JP2009117975A - Image pickup apparatus and method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image pickup apparatus that is more convenient in use for users and hardly causes erroneous operations. <P>SOLUTION: The image pickup apparatus 10 includes a manipulation part 12, a control part 14, an image pickup part 16, a face extracting part 18, a face identifying part 20, a range setting part 22, a pattern identifying part 24, a memory 26, and a displaying part 28. The image pickup part 16 generates an image by picking up an object and detects whether or not this image includes a face image of a person giving an instruction to the image pickup apparatus 10 previously registered in the memory part 26. When the image includes the face image, an area to detect a pattern to send the instruction to the image pickup apparatus is set near the face image included and the process designated by the pattern is executed when the pattern is detected in this area. Therefore, the pattern can be detected accurately and erroneous operations can be prevented. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to an image capturing apparatus and an image capturing method, and more particularly to an image capturing apparatus and method for automatically controlling an image capturing apparatus in response to a signal from a subject.

  Conventionally, when a photographing device such as a digital camera automatically starts processing, such as when a photographer himself is a subject, the photographing device is controlled by using a self-timer function or a remote control. It was.

  However, in the case of using the self-timer function, the photographer himself must operate the photographing apparatus to set the self-timer function and then move to become a subject. Therefore, there have been problems such as the movement time being short or long, and a problem that it is difficult for the subject to understand the shooting timing. In the case of using a remote controller, it is possible to determine the timing of shooting, but the remote controller must be carried. Furthermore, there is a problem that it cannot be used depending on the shooting environment. Therefore, there has been a demand for a method that can control the photographing apparatus without using a self-timer function or a remote controller.

Japanese Patent Application Laid-Open No. 2004-228561 discloses a method of performing shooting in response to a change in a specific portion in a captured image. Specifically, in Patent Document 1, a photographer designates an area in which a change is detected in advance, detects a change in the designated area, and imaging is started when a change is detected.
JP 2006-121761 A

  However, in the imaging method disclosed in Patent Document 1, since the photographer designates an area in which a change is detected in advance, for example, when the photographer is a subject alone, detection of which area is designated as the area Is difficult, and photography is not started properly. In addition, even when there are a plurality of subjects, there is a problem that a moment when the subject is arranged or the pose changes is recognized as a change and an appropriate picture cannot be taken.

  The present invention eliminates the disadvantages of the prior art, and provides a photographing apparatus and a photographing start method that are less likely to malfunction when operating the photographing apparatus from a remote location and that are more convenient for the user. Objective.

  In order to solve the above-described problems, in the photographing apparatus and the photographing method according to the present invention, a face of a person giving an instruction to the photographing apparatus and a pattern for giving the instruction to the photographing apparatus are registered in the photographing apparatus in advance. A registered photographer's face image is detected from an image created by the photographing apparatus, and a range for detecting a registered pattern is set near the detected face image. Thereafter, when a pre-registered pattern is detected from the set range, the process designated by the pattern is started with this as a trigger.

  Since this pattern can be associated with an arbitrary process, for example, by registering a certain pattern in the photographing apparatus as a pattern for starting photographing, the photographing apparatus can automatically start photographing. Become. In the present invention, in addition to the pattern for starting shooting, it is possible to register an arbitrary pattern corresponding to an arbitrary instruction. For example, a pattern for causing the shooting apparatus to end shooting or a flash for the shooting apparatus. Examples of such a pattern may be mentioned, but the present invention is not limited to this.

  According to the present invention, in order to detect a photographer's face registered in advance from an image obtained by the photographing apparatus and set a region for recognizing a pattern for controlling the photographing apparatus according to the position of the face, The person who gives an instruction to the photographing apparatus is in any pose in the area captured by photographing, and at any position, the photographing apparatus performs predetermined processing, for example, photographing at a timing desired by the photographer. It is also possible to perform processing for ending the shooting. Therefore, it is possible to provide a photographing apparatus and a photographing method that are less likely to malfunction and that are more convenient for the user.

