collage image), which is an operation of the present embodiment, will be described with reference to the flowchart of FIG.

  Before starting the processing shown in the flowchart of FIG. 3, the user stores images in the image database 14 and inputs various types of templates in the template storage memory 12. At that time, attribute information for each area is also specified in the template.

  First, in step S100 of FIG. 3, the user selects a template. For this purpose, the user issues an instruction to the computer 10 via the input means 20 such as a keyboard as the template selection means 22, calls a template from the template storage memory 12, and displays it on the display means 30. The user sees the display and selects a template. For example, assume that a template 40 as shown in FIG. 2 is selected.

  The template 40 has five areas 42, and the central area 3 is designated with the attribute that the face direction is front and the number of people is one. Similarly, in the upper left area 1, the face direction is lower left and the number of people is 1, the upper right area 2 is the face direction of the lower right and the number of people is 1, and the lower left area 4 is the face direction of the upper left and the number of people. The attribute of the image to be arranged for each area 42 is designated such that one person is in the lower right area 5 and the face direction is in the upper right and the number of people is one person.

  Here, when changing the attribute of each area 42 of the selected template 40, the user inputs an attribute information change instruction to the computer 10 via the input means 20 such as a keyboard as the attribute instruction means 23. . Users can also specify attributes by tagging. When the attributes in the image database 14 are pre-assigned, an event name such as an athletic meet is designated, and a favorite rank indicating the degree of favorite of the image (for example, one star or two stars) ) Or the name of the person in the picture.

  Next, in step S110, a target is selected. For example, when a collage image is to be created by arranging various images of one person in one template, the image of the person is selected. The user issues an instruction from the input unit 20 as the target selection unit 24 and selects a target image (person) from the images in the image database 14. At this time, the user may input a separate image to the computer 10 instead of selecting a target from the image database 14.

  In particular, when a person name is registered as an attribute, the target can be selected by designating the target person name.

  Next, in step S120, the image extracting means 16 is an image including a target from the image database 14 based on the attribute information and target (image) specified in the template 40 specified by the user above, and each of the above areas. Images that match the attributes specified for each 42 are extracted in correspondence with each area 42.

  The image extraction means 16 first extracts the face of a target person from the target image. A subject person's specific target, that is, a face, eyes, pupils, nose, mouth, and the like are detected from the target image, and their feature points and patterns are identified to detect the face of the subject person. A specific method of face detection is a face detection method by edge detection or shape pattern detection, a feature point vector approximation method by vectorizing feature points of each specific target and approximating feature point vectors, hue detection or skin color detection A known method such as a face detection method by the method can be used. Many face detection techniques are known. For example, as a device for recognizing the shape of a recognition object shown in an image, a device that recognizes the shape of a human eye pupil in particular. This is described in JP-A-9-73544.

  Similarly, the image extraction means 16 detects the face of the subject person from each image in the image database 14. Then, an image showing the same person as the person in the target image is selected. Various methods are known as methods for determining whether or not a subject person in two images is the same person. For example, as described in Japanese Patent Application Laid-Open No. 2006-31388, the similarity between two images is calculated based on the correlation value of the region values of the partial regions in the feature region set in common for the two images. By doing so, the same person may be determined.

  In the case of the template 40 as shown in FIG. 2, for example, the face direction is specified as an attribute in each area 42. In order to extract an image suitable for the attribute, the face direction must be detected. Don't be.

  Various methods for detecting the orientation of the face are also known. For example, in Japanese Patent Laid-Open No. 2001-291108, a face area is extracted from image data, and the input face image data is obtained by normalizing the size of the face area image. Mosaic processing is performed to obtain low-frequency input face image data. This low-frequency input face image and the three-dimensional face model are rotated by a predetermined angle step, and average face image data by angle obtained by projecting from a fixed direction in each step. The direction of the face image is discriminated by comparing the data of each angle and determining the image data having the highest degree of correlation.

  Further, for example, in Japanese Patent Application Laid-Open No. 2004-297274, image data is analyzed to automatically extract a person region and a face part from the image, and the face orientation and the like are calculated from the extracted face part information. I have to. Specifically, the direction of the face is calculated by obtaining the difference between the left and right of the distance between the center coordinates of the eyes and the coordinates of the end points of the face contour.

  In addition to the face facing left, right, up and down, this face may be rotated in the image (photo) plane that faces the front and tilts the face to the side. Similarly, the inclination of can be detected.

  In addition, it may be necessary to detect a shooting scene of an image in order to match the attribute specified in the template area. For example, when a shooting scene (image classification) such as “person face”, “underwater”, “high saturation”, “evening scene”, “night scene”, “blue sky” is specified as an attribute, an image having the attribute Need to be detected. In such a case, for example, a technique as described in JP-A-2006-236181 can be used.

