JP2006211139A - Imaging apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an imaging apparatus which can automatically detect a figure's face which a photography person is going to photo, and which becomes possible to perform an automatic focus control to the figure who is going to take a photograph, an automatic exposure control, an automatic white balance regulation, etc. <P>SOLUTION: The imaging apparatus includes an image pick-up means for photographing an image; a display device for displaying the image photographed by the image pick-up means; a means for adding a mark showing a face detecting point to the image data inputted from the image pick-up means, and displaying to the display device; a face detection means for detecting the figure's face region of the person which laps with the face detection point from the image data inputted from the image pick-up means; and a means for performing at least one of automatic focus control, the automatic exposure control, and the automatic white balance regulations with the part of the face region at least detected by the face detection means as the regulation object region. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an imaging apparatus such as a digital camera.

  In an imaging apparatus having a conventional automatic focusing function, automatic focusing is generally performed using one or a plurality of areas at the center of the screen as a ranging area. However, it is not easy to keep the moving person's face accurately aligned with the distance measurement area with the shutter button pressed halfway.

  Japanese Patent Laid-Open No. 2003-107335 focuses on a person's face regardless of the position of the person by focusing on the person's face in the screen using a means for detecting the person's face. It is disclosed.

However, when a plurality of faces are detected on the screen, the photographer has to select one face from the plurality of faces, which is troublesome. The above publication also discloses a technique for automatically selecting a person at the center of the screen or a person whose face is greatly reflected when a plurality of faces are detected in the screen. The user needs to set a mode for deciding whether to automatically select a correct face.
JP 2003-107335 A

  The present invention can automatically detect the face of the person to be photographed by the photographer, and can perform automatic focusing control, automatic exposure control, automatic white balance adjustment, and the like on the person to be photographed. An object is to provide an imaging device.

  According to the first aspect of the present invention, an image pickup means for picking up an image, a display device for displaying an image picked up by the image pickup means, and a mark indicating a face detection point are added to the image data input from the image pickup means. And displaying on the display device, face detection means for detecting the face area of the person overlapping the face detection point from the image data input from the imaging means, and the face detected by the face detection means Means is provided for performing at least one of automatic focusing control, automatic exposure control, and automatic white balance adjustment with at least a part of the region as an adjustment target region.

  The invention described in claim 2 is characterized in that in the invention described in claim 1, when a face region is detected by the face detection means, a means for erasing the mark indicating the face detection point is provided. .

  According to a third aspect of the present invention, in the second aspect of the present invention, when the face area is detected by the face detecting means, the center of the detected face area is set as a face detection point. It is characterized by that.

  According to the present invention, the photographer can automatically detect the face of the person to be photographed, and can perform automatic focusing control, automatic exposure control, automatic white balance adjustment, etc. on the person to be photographed. It becomes.

  Hereinafter, embodiments of the present invention applied to a digital camera will be described with reference to the drawings.

[1] Explanation of the configuration of the digital camera

  FIG. 1 shows the configuration of a digital camera.

  The digital camera includes a control unit 10 including a CPU, a ROM, a RAM, and the like. An imaging unit 1, focus / exposure adjustment unit 2, ranging unit 3, operation unit 4, face detection unit 5, display unit 6, and nonvolatile memory 7 are connected to the control unit 10.

  The imaging unit 1 includes an imaging element, an optical system, a zoom mechanism, a diaphragm mechanism, and the like. The distance measuring unit 3 measures a distance to an area (an adjustment target area) that is a target of focus adjustment. The focus / exposure adjustment unit 2 performs focus adjustment by controlling the lens mechanism based on the distance measured by the distance measurement unit 3. Further, the focus / exposure adjustment unit 2 performs exposure adjustment by controlling the aperture mechanism and the shutter speed so that the adjustment target region has a suitable brightness. Note that focus adjustment may be performed based on the contrast of the captured image. In other words, the focus may be changed so that the focus with the highest contrast in the adjustment target region may be selected. In this case, the distance measuring unit 3 is unnecessary.

  The operation unit 4 includes various buttons such as a shutter button. The face detection unit 5 detects a face area of a person in a face search area described later in the input image. An input image or the like is displayed on the display unit 6. The non-volatile memory 7 holds captured images.

  This digital camera is provided with a portrait mode for imaging a person. Then, when the portrait mode is set, this digital camera automatically detects the face of the person that the photographer intends to shoot, and uses at least a part of the detected face area as an adjustment target area. It has a function to automatically adjust the focus and exposure.

[2] Explanation of face detection method

  A face detection method used in this embodiment will be described. When photographing in the portrait mode, a “face detection point” is set at the center of the display screen of the display unit 6. As shown in FIG. 2, a mark (face detection point mark) M indicating the position of the “face detection point” M is displayed on the display screen of the display unit 6 so that the photographer can know the position of the “face detection point”. . In the example of FIG. 2, a cross mark is used as the face detection point mark M. As the face detection point mark, an X mark, a circle mark, a dot mark, a bracket, or the like can be used.

