JP2006109005A - Image output apparatus and imaging apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image output device enabling a user to easily confirm the magnified image of a specific image included in image data. <P>SOLUTION: The image output device is provided with a detection means 120 for detecting the specific image included in the image data, a decision means 107 for deciding a specific point in the image data on the basis of the detection result of the specific image, and an output means 110 for outputting the magnified image of the specific image on the basis of the specific point. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an image output apparatus and an image output method capable of outputting an enlarged image of a specific image portion included in image data.

  A method for a user to confirm a captured image in a conventional digital camera will be described.

  When the camera is set to the image reproduction mode, the image data recorded on the recording medium or the like is read out to the buffer and displayed on the display unit of the camera.

  Here, when the zoom button for enlarging and reducing the image data is operated, the image displayed on the display unit is enlarged or reduced. In addition, when the cross button is operated while the image is enlarged, the position of the enlarged portion in the captured image can be changed.

On the other hand, when photographing, there is a camera that detects the face of a person as a subject and sets the person to a predetermined size using optical zoom and electronic zoom (see, for example, Patent Document 1). . Further, in Patent Document 1, when performing electronic zoom, the cutout position of an image is determined according to the face position.
JP-A-6-217187 (paragraph numbers 0092 to 0094, FIG. 4 etc.)

  In general, when a captured image is enlarged and displayed, it is often desired to enlarge and view a subject portion that is noticed by the user. However, in the conventional digital camera, the central portion of the photographed image is first enlarged and displayed, and the subject portion of interest cannot be directly enlarged and displayed.

  Here, in order to enlarge and display the subject portion of interest, it is necessary to further enlarge the image or change the position of the enlarged portion in the captured image by operating the zoom button or the cross button. The operation is troublesome.

  On the other hand, the camera disclosed in Patent Document 1 uses an optical zoom and an electronic zoom in combination to perform shooting so that the size of the subject with respect to the shooting screen is appropriate, so that it is used in ID photos and security systems. This is effective when shooting a person with a specific size.

  However, since the size of the subject is automatically determined in the camera of Patent Document 1, a part of the captured image is changed to a size that the user likes, and a person including a landscape or the like is viewed. Cannot be viewed as a subject and an enlarged image of a person's face cannot be viewed after shooting.

  Further, since the camera of Patent Document 1 cuts out an image according to the position of the face, the display position is changed when an image in a state before being cut out is enlarged or reduced, or when a plurality of persons are included. It cannot be changed.

  The present invention has been made in view of the above-described problems, and provides an image output apparatus and an image output method capable of enlarging a specific image portion in a captured image and easily confirming the specific image portion. The purpose is to do.

  An image output apparatus according to the present invention includes a detection unit that detects a specific image portion included in image data, a determination unit that determines a specific point in the image data based on a detection result of the specific image portion, and the specific point. Output means for outputting an enlarged image of the specific image portion with reference to the above.

  The image output method of the present invention includes a detection step of detecting a specific image portion included in image data, a determination step of determining a specific point in the image data based on a detection result of the specific image portion, And an output step of outputting an enlarged image of the specific image portion based on the specific point.

  According to the present invention, an enlarged image of a specific image portion with a specific point as a reference can be easily confirmed.

  Examples of the present invention will be described below.

  A camera that is Embodiment 1 of the present invention will be described. FIG. 1 is a block diagram showing the configuration of the camera of this embodiment.

  The image sensor (image acquisition means) 101 photoelectrically converts a subject image (optical image) formed by the photographing lens 114 into an electrical signal. The signal processing circuit (image acquisition unit) 102 generates image data by performing predetermined processing (color processing, gamma correction, etc.) on the output signal of the image sensor 101. The image data is stored in the buffer memory 103.

  When recording image data, the image data in the buffer memory 103 is sent to the compression / decompression circuit 104 to perform compression processing, and the compressed image data is recorded via the recording / reading circuit 105 to the recording medium. 106.

  On the other hand, when displaying recorded image data such as post-view (display of a photographed image immediately after photographing), the image data recorded on the recording medium 106 is read by the recording / reading circuit 105. The read image data is subjected to image expansion processing by the compression / decompression circuit 104 and then stored in the buffer memory 103.

  The face position detection circuit (detection means) 120 detects the face of a person in the captured image using the image data stored in the buffer memory 103, and obtains the position coordinates of the face in the captured image. The operation for detecting a person's face will be described later.

  The zoom center position setting circuit 111 sets the zoom center position (specific point) based on the output from the user interface 113. Here, the zoom center position is a position that serves as a reference when enlarging and displaying a partial area of the captured image, as will be described later. That is, the image in the partial area is enlarged at a preset enlargement ratio with the zoom center position as the center.

  In the user interface 113, the zoom center position setting condition can be changed.

  The zoom center position determining circuit (determining means) 107 is based on the zoom center position set by the zoom center position setting circuit 111 and the area (face detection area) including the face detected by the face position detecting circuit 120 is used. Determine the zoom center position.

  The image magnification setting circuit 112 sets the image magnification based on a signal from the user interface 113.

