KR20090017355A - Apparatus and method for taking pictures after adjusting composition of the pictures automatically - Google Patents

Abstract

An image photographing apparatus and a method thereof are provided to select a photographing point of time so as to automatically photograph an image where an image tool is controlled after automatically controlling the image tool. An image photographing module changes an angle and focus of a camera lens to photograph an image. A face area extraction module(120) detects and traces a face area in a preview image of the photographed image. An image photographing control module(130) performs a control operation so as to change the angle and focus of the camera lens so that an image structure of the traced and defected face area can satisfy a predetermined criteria.

Description

Apparatus and method for automatically adjusting image composition {APPARATUS AND METHOD FOR TAKING PICTURES AFTER ADJUSTING COMPOSITION OF THE PICTURES AUTOMATICALLY}

The present invention relates to an image capturing apparatus and method for automatically adjusting an image composition. In particular, it is useful as an apparatus and method applied to a mobile terminal or a robot.

In a conventional image capturing apparatus, an image is mainly captured by a command of a user. As a more advanced technology that automatically prepares for image capturing, a digital camera automatically detects a face area using a technology of detecting a face area in a preview image and focuses on the face area shown above. There is a technology that automatically recalibrates.

However, these conventional technologies only refocus the image to be photographed, and there is a technology that automatically adjusts the image composition so that the image has a stable composition, or automatically captures the image in the image composition automatically adjusted above. I never do that.

The present invention provides an image photographing apparatus and method for automatically adjusting an image composition that does not exist in the prior art.

In addition, the present invention provides an image capturing apparatus and method for selecting a photographing time point so that the image composition can be automatically photographed after the image composition is automatically adjusted.

In particular, the present invention provides an image photographing apparatus and method for automatically providing an image composition, such that a face of a person can be photographed stably without biasing one side of the image.

On the other hand, such an image recording apparatus and method is applied to a mobile terminal or a robot (Robot) to be particularly useful.

In order to solve the problem to be solved, the present invention can take an image, change the angle of the camera lens, the focus and the zoom (Zoom) function state for taking the image, the preview image for taking the image An image photographing module for providing a face, a face region extraction module for detecting and tracking a face region in the provided preview image, and an angle and a focus of the camera lens such that an image composition of the detected and traced face regions satisfies a predetermined criterion; And an image photographing control module for controlling to change the zoom function state. In this case, in the case of a mobile communication terminal or a robot, the image photographing device may use a wheel, a pan of the robot head, a tilt function, and a wheel leg below the robot. .

Meanwhile, in order to solve the problem to be solved, the present invention detects and tracks a face region in a preview image, and determines whether an image composition of the face region detected and tracked in the preview image satisfies a predetermined criterion. And controlling the camera lens to change an angle, a focus, and a zoom function state of the camera lens until the predetermined criterion is satisfied, when the predetermined criterion is not satisfied. do.

The present invention configured as described above can automatically adjust the image composition of the face, there is an effect that can be automatically taken when the image composition is adjusted.

In particular, people's faces are photographed in a stable manner without biasing one side of the image, and can be particularly useful by being applied to a mobile communication terminal or a moving robot.

In addition, when the image of the children is a lot of movement when it is difficult to compose the image, there is an effect that can easily take the image by adjusting the image composition.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. Note that the same components in the drawings are represented by the same reference numerals and symbols as much as possible even if shown in different drawings. In addition, in the following description of the present invention, if it is determined that a detailed description of related known functions or configurations may unnecessarily obscure the subject matter of the present invention, the detailed description thereof will be omitted.

1 is a configuration diagram of an image photographing apparatus 100 that automatically adjusts an image composition according to an embodiment of the present invention. 1 shows a connection state between the image capturing apparatus 100 that automatically adjusts the image composition and each component. Referring to FIG. 1, the image capturing apparatus 100 includes an image capturing module 110, a face region extraction module 120, an image capturing control module 130, and an image storing module 140. Each configuration is as follows.

The image capturing module 110 is a component for capturing an image and includes a camera 111, a camera driver 112, and an image display unit. Each configuration is as follows.

The camera 111 includes a general configuration such as a lens unit (not shown) and a photosensitive unit (not shown) for capturing an image. In particular, it is assumed that a camera 111 that can provide a preview image before taking an image. For example, it may be a video camera such as a digital still camera or a camcorder provided with a preview image.