  As shown in FIG. 3, the photographer adjusts the orientation of the digital camera so that the face detection point mark M overlaps the face of the person to be photographed that exists in the input image displayed on the display unit 6. . In FIG. 3, a indicates the face of the person A, and b indicates the face of the person B. The face detection unit 5 detects only the face matched with the face detection point mark.

  As the face detection method, a known method (for example, a method described in Document A) is used.

  Reference A: P. Viola and M. Jones, “Robust Real Time Object Detection”, IEEE ICCV Workshop Statistical and Computational Theories of Vision, July 2001.

  An outline of a general face detection method will be described. Effective for extracting whether or not a face is extracted by extracting various edge features from the face area including the eyes, nose, mouth, face contour, etc. A face classifier is constructed by learning feature quantities based on statistical methods.

  In order to detect a face from an input image, a similar feature amount is extracted while raster scanning from the edge of the input image with a face size (face matching frame) normalized at the time of learning. The discriminator determines whether or not the area is a face. The feature amount includes, for example, a horizontal edge, a vertical edge, a right diagonal edge, a left diagonal edge, and the like.

  If no face is detected, the input image is reduced at a certain rate, and the face is searched for the reduced image while performing raster scanning in the same manner as described above. By repeating such processing, a face of any size can be found in the image.

  In the present embodiment, the face search area S is limited as shown in FIG. 3 in order to appropriately detect only the face that matches the face detection point mark. That is, the face search area S centering on the face detection point is set. Then, while raster scanning the face search area S with the face size (face matching frame) normalized at the time of learning, feature quantities are extracted, and a face discriminator is used to determine whether or not the face is a face. Do. In this way, by providing face detection points and limiting the face search area, it is possible to detect only the faces that overlap the face detection points. In the example of FIG. 3, there are two faces a and b in the image, but the face b is excluded from the search target, and therefore only one face a that overlaps the face detection point can be detected.

  The size of the face search area S is set larger than the face matching frame F in consideration of the difficulty of aligning the center of the face with the center of the face detection point mark by the photographer as shown in FIG. . However, if the size of the face search area S is too large, a plurality of faces may be detected. Therefore, the size of the face search area S is set so that two faces do not fit. Here, the size of the face search area S is set to 1.5 times the face matching frame F.

  By the way, a face cannot be detected unless the size of the face in the image is substantially the same as the size of the face matching frame F, but the size of the face in the image is not constant. Therefore, when a face cannot be detected, the size of the face search area S is kept as it is, and the image is reduced at a certain rate to search for a face. By repeating such processing, a face of an arbitrary size is detected from the face search area S. The image reduction ratio and the number of reductions are determined in advance.

  Note that reducing the image while keeping the size of the face search region S is equivalent to increasing the size of the face search region S while keeping the image as it is.

  FIG. 5 shows the relative relationship between the face search area and the face size. In FIG. 5, in order to express the relative relationship between the face search area and the face size in an easy-to-understand manner, the size of the face search area S is increased without changing the image. S1 to S4 represent face search areas. A small face is searched in a small area near the face detection point. When detecting a large face, the face search area is expanded.

  When a face is detected, in order to make the photographer recognize this, as shown in FIG. 6, the face detection point mark (cross mark) M is deleted, and a rectangular mark (face area) surrounding the detected face is deleted. Display mark) W is displayed.

  The subject is also expected to move. Therefore, in this embodiment, after the face is detected, the face detection point is moved to the center point of the face detection area, and the face is searched similarly using the position as a reference. The position of the face detection point may be determined by predicting the movement of the subject. In this way, the subject can be tracked by continuously detecting the face. Further, in order to track the subject, a known tracking method (see Document B) may be used. Alternatively, the color information of the detected face area may be acquired and the subject may be tracked based on the color information (see Document C).

  Reference B: "An iterative image registration technique with an application to stereo vision" Proc. Of the international Joint Conference on Artificial Intelligence, pp.674-679, 1981.

  Reference C: “Proposal of a modified HSV color system effective for face region extraction”, Journal of Television Society Vol.49, No.6, pp.787-797 (1995).

[3] Explanation of processing procedure

  FIG. 7 shows a focus and exposure adjustment processing procedure based on face detection executed by the digital camera when the portrait mode is set.

  An image picked up by the image pickup unit 1 is input (step S1). Then, it is determined whether the flag F is reset (F = 0) or set (F = 1). The flag F is reset when the portrait mode is set, and is reset when the shutter button is pressed in the portrait mode. This flag F is set when a face is detected in the face detection process of step S5 described later.

