KR20170093775A - Group photo amednding system - Google Patents

Abstract

Provided are a group portrait correction system and a group portrait correction method. The group portrait correction method according to an aspect of the present invention comprises the steps of: recognizing a subject from a plurality of photographs and forming a subject contour line; selecting a main photograph to be used as a base for correction; forming a person contour line by combining objects constituting a person among objects in the photographs; comparing people in the photographs to determine the same people; and positioning objects of a person to be synthesized in a background area of the main photograph.

Description

GROUP PHOTO AMEDNDING SYSTEM}

The present invention relates to a group portrait correction system for correcting a photograph including a plurality of portrait images.

Photography has been used for a long time as a medium to remember past times. In recent years, digital cameras and the like have become popular, and they are not stored on paper but stored in digital form or uploaded to a web page.

In general, an image acquired through a digital camera can be corrected or synthesized in various ways and applied to the modification. The image can be corrected using commercial programs such as Photoshop and Illustrator.

When taking a ceremony or traveling, usually take commemorative photos. It is very difficult to recapture and recapture everyone when a person is missing because the commemorative photographs involve many people.

After shooting only missing persons, they can be synthesized into a group portrait using a commercial program, but in this case, there is a problem that requires a lot of time and a high level of skill.

SUMMARY OF THE INVENTION The present invention has been made to solve the above problems, and it is an object of the present invention to provide a group portrait correction system and a group portrait correction method which can easily correct a group portrait.

A group portrait correction system according to an aspect of the present invention includes a subject contour generation unit for recognizing a subject in a plurality of photographs and forming a subject contour, a main photograph selection unit for selecting a main photograph to be used as a base of correction, A character outline generation unit for forming a character outline by combining objects constituting a character among objects displayed in the images, a same person recognition unit for comparing the characters displayed in the photographs to determine the same characters, And a portrait synthesizing unit for positioning the background image in the background area of the photograph.

Here, the subject contour generating unit may include a pixel discriminating module for grouping the unit objects included in the images, and a type determining module for matching the shape of the pixel grouped region to a plurality of polygons.

In addition, the pixel discrimination module may select a pixel having a similar color value among the pixels, and select a pixel within a set range of the distance between the pixels.

The pixel discrimination module may define a value of the color value as a degree of color similarity, define a value obtained by dividing the color similarity by the distance between the pixels as a binding force, and group the pixels when the binding force is larger than the predetermined value .

In addition, the group portrait correction system may further include an alignment portion for adjusting the size, brightness, or contrast of the plurality of photographs.

In addition, the aligner may include a reference setting module for selecting a reference object common to all selected pictures.

In addition, the aligner may include a resizing module for comparing sizes of the reference objects to adjust sizes of the pictures so that the sizes of the reference objects indicated by the pictures are identical to each other, and a resizing module for adjusting the sizes of the reference objects, And a contrast adjusting module for adjusting the contrast of the photographs so that the contrast of the reference objects indicated in the photographs coincides with each other.

In addition, the figure outline generation unit may compare the object shape data stored in the recording medium with the objects shown in the photographs, and combine objects to generate a character outline.

In addition, the same person recognizing unit may identify the characters displayed on the main photograph, and identify the same characters among the characters displayed on the main photograph and the photographs other than the main photograph.

Further, the same person recognizing unit can specify a synthesizing target person, which is shown in the seed photograph but is not shown in the main photograph.

In addition, the group portrait correction system may further include a fine adjustment unit for adjusting an object synthesized in the main photo by edge smoothing, exposure adjustment, contrast adjustment, and masking to harmonize the object with the surroundings.

As described above, according to the present invention, it is possible to more easily correct a group portrait by recognizing the characters arranged in the plurality of images and correcting the characters into a single photograph.

FIG. 1 is a block diagram illustrating a system for correcting a group portrait photograph according to an exemplary embodiment of the present invention.
FIG. 2 is a flowchart illustrating a method of correcting a group portrait according to an exemplary embodiment of the present invention. Referring to FIG.
3 is a view showing a main photograph to be corrected.
Fig. 4 is a view showing a species photograph to be corrected.
FIG. 5 is a photograph showing a state after correction is completed. FIG.