The camera driver 112 is configured to automatically adjust an image composition by capturing an object to be photographed. The direction of the camera 111 may be changed by the camera driver 112, and in particular, the direction of the camera lens may be changed or rotated up, down, left, or right, or the focus may be changed by performing a zoom function of the camera lens. For example, when applied to the mobile terminal may be a wheel (Wheel) attached to the camera 111, when applied to the robot head of the robot that can move freely with the camera 111 is attached The part and wheel or leg may correspond to this.

The image display unit is configured to display a preview image provided by the camera 111 so that the user can check it. Of course, it is possible to check the captured image.

The face region extraction module 120 detects a face region in the preview image and tracks the detected face region when it moves. The face region extraction module 120 includes an omega detection unit 122, a face shape detection unit 121, and a face area detection and tracking unit 123. Each configuration is as follows.

The omega detector 122 is configured to detect an omega shape from the preview image. Since the omega shape is similar to the shape of the head and shoulders of a person, when an image is taken with a subject as a person, the object to be photographed can be found by detecting the omega shape, and it is easy to detect the head of the side or the back. The omega detection unit 122 detects information on the size, number, location, and the like of the omega shape. The information about the number is information that can be used when automatically divided into a mode for capturing a group image to be described later and a mode for capturing a single personal image.

The face shape detector 121 is configured to detect a face shape in the preview image. To detect facial shape, there are methods using skin color distribution and methods using Active Contour. Specifically, one method may be to detect the skin color in the preview image to recognize it as a face shape, or to detect a shape that is considered to have features such as eyes, nose, and mouth. The face shape detection unit 121 also detects information about the size, number, and position of the face shape. The information on the number is also information that can be used to distinguish between the mode for photographing a group image and the mode for photographing a single individual image.

However, only one of the above-described omega detection unit 122 and the face shape detection unit 121 may not be regarded as a shape of a person. When the output values of the above two elements are combined, there is a high probability of the shape of a person. Therefore, the face area detection and tracking unit 123 combines the omega shape detected by the omega detection unit 122 and the face shape detected by the face shape detection unit 121, and then detects it as a face area. When the detected facial areas move each time, that is, when the subjects move or when the detected facial areas move in the adjacent preview image, the accuracy of the detected facial areas is improved and tracked in order to capture a stable composition.

The aforementioned group video mode is a shooting mode for capturing a group video, and the personal video mode is a mode for capturing a single video. This may automatically determine whether the group image mode or the personal image mode through the information on the number of faces in the result information of the face region detection and tracking unit. If it is determined that the group image mode, if there is a portion where a plurality of face areas are densely detected in the preview image, the direction of the camera 111 is enlarged by zooming, and then the image is determined again to determine the dense portion. In the case where the densely detected portion does not exist specifically, the following predetermined criteria are given.

The image capturing control module 130 detects an image composition of the detected and traced face region in the face region extraction mode and then determines whether the image composition satisfies a predetermined predetermined criterion. If the predetermined criterion is satisfied, the image composition based on the current preview image may be set to an image composition suitable for photographing. If the predetermined criterion is not satisfied, the image capturing module 110 and the image composition satisfy the predetermined criterion. The face region extraction module 120 is controlled. In other words, to control the face region extraction module 120 to detect and track the face region in the current preview image, and to make the image composition satisfy the predetermined criteria in the preview image where the face region is detected and tracked. The image capturing module 110 is controlled to adjust the angle of the camera 111. The angle adjustment of the camera may be a wheel attached to the camera 111 when the device is applied to a mobile communication terminal, and when applied to the robot, the robot may move freely with the camera 111 attached thereto. Use the pan, tilt of the head and the wheel or legs under the robot.

More specifically, the data output from the respective components of the image capturing module 110 and the face region extraction module 120 are determined, and the image composition satisfying a predetermined criterion in the current preview image is controlled. On the other hand, if the image composition is properly adjusted, this time point is selected as the time to shoot the image and control to automatically shoot the image. The data flow and control direction between the components are as shown in FIG.

In this case, the predetermined criterion is an algorithm that enables a best image composition when capturing a face image. Such predetermined criteria may be replaced according to development of various algorithms, and the most optimal imaging apparatus 100 may be configured according to a test value. As an example of an algorithm corresponding to a predetermined criterion, a head room technique, a third segment technique, and a dynamic symmetry scheme may be used. Any of these algorithms can be used or used. Each is described as follows.

The headroom technique presupposes that the space from the head of the person in the image to the top edge of the image should not be too wide. This condition is a condition for stable and good image composition. This headroom technique applies differently depending on the size of the face area, the number of detected face areas, or whether it is a group image mode or a personal image mode. Since the closer the distance between the camera 111 and the person is, the better the image composition is, the image is taken after being enlarged in consideration of the number of detected face areas, the size of the face, and the like. For this enlargement, the camera 111 itself may be moved forward or a zoom function may be used.