  If it is immediately after the port trade mode is set, since the flag F is reset, the process proceeds to step S3, where the face detection point mark M is combined with the input image and displayed on the display unit 6. Then, the process proceeds to step S5. The photographer adjusts the orientation of the digital camera so that the face detection point mark M overlaps the face of the person who wants to photograph.

  In step S5, a face search area centered on the face detection point is set, and face detection processing is performed in the face search area as described above.

  If the face cannot be detected (NO in step S6), the process returns to step S1 and the image captured by the imaging unit 1 is input. When the face can be detected, the flag F is set (F = 1) (step S7), the face detection point mark M is deleted (when M is displayed), and the face surrounding the detected face is detected. An area display mark W is displayed (step S8). Further, focus adjustment and exposure adjustment are performed using at least a part of the detected face area as an adjustment target area (step S9).

  Next, the center of the detected face area is set as a face detection point (step S10). Then, the process returns to step S1. When the process returns to step S1 after the face is detected, the flag F is set, so when an image captured by the imaging unit 1 is input, the image is displayed as it is on the display unit 6 (step S4). ). Then, face detection processing is performed on the input image using the face search area centered on the face detection point set in step S10 (step S5).

  If the shutter button is pressed while such processing is being performed, the flag F is reset (F = 0). Therefore, when returning to step S1 thereafter, the image is captured by the imaging unit 1. When an image is input, a face detection point mark is combined with the input image and displayed on the display unit 6. Then, face detection processing is performed on the input image using a face search area centered on the face detection point (step S5).

  In step S9, focus adjustment and exposure adjustment are performed using the detected face area as an adjustment target area, but only one of the adjustments may be performed. Further, the white balance may be adjusted using at least a part of the detected face area as an adjustment target area.

  In the above embodiment, every time it is determined in step S6 that face detection has been performed, focus adjustment and exposure adjustment are performed, but the face is detected in the previous frame, and from the camera to the face detected in the previous frame. If the distance between the camera and the distance from the camera to the face detected in the current frame is not different, the focus adjustment may not be performed. Whether the distance from the camera to the face detected in the previous frame and the distance from the camera to the face detected in the current frame are different depends on the size of the face detected in the previous frame and the face detected in the current frame. It can be determined based on whether or not the size is different. The size of the face can be obtained from the relative relationship between the face search area and the face size when the face can be detected.

  Similarly, when a face is detected in the previous frame and the brightness of the face area detected in the previous frame is not different from the brightness of the face area detected in the current frame from the camera, exposure adjustment is not performed. You may do it.

  In the above embodiment, when the portrait mode is set, focus and exposure adjustment based on face detection is always performed. However, a switch for switching the face detection mode is provided, and this switch is turned on. The focus and exposure adjustment based on the face detection may be performed only in the case where it is performed. Regardless of the photographing mode such as the portrait mode, focus and exposure adjustment based on face detection may be performed when the shutter button is half-pressed. In this case, when a face cannot be detected even if face detection is performed on a plurality of frames, the focus and exposure adjustment based on the face detection is ended. When the face is not detected in this way and the shutter button is pressed, shooting is performed according to the “automatic scene determination function” that has been conventionally provided.

  The “automatic scene determination function” refers to a function that automatically determines each scene such as portrait, landscape, night view, etc., and automatically sets an aperture and shutter speed suitable for the determination result.

It is a block diagram which shows the structure of a digital camera. 5 is a schematic diagram showing a face detection point mark M. FIG. The schematic diagram which shows the example of a display image when direction of a digital camera is adjusted so that the face detection point mark M may overlap with the face of the person who wants to image | photograph which exists in the input image currently displayed on the display part 6 It is. 5 is a schematic diagram showing a face search area S. FIG. It is a schematic diagram which shows the relative relationship between a face search area | region and face size. FIG. 6 is a schematic diagram showing that when a face is detected, the face detection point mark M is erased and a rectangular mark (face area display mark) W surrounding the detected face is displayed. It is a flowchart which shows the focus and exposure adjustment processing procedure based on the face detection performed by a digital camera when portrait mode is set.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Control part 1 Imaging part 2 Focus and exposure adjustment part 3 Distance measuring part 4 Operation part 5 Face detection part 6 Display part 7 Non-volatile memory

Claims (3)

Imaging means for capturing an image;
A display device for displaying an image captured by the imaging means;
Means for adding a mark indicating a face detection point to the image data input from the imaging means and causing the display device to display the mark;
Face detection means for detecting a face area of a person overlapping a face detection point from image data input from the imaging means, and at least a part of the face area detected by the face detection means is an adjustment target area Means for performing at least one of automatic focusing control, automatic exposure control, and automatic white balance adjustment,
An imaging apparatus comprising: The imaging apparatus according to claim 1, further comprising a unit that erases a mark indicating a face detection point when a face area is detected by the face detection unit. The imaging apparatus according to claim 2, further comprising means for setting a center of the detected face area as a face detection point when the face area is detected by the face detecting means.

Images