The present invention can be variously modified and may have various embodiments, and specific embodiments will be described in detail with reference to the drawings. It should be understood, however, that the invention is not intended to be limited to the particular embodiments, but includes all modifications, equivalents, and alternatives falling within the spirit and scope of the invention.

Terms including ordinals, such as first, second, etc., may be used to describe various elements, but the elements are not limited to these terms. The terms are used only for the purpose of distinguishing one component from another.

For example, without departing from the scope of the present invention, the first component may be referred to as a second component, and similarly, the second component may also be referred to as a first component. And / or < / RTI > includes any combination of a plurality of related listed items or any of a plurality of related listed items.

It is to be understood that when an element is referred to as being "connected" or "connected" to another element, it may be directly connected or connected to the other element, . On the other hand, when an element is referred to as being "directly connected" or "directly connected" to another element, it should be understood that there are no other elements in between.

The terminology used in this application is used only to describe a specific embodiment and is not intended to limit the invention. The singular expressions include plural expressions unless the context clearly dictates otherwise. In the present application, the terms "comprises" or "having" and the like are used to specify that there is a feature, a number, a step, an operation, an element, a component or a combination thereof described in the specification, But do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, or combinations thereof.

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Terms such as those defined in commonly used dictionaries are to be interpreted as having a meaning consistent with the contextual meaning of the related art and are to be interpreted as either ideal or overly formal in the sense of the present application Do not.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art can easily carry out the present invention. The present invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein.

FIG. 1 is a block diagram illustrating a system for correcting a group portrait photograph according to an exemplary embodiment of the present invention.

1, a group portrait correction system 100 according to the present embodiment includes a subject contour generation unit 10, an alignment unit 20, a main picture selection unit 30, a portrait contour generation unit 40, A same person recognizing unit 50, a person synthesizing unit 60, and a fine-tuning unit 70.

When a plurality of photographs are selected, the subject contour generating unit 10 recognizes a subject such as a person and a background from a plurality of photographs and generates a contour line for the subject. The object contour generating unit 10 includes a pixel discriminating module 11 for grouping pixels of a unit object included in images and a type discriminating module 12 for matching a shape of a region of a pixel grouping into a plurality of polygons .

First, the pixel discrimination module 11 digitizes the color values of the pixels existing on the digital image, selects pixels having a similar number among them, and performs grouping of pixels by collecting the distances between the pixels within the set range . Accordingly, in the case of a single person or object, since the image is constituted by a majority of consecutive pixels, it can be recognized as one object.

More specifically, the pixel discrimination module 11 defines the color value as the color similarity. The color discrimination module 11 defines a value obtained by dividing the color similarity by the distance between the pixels as a cohesion force. If the cohesion force is larger than the set value, The aggregation can be further improved by collecting.

The shape determination module 12 performs matching distribution of the shape of the region of the pixel grouping using a plurality of polygons in order to further enhance the accuracy in which the pixels grouped by the pixel determination module 11 are recognized as objects.

The shape determination module 12 forms a rectangle and two consecutive long rectangles in the group of the grouped pixels on the grouped pixels to remove the protruded pixels through the pixel grouping, So that it is closer to the actual object image.

The alignment portion 20 adjusts the size of the plurality of photographs, adjusts the brightness, or adjusts the contrast. The alignment section 20 includes a reference setting module 21 for recognizing the same object shown in the photographs, a resizing module 22 for adjusting the size of the photographs, a brightness adjusting module 23 for adjusting the brightness of the photographs, And a contrast adjustment module 24 for adjusting the contrast.

The reference setting module 21 selects a reference object common to all the selected photographs, and sets a reference object in the background of the photograph and the figure. The resizing module 22 compares the size of the reference objects and adjusts the size of the pictures so that the sizes of the reference objects shown in the photographs are equal. The brightness adjustment module 23 adjusts the brightness of the photographs so that the brightness of the reference objects shown in the photographs coincide with each other. The contrast adjustment module 24 adjusts the contrast of the photographs so that the contrast of the reference objects shown in the photographs coincide.

3 and 4, the main photograph selecting unit 30 compares the sharpness, the exposure, and the white balance of the photographs to select the main photograph 81 to be used as the base of the correction, . The main photograph 81 indicates a photograph to be corrected, and a portrait image displayed only on a bell photograph is pasted on a main photograph to be combined.