  In this way, the image extracting means 16 extracts an image (candidate image) that has the attribute designated in each area 42 of the template 40 and that is an image showing the target person selected by the user. When the trimming information is specified as the attribute information, not the entire image itself but a face image obtained by trimming the face of the target person is extracted.

  Next, in step S130, the image arrangement unit 18 arranges the image extracted above in each area 42 according to the attribute. Here, when there are a plurality of images (candidate images) to be arranged in one area 42, a plurality of candidate images having a high priority are arranged according to the separately designated priority. . Note that the priority may be designated on the spot via the priority designation means 26.

  In this way, an image suitable for each attribute is arranged in each area 42 of the template 40 and displayed on the display means 30. FIG. 4 shows an example of the template 40 in which an image is arranged in each area 42 in this way and a composite image is formed. The composite image shown in FIG. 4 is a collage image in which a face facing the front is arranged in a central area and a face facing the center is arranged around the face.

  Next, in step S140, the user looks at this display, and if there is something that he does not like, the image is changed (reselected) in step S150.

  For example, on the display screen of the display means 30, the user clicks the area 42 of the image that the user wants to change via the input means 20 as the image change instruction means 28, and displays other candidate images. This is done by selecting a new image from.

  In this way, a collage image that satisfies the user is created. In the simplest case, the user simply selects a template and a target image, and a collage image is automatically created.

  The created collage image may be stored in a memory or printed out from a printer (not shown).

  As described above, according to the present embodiment, even a non-technical user can easily create a collage image by arranging a plurality of images as desired.

  Although the image arrangement method and apparatus of the present invention have been described in detail above, the present invention is not limited to the above examples, and various improvements and modifications may be made without departing from the spirit of the present invention. Of course.

1 is a system configuration diagram showing an outline of an embodiment of an image arrangement device according to the present invention. It is explanatory drawing which shows an example of a template. It is a flowchart which shows the preparation method of the synthesized image which is an effect | action of this embodiment. It is explanatory drawing which shows an example of the produced synthesized image.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 ... Image arrangement | positioning apparatus, 10 ... Computer, 12 ... Template storage memory, 14 ... Image database, 16 ... Image extraction means, 18 ... Image arrangement means, 20 ... Input means, 22 ... Template selection means, 23 ... Attribute instruction means, 24 ... target selection means, 26 ... priority designation means, 28 ... image change instruction means, 30 ... display means, 40 ... template, 42 ... area

Claims (10)

A step of selecting a template from a plurality of templates having a plurality of areas in which images are arranged and the attributes of the images arranged there are designated for each area;
Selecting a target person image from a plurality of images stored in an image database;
Extracting an image of the selected target person from the image database;
Automatically placing the extracted images in the areas according to the attributes specified for each area of the template;
A method for arranging images.   The image placement method according to claim 1, wherein the attribute is one or more of the number of faces, face orientation, face size, and image orientation.   3. The image arrangement method according to claim 1, further comprising: extracting a plurality of extracted images when a plurality of images that match an attribute specified for each area are extracted for one area. An image arrangement method comprising a step of designating a priority of an image to be arranged in the area.   The image arrangement method according to any one of claims 1 to 3, further comprising: extracting a plurality of images that match an attribute designated for the area of the template, and one of the images is the area. An image arrangement method comprising: changing the image arranged in the area to another image out of the plurality of extracted images.   5. The image arrangement method according to claim 1, further comprising a step of adding or changing an attribute specified for each area of the template. 6. A template selecting means for selecting a template from a plurality of templates having a plurality of areas in which images are arranged and the attributes of the images arranged there are designated for each area;
Target selection means for selecting a target person image from a plurality of images stored in an image database;
Image extracting means for extracting an image of the selected target person from the image database;
Image placement means for automatically placing the extracted images in the areas according to the attributes designated for each area of the template;
An image arrangement apparatus comprising:   The image placement apparatus according to claim 6, wherein the attribute is at least one of the number of faces, the face orientation, the face size, and the image orientation.   8. The image arrangement device according to claim 6 or 7, further comprising: extracting a plurality of extracted images when a plurality of images matching an attribute designated for each area are extracted for one area. An image placement apparatus comprising priority designation means for designating a priority of an image placed in the area.   9. The image placement device according to claim 6, further comprising: extracting a plurality of images that match an attribute specified for the area of the template, and one of the images is the area. And an image changing means for instructing a change from the image arranged in the area to another image extracted from the images arranged in the area.   10. The image placement device according to claim 6, further comprising an attribute instruction unit that adds or changes an attribute designated in each area of the template. 11. Placement device.

Images