  The display image generation circuit (output unit) 108 is configured to determine the zoom center position determined by the zoom center position determination circuit 107 and the image magnification setting circuit for the image data (recorded image data) read from the buffer memory 103. By performing image processing based on the enlargement ratio set in 112, image data for display that matches the display screen of the monitor 110 is generated. The image processing includes processing for enlarging or reducing an image from the buffer memory 103, or cutting out a part of the image.

  The image data generated by the display image generation circuit 108 is subjected to conversion processing by the D / A converter 109 and then output to the monitor 110 and displayed.

  Next, a method for detecting a person's face in a captured image will be described.

  As a face detection method, a detection method using learning typified by a neural network, a method of detecting a characteristic part of a physical shape such as an eye or a nose from an image region using template matching, a skin color There are detection methods using statistical analysis by detecting image feature quantities such as eye shapes.

  In general, face detection is performed by combining a plurality of methods described above. For example, there are a method of detecting a face using wavelet transform and image feature amount, and a detection method combining template matching and the like.

  In this embodiment, an area corresponding to a person's face is extracted from a captured image using color information, and a part having a characteristic physical shape (in the following description, a person's eyes) is extracted from the image area as a template. The person's face is detected by specifying by matching. Hereinafter, the face detection operation in this embodiment will be briefly described.

  FIG. 4 is a flowchart for explaining the face detection operation in this embodiment. FIG. 3 is a diagram for explaining a face detection method in this embodiment. Here, FIG. 3A shows a captured image (original image) 200 that is a target when detecting a face.

  In step S <b> 101 of FIG. 4, a skin color region is extracted from the captured image 200. Here, FIG. 2 is a chromaticity diagram showing representative colors in the CIELAB Lab color space, and a region (skin color region) A surrounded by an ellipse in the diagram indicates a region that is likely to be a skin color.

  Therefore, in step S101, a region corresponding to the color in the skin color region A is extracted from the captured image based on the color information in the captured image and the color information in the skin color region A shown in FIG. FIG. 3B shows an image 201 obtained by extracting an area corresponding to the chromaticity in the skin color area A from the photographed image 200.

  In step S102 of FIG. 4, high-pass filter processing is performed on the image data extracted in step S101. FIG. 3C shows an image 202 generated by high-pass filter processing. FIG. 5 shows filter coefficients of the two-dimensional high-pass filter used in this embodiment.

  In step S103, template matching is performed, and eyes are detected in the extracted image. Specifically, as shown in FIG. 3D, a high-pass filter process is performed while sequentially moving a template 203 corresponding to a human eye prepared in advance from the upper side to the lower side in the direction indicated by the arrow. The image 202 is overlapped. The eye position can be detected by examining the correlation between the template 203 and the image 202.

  In step S104, a face is detected based on the eye position, and feature quantities such as the orientation and size of the face are extracted. FIG. 6 shows a state in which a human face is detected in the photographed image.

  Next, the operation of the zoom center position determination circuit 107 will be described.

  FIG. 7A shows a state where the face of one person included in the photographed image is detected by the face detection operation described above. FIG. 7B shows a state where the center 301a of the face detection area 301 is determined as a reference for the zoom center position. Here, the zoom center position may be approximately the center of the face detection area 301.

  In the present embodiment, the center of the face detection area 301 is determined as the zoom center position, but other positions may be used as the zoom center position. For example, a position that satisfies a predetermined condition, such as a center position between two eyes of a person or a position having a specific ratio in the vertical and horizontal directions in the face detection area, is set as the zoom center position. Can do. Conditions regarding the setting of the zoom center position can be changed in the user interface 113.

  FIG. 7C shows an image generated by the display image generation circuit 108 based on the zoom center position determined by the zoom center position determination circuit 107 and the enlargement ratio set by the image enlargement ratio setting circuit 112. Show.

  As described above, if the zoom center position and the enlargement ratio are set in advance by operating the user interface 113, when the recorded image data is displayed on the monitor 110, the image in the face detection area is displayed at a predetermined magnification. Enlarged display is possible. That is, when displaying a photographed image, an enlarged view of a person's face included in the photographed image can be observed immediately, and the troublesome operation of enlarging a person's face by operating a zoom button or a cross button like a conventional camera. It is not necessary to perform a special operation.

  FIG. 8 is a diagram illustrating a state in which face detection is performed on faces of a plurality of persons in a case where a plurality of persons are included in a captured image. In this case, a plurality of face detection areas 401 to 405 exist in one image. In the present embodiment, the center in a specific face detection area selected in advance in the user interface 113 is determined as the zoom center position.

  Here, the face detection area selected by the operation of the user interface 113 is set to a face detection area (face detection area 403 in FIG. 8) located in the center of the photographed image, or face detection located at the edge of the photographed image. Or an area (face detection area 401 or 405 in FIG. 8). In addition, the user interface 113 can change the selection condition of the face detection area.

  On the other hand, when there are a plurality of face detection areas, these face detection areas may be displayed on the monitor 110 and one face detection area may be selected by operating the user interface 113. In this case, since the selected face detection area can be enlarged and displayed only by the user selecting the face detection area, the person included in the captured image can be easily confirmed as compared with the conventional camera. .