The three-division technique or dynamic symmetry scheme is based on the premise that the face region is located at or within the specified boundary of the preview image. In the case of the person image, the image composition of placing the person in the middle of the image is a bad image composition. Therefore, the image is divided into three horizontally and vertically, and the face area of the subject is disposed at the intersection point of the image.

In addition to the predetermined criteria, an embodiment of the criteria applicable to the present invention may be specified as follows.

1. Not bounded by boundaries

1.1 No face / head detected outside the upper / lower / left / right boundaries.

1.2 The specific range of the boundary may vary depending on the number or size of faces / heads detected in the image.

2. When there is only one face / head detected, the appropriate size and face / head should be centered on the screen.

2.1 Detected face / head size must be above a certain value.

2.2 The center of the face / head shall be located within a specific area pre-defined horizontally / vertically.

3. When there is more than one face / head detected, the appropriate size and distribution should be even.

3.1 The detected face / head size sum should be above a certain value.

3.2 All detected faces / heads should not be present only in certain areas of the left or right side (should not be biased to one side).

3.3 The average center of detected faces / heads shall be inside a predefined area in the horizontal / vertical direction.

3.4 The maximum distance between detected faces / heads should be less than a certain percentage of the screen width.

4. The detected face must satisfy the skin color condition (applies only to the face shape detection unit).

4.1. The condition of skin color should be satisfied in the YCbCr / RGB domain. If you have a face that is too bright, too dark, or distorted, discard the picture.

In addition, various algorithms may be replaced or added to the imaging apparatus 100 and the method of the present invention.

The image storage module 140 is a component for storing an image and a preview image taken by the camera 111.

2 is a flowchart of an image capturing method for automatically adjusting an image composition according to an exemplary embodiment of the present invention. Referring to FIG. 2, an image capturing method according to the present invention includes a process of detecting and tracking a face region largely, determining whether an image composition of a face region satisfies a predetermined criterion, and an image composition does not satisfy a predetermined criterion. And changing the angle, focus, and zoom function of the camera lens to satisfy a predetermined criterion. Each configuration is as follows.

First, it is determined whether the image photographing apparatus 100 is set to the automatic composition adjusting mode (S210). If not, the user manually composes the image and captures and stores the image at a desired time point (S215). In the adjustment mode, the face region extraction module 120 detects and tracks the face region on the preview image (S220). The image capture control module 130 detects the face region and detects the face region. 100 is automatically determined whether the group image mode or the personal image mode (S230). In the case of the personal image mode, after determining whether the image composition of the preview image satisfies a predetermined standard (S260), if not, the camera lens and the zoom Change the state (S270). When applied to a mobile communication terminal or a robot, it can be replaced by an automatic movement function using a wheel or a leg instead of a zoom function.

However, in the group image mode, the area in which the detected and tracked face areas are gathered in the preview image, that is, the area of the face is determined is determined (S240), and the camera lens and the focus are changed to the dense area. If there is no dense area, it is determined whether it meets a predetermined standard as in the personal image mode (S260), and if not, the camera lens and zoom state are changed. (S270) Similarly, it is applied to a mobile communication terminal or a robot. In this case, instead of the zoom function, it can be replaced by an automatic moving function using a wheel or a leg. After adjusting the image composition in the group image mode and the personal image mode (S270), the above steps are repeated until the predetermined criteria are satisfied by performing the face detection and tracking (S220) of the preview image again. If the predetermined criterion is satisfied, the image is captured and stored (S280).

1 is a block diagram of an image photographing apparatus for automatically adjusting an image composition according to an embodiment of the present invention.

2 is a flowchart of an image capturing method for automatically adjusting an image composition according to an exemplary embodiment of the present invention.

Claims (23)