The character outline generation unit 40 generates a character outline by combining objects constituting a person among objects displayed in photographs. The person contour generating unit 40 compares the objects shown on the photographs with the person shape data stored in the recording medium, and combines the objects to generate a character contour line. One person includes a face, an arm, a leg, and a shoe, and the subject contour forming unit 10 recognizes portions of different colors as separate objects. The person contour generating unit 40 compares the objects constituting one character by comparing them with the character shape data in order to recognize these objects as one person. The person shape data generally refers to image data that has a human appearance.

The same person recognition unit 50 compares the characters displayed on a plurality of photographs to determine the same characters. The same person recognizing unit 50 specifies the characters displayed on the main photograph 81 and displays the same character 85 and characters 86 and 86 of the characters displayed on the main photograph 81 and the portrait photograph 82, 87). The same person recognition unit 50 first compares the keys of the person to determine whether they are the same person or not. If the keys are similar to each other, it compares the faces to determine whether they are the same person or not. In addition, the same person recognizing section 50 specifies a synthesizing target person 87 which is shown in the seed photograph 82 but not in the main photograph 81. [

5, the person synthesizing unit 60 synthesizes the objects of the synthesizing target object 87 by superimposing them on the main picture 81. The person synthesizing unit 60 synthesizes the objects The object of the synthesizing target person 87 is positioned in the background area which is not positioned. The fine adjustment unit 70 adjusts the object synthesized in the main picture 81 in a manner such as edge smoothing, exposure adjustment, contrast adjustment, masking, etc. to match the object with the surroundings.

As described above, according to the present embodiment, if only the photographs are selected, the characters displayed in the respective photographs can be synthesized to automatically form one group photo.

FIG. 2 is a flowchart illustrating a method of correcting a group portrait according to an exemplary embodiment of the present invention. Referring to FIG.

2, a group portrait photographing method according to the present embodiment includes a subject contour creating step S101, an aligning step S102, a main photograph selecting step S103, a person contour creating step S104, The same person recognition step S105, the person synthesis step S106, and the fine adjustment step S107.

The group portrait correction method according to the present embodiment is performed by the above-described group portrait correction system.

The subject contour generation step S101 is performed by the subject contour generation unit 10, and when a plurality of photographs are selected, the subject generates an outline for a person and a background located in the photograph. The object contour generation step S101 includes a pixel discrimination step of grouping the unit objects included in the images and a shape discrimination step of matching the shape of the area of the pixel grouping to a plurality of polygons.

The pixel discrimination step is performed by the pixel discrimination module 11, and the color values of the pixels existing on the digital image are digitized. After selecting the pixels having the similar numerical values, the distance between the pixels is determined to be within the set range Thereby performing pixel grouping. Accordingly, in the case of a single person or object, since the image is constituted by a majority of consecutive pixels, it can be recognized as one object.

More specifically, the pixel discrimination step defines the numerical value of the color value as a color similarity degree. When a value obtained by dividing the color similarity degree by the distance between pixels is defined as a cohesion force and the object when the cohesion force is larger than the set value is collected And the matching performance can be improved.

The type determination step is performed by the type determination module 12, and in order to further increase the accuracy in which the pixels grouped in the pixel determination step are recognized as an object, the shape of the pixel grouped area is matched and distributed using a plurality of polygons.

The shape determination step forms a rectangle and two consecutive long rectangles on the grouped pixel in the group of the grouped pixels to remove the protruded pixel through the pixel grouping or to fill the pixel area where the part of the groove is continuous on the continuous line So that it is closer to the actual object image.

The aligning step S102 is performed by the aligning unit 20 to adjust the size of the plurality of pictures, adjust the brightness, or adjust the contrast. The aligning step S102 includes a reference setting step of recognizing the same object shown in the photographs, a resizing step of adjusting the size of the photographs, a brightness adjusting step of adjusting the brightness of the photographs, and a contrast adjusting step of adjusting the contrast of the photographs .