  Even when the face detection area to be enlarged is selected in advance, another face detection area may be selected by operating the user interface 113. That is, in the state shown in FIG. 8, the face detection area selected in advance can be displayed on the monitor 110 in a state where it can be distinguished from other face detection areas, and the face detection area can be changed by operating the user interface 113. .

  When the face detection area selected in advance is enlarged and displayed, if another face detection area is included in the display screen, the other face detection area is selected via the user interface 113. You may be able to do it. Further, another face detection area outside the display screen may be selected.

  In the user interface 113, the setting of the zoom center position may be changed.

  On the other hand, when there are a plurality of face detection areas, the center of gravity of the area including all these face detection areas may be set as the zoom center position. Here, in FIG. 9, the position serving as the center of gravity for the face detection areas 401 to 405 is indicated by a black circle 406.

  In this way, if the center of gravity 406 is used as the zoom center position and enlarged display is performed at a predetermined magnification, a part of the captured image including a plurality of persons can be enlarged and displayed. It can be done easily.

  On the other hand, when there are a plurality of face detection areas, the face detection area having the largest face detection area size may be selected. For example, when shooting a crowded place with people, the face of a person other than the person who is the main subject (the main subject to be shot) may enter the periphery of the shooting screen. The face of the subject is often the largest.

  Therefore, by selecting the largest size face detection area as described above, the area including the face of the main subject can be enlarged and displayed, and the main subject can be easily confirmed.

  Further, when there are a plurality of face detection areas, the face detection area having the brightest brightness may be selected. For example, when a place crowded with people is photographed using illumination light, the illumination light is most irradiated to the face of the main subject and is often brightened. Therefore, by selecting the brightest face detection area as described above, the area including the face of the main subject can be enlarged and displayed, and the main subject can be easily confirmed.

  In this embodiment, the face is detected after the recorded image data is read out. However, when the photographed image data is generated by the photographing operation, the face position detection circuit 120 detects the face position in advance. In addition, the detected face detection information (information regarding the face detection area) may be stored. In this case, when the image data is reproduced, the stored face detection information is read, and the zoom center position determination circuit 107 determines the zoom center position based on the face detection information.

  In the present embodiment, the case where a human face is detected from a captured image has been described, but an area other than the face may be extracted. Specifically, a part having a characteristic shape, color, or the like can be extracted.

  Here, when a car traveling on a road is photographed, an enlarged image of a portion corresponding to the license plate can be generated from the photographed image.

  In this embodiment, the case where a part of the photographed image is enlarged and displayed on the monitor 110 has been described. However, the enlarged image data may be output to a printer and printed.

  Furthermore, in the present embodiment, a case has been described in which a part of a captured image is enlarged and displayed using the monitor 110 provided in the camera, but the present invention is not limited to this. For example, the present invention can also be applied to a case where image data generated by a camera is transferred to an external device (image output device) such as a PC and a part of a captured image is enlarged and displayed by the external device.

1 is a block diagram illustrating a configuration of a camera that is Embodiment 1 of the present invention. FIG. The figure which shows the typical chromaticity in CIELab color space. FIG. 5 is a diagram illustrating a face detection operation according to the first embodiment. 5 is a flowchart illustrating a face detection operation according to the first exemplary embodiment. The figure which shows the filter coefficient of a two-dimensional high pass filter. In Example 1, it is a figure which shows a face detection result. FIG. 6 is a diagram illustrating a face detection state (A), a zoom center position determination state (B), and an image enlargement state (C) in the first embodiment. The figure which shows the face detection result with respect to the picked-up image containing a several person. The figure which shows the gravity center position in several face detection area | region.

Explanation of symbols

107: Zoom center position determining circuit (corresponding to determining means)
108: Display image generation circuit (corresponding to output means)
110: Monitor 120: Face position detection circuit (corresponding to detection means)

Claims (6)

Detecting means for detecting a specific image portion included in the image data;
Determining means for determining a specific point in the image data based on a detection result of the specific image portion;
And an output unit that outputs an enlarged image of the specific image portion with the specific point as a reference.   The image output apparatus according to claim 1, wherein the specific image portion is an image portion of a human face.   3. The image output apparatus according to claim 1, wherein the determining unit determines the specific point based on a center position of the specific image portion or a position that matches a predetermined condition in the specific image portion. .   In the case where a plurality of the specific image portions are detected by the detecting means, the determining means is based on at least one of the size and brightness of the specific image portions and the centers or centroids of the plurality of specific image portions. The image output apparatus according to claim 1, wherein a specific point is determined. An image output apparatus according to any one of claims 1 to 4,
An image pickup device comprising: an image acquisition unit that photoelectrically converts a subject image and generates image data in which the specific image portion is detected by the image output device. A detection step of detecting a specific image portion included in the image data;
A determination step of determining a specific point in the image data based on a detection result of the specific image portion;
And an output step of outputting an enlarged image of the specific image portion with the specific point as a reference.

Images