In the video recording device for automatically adjusting the video composition, An imaging module for capturing an image by changing an angle and a focus of a camera lens; A face region extraction module for detecting and tracking a face region in the preview image of the captured image; And an image photographing control module for controlling to change an angle and a focus of the camera lens so that the image composition of the detected and tracked face region satisfies a predetermined criterion. The method of claim 1, wherein when the image composition satisfies a predetermined criterion, When the predetermined criteria are satisfied, the image capturing control module controls the image capturing module to capture an image, and the image capturing module captures an image according to the control. Imaging device. The method of claim 2, And an image storage module for storing the image captured by the image capturing module. The method of claim 1 or 2, wherein the image recording module, A camera for photographing the image, A camera driver which moves or rotates the camera in up, down, left and right directions or performs a zoom function of the camera lens according to the control of the image capturing control module; And an image display unit for displaying the preview image. According to claim 1 or 2, wherein the face area extraction module, An omega detector for detecting an omega shape from the preview image; A face shape detection unit for detecting a face shape in the preview image; And a face region detection and tracking unit for detecting and tracking a face region by combining the detected omega and face shapes. The method of claim 1 or 2, wherein the face area detection unit, And the facial area is detected using the skin color distribution or the active coutour in the preview image. The method according to claim 1 or 2, wherein the shooting mode of the video recording device is The image capturing apparatus for automatically adjusting the image composition, characterized in that the individual image mode or group image mode. The method according to claim 7, wherein the photographing mode of the image photographing apparatus is a group image mode. The image capturing control module is configured to change an angle, a focus, and a zoom function state of the camera lens to a region in which a face region detected and tracked by the face region extraction module is distributed by a predetermined pixel or more on the preview image. An image capturing apparatus for automatically adjusting an image composition. The method according to claim 1 or 2, wherein the predetermined criterion is An apparatus for automatically adjusting an image composition, characterized in that a criterion is set using at least one of a head room technique, a three-division technique, and a dynamic symmetric composition technique. The method of claim 9, wherein the predetermined criterion is The detected and tracked face area and the detected and tracked head area together are (1) not present on the upper, left and right boundaries of the preview image, and (2) not present on the lower region of the preview image. Not, And the size of the predetermined area is determined according to the number of pixels of the face area and the head area. The method of claim 10, wherein the predetermined criterion is When the number of the face area and the head area is one, (1) the predetermined number of pixels of the face area and the head area is more than a predetermined number, and (2) the predetermined area where the face area and the head area are located at the center of the preview image. An image capturing apparatus for automatically adjusting an image composition, characterized in that located in an area of a size. The method of claim 10, wherein the predetermined criterion is When the number of the face area and the head area is plural, (1) the number of pixels of the face area and the head area is more than a predetermined number, and (2) the face area and the head area are predetermined on the left or right side of the preview image. (3) the face region and the head region exist in a region of a predetermined size located at the center of the preview image, and (4) the maximum distance between the face region and the head region is An image capturing apparatus for automatically adjusting an image composition, characterized in that it is equal to or less than a predetermined ratio of the horizontal length of the preview image. The method according to claim 11, wherein the predetermined criterion is And the color of the face region is a skin color. The method of claim 12, wherein the predetermined criterion is And the color of the face region is a skin color. In the image recording method for automatically adjusting the image composition, Detecting and tracking the face area in the preview image; Determining whether an image composition of the face region detected and tracked in the preview image satisfies a predetermined criterion; And changing an angle and a focus of the camera lens until the predetermined criterion is not satisfied when the predetermined criterion is not satisfied. The method of claim 15, wherein the predetermined criterion is satisfied. And photographing the image at a time point when the predetermined criterion is satisfied. The method of claim 16, wherein after the photographing of the image, The method of claim 1, further comprising the step of storing the captured image. The method of claim 15, wherein the predetermined criteria, An image capturing method for automatically adjusting an image composition, characterized in that a criterion is set using at least one of a head room technique, a third segment technique, and a dynamic symmetric composition technique. 19. The method of claim 18, wherein the predetermined criterion is The detected and tracked face area and the detected and tracked head area together are (1) not present on the upper, left and right boundaries of the preview image, and (2) not present on the lower region of the preview image. Not, And the size of the predetermined area is determined according to the number of pixels of the face area and the head area. The method of claim 19, wherein the predetermined criterion is When the number of the face area and the head area is one, (1) the predetermined number of pixels of the face area and the head area is more than a predetermined number, and (2) the predetermined area where the face area and the head area are located at the center of the preview image. An image capturing method for automatically adjusting an image composition, characterized in that located in an area of a size. The method of claim 19, wherein the predetermined criterion is When the number of the face area and the head area is plural, (1) the number of pixels of the face area and the head area is more than a predetermined number, and (2) the face area and the head area are predetermined on the left or right side of the preview image. (3) the face region and the head region exist in a region of a predetermined size located at the center of the preview image, and (4) the maximum distance between the face region and the head region is An image capturing method for automatically adjusting an image composition, characterized by being less than or equal to a predetermined ratio of the width of a preview image. The method of claim 20, wherein the predetermined criterion is And the color of the face region is a skin color. The method of claim 21, wherein the predetermined criterion is And the color of the face region is a skin color.

Images