The reference setting step selects a reference object which is performed by the reference setting module 21 and is common to all selected pictures, and sets a reference object in the background of the picture and the figure. The resizing step is performed by the resizing module 22 and compares the size of the reference objects to adjust the size of the pictures so that the sizes of the reference objects shown in the photographs are equal. The brightness adjusting step is performed by the brightness adjusting module 23 and adjusts the brightness of the photographs so that the brightness of the reference objects shown in the photographs matches. The contrast adjustment step is performed by the contrast adjustment module 24 and adjusts the contrast of the photographs so that the contrast of the reference objects shown in the photographs coincide.

The main photograph selection step S103 is performed by the main photograph selection unit 30 and compares the sharpness, exposure, and white balance of the photographs to select a main photograph 81 to be used as a basis for correction, ). The main photograph 81 indicates a photograph to be corrected, and a portrait image displayed only on a bell photograph is pasted on a main photograph to be combined.

The person contour generation step S104 is performed by the person contour generation unit 40, compares the objects shown in the pictures with the person shape data stored in the recording medium, and combines the objects to form a character contour line. One person includes a face, an arm, a leg, and a shoe. The object contour forming step recognizes portions of different colors as separate objects. In the figure outline generation step, in order to recognize these objects as one person, it compares the objects constituting one person with the figure shape data. The person shape data generally refers to image data that has a human appearance.

The same person recognition step S105 is performed by the same person recognition unit 50, and the same persons are judged by comparing the characters displayed on the plurality of photographs. In the same person recognition step, the characters displayed on the main photograph 81 are identified and the same character 85 and different characters 86 and 87 among the characters displayed on the main photograph 81 and the characters 82 displayed on the photograph 82 Specify. In the same person recognition step (S105), first, the keys of the person are compared to determine whether they are the same person. If the keys are similar to each other, the faces are compared to determine whether they are the same person or not. The same person recognizing step S105 judges the synthesizing target person 87 which is shown in the seed photograph 82 but not in the main photograph 81. [

The person synthesizing step S 106 is performed by the person synthesizing unit 60 and composites the images of the synthesizing target person 87 by superimposing them on the main picture 81. The person synthesizing unit 60 synthesizes The image of the object person 87 to be synthesized is placed in the background area in which the characters are not located. The fine adjustment step S107 is performed by the fine adjustment unit 70 and adjusts the object synthesized in the main picture 81 in a manner such as edge smoothing, exposure adjustment, contrast adjustment, masking, etc. to match the object with the surroundings.

While the present invention has been described in connection with what is presently considered to be practical exemplary embodiments, it is to be understood that the invention is not limited to the disclosed embodiments, but, on the contrary, Of course.

100: Photo correction system
10: Subject contour forming part
11: Pixel discrimination module
12: type determination module
20: Allied worker
21: Reference setting module
22: Resizing module
23: Brightness adjustment module
24: Contrast adjustment module
30: main picture selection unit
40: Character contour generation unit
50:
60:
70:
81: State photo
82: Bell picture

Claims (1)

A subject contour generation unit for recognizing a subject and forming a subject contour line from a plurality of pictures;
A main picture selection unit for selecting a main picture to be used as a base of correction;
A figure contour generating unit for forming a figure contour by combining objects constituting a character among the objects shown in the photographs;
A same person recognition unit for comparing the characters displayed on the photographs to determine the same characters;
A person synthesizing unit for positioning an object of a person to be synthesized in a background area of the main picture; And
An aligner for adjusting the size, brightness, or contrast of the plurality of pictures;
Lt; / RTI >
The subject contour generating unit may include a pixel discriminating module for grouping pixels of the unit objects included in the images and a type discriminating module for matching the shape of the pixel grouped region into a plurality of polygons
Wherein the type determination module forms a series of long rectangles attached to a rectangle and a rectangle on a grouped pixel set on a grouped pixel so as to project a long rectangle contiguous to the rectangle and the rectangle among the group of the grouped pixels, Removing a pixel having a groove that is contiguous with the long rectangle connected to the rectangle and the rectangle among the group of the grouped pixels,
The aligner includes a reference setting module for selecting a reference object common to all selected pictures
The figure outline generation unit compares the figure data stored in the recording medium with the objects shown in the photographs, combines the objects to generate a figure outline,
The same person recognizing unit specifies the characters displayed on the main photograph, identifies the same person among the characters displayed on the main photograph and the species photographs other than the main photograph, and the same person recognizing unit identifies the same person But does not appear in the main picture, is specified.

Images