  Embodiments of the present invention will now be described in detail with reference to the accompanying drawings. FIG. 1 is a block diagram schematically showing a photographing apparatus 10 according to the present invention. In FIG. 1, an imaging device 10 according to the present invention includes an operation unit 12, a control unit 14, an imaging unit 16, a face extraction unit 18, a face identification unit 20, a range setting unit 22, a pattern identification unit 24, a storage unit 26, and a display. The image capturing unit 16 shoots the subject and creates an image, and detects whether or not the face image of the person who gives instructions to the image capturing device 10 registered in the storage unit 26 is included in the image. If it is captured, an area for detecting a pattern for giving an instruction to the photographing apparatus is set near the captured face image, and a pattern registered in advance in this area is detected. The corresponding control is performed. Parts that are not directly related to understanding the present invention are not shown and redundant description is avoided.

  In FIG. 1, an operation unit 12 is a manual operation device to which an operator's instruction is input, and an operation signal 30 is sent to the control unit 14 in accordance with an operator's manual operation state, for example, a stroke operation of a shutter button (not shown). To supply. Each signal is identified by the reference number of the connecting line in which it appears. The control unit 14 controls and controls the entire photographing apparatus 10. For example, in this embodiment, the control signal is supplied to each unit in response to the operation signal 30 supplied from the operation unit 12, and the imaging unit 16, the face extraction unit 18, the face identification unit 20, the range setting unit 22, and the pattern identification unit 24 are supplied. Then, the storage unit 26 and the display unit 28 are made to perform necessary processes. For example, the control unit supplies the control signal 32 to the imaging unit 16 to start or end imaging.

  The imaging unit 16 includes an optical system that captures a desired subject image and enters the imaging unit 16, and performs imaging using incident light that has entered through the optical system, thereby creating captured image data 34. The photographing unit 16 is configured using a known one, and can take a picture using a known method. For example, in the present embodiment, the imaging unit 16 uses the solid-state imaging device and creates the captured image data 34 by photoelectrically converting incident light incident on the solid-state imaging device.

  Although a specific configuration is not illustrated, the optical system may include a known configuration. For example, in this embodiment, a lens, a diaphragm adjustment mechanism, a shutter mechanism, a zoom mechanism, an automatic focus (AF) adjustment mechanism, In addition, an automatic exposure (AE) adjustment mechanism is included, and a lens, an automatic focus adjustment mechanism, and an automatic exposure adjustment mechanism are driven by control from the control unit 14, and a desired image is captured and incident on the photographing unit 16. . The photographed image data 34 created by the photographing unit 16 is sent to the face extracting unit 18.

  The face extraction unit 18 extracts, from the supplied captured image data 34, information related to the face image of the person shown in the captured image necessary for the subsequent face identification unit 20, that is, the facial image information 36. . The face image information 36 can be extracted using a known method.

  For example, in this embodiment, the face extraction unit 18 is disclosed in IEICE Journal, D-II, Vol. J80-D-II No. 8 “Personal Identification from Input Face Images with Various Postures”. The feature amount data indicating the feature amount of the face is extracted from the photographed image data 34 as the face image information 36 using the hierarchical Gabor filter. Further, the face extraction unit 18 extracts the position of the face from which the feature amount data has been extracted and the size in the vertical direction and the horizontal direction from the captured image data in a known manner. Note that the reason for extracting the face position and its size in this way is that the pattern setting range is set using the extracted face position and size in the range setting unit 22 in the subsequent stage. Is not limited to extracting them.

  The present invention is not limited to extracting the face image information 36 using a hierarchical Gabor filter. The face image information 36 may be arbitrary, for example, using a known template or the like to extract the shape of the eyes or mouth, or their positions, etc., and the shape of the eyes or mouth, these positions, or shapes A combination of the position and the like can be used as the face image information 36. Furthermore, in addition to the face image information 36, arbitrary information can be acquired from the captured image data 34 in accordance with processing performed by the face identification unit 20 and the range setting unit 22 at the subsequent stage. The extracted face image information 36 is sent to the face identification unit 20.

  The face identification unit 20 collates the face image information 36 supplied from the face extraction unit 18 with the registration information 38 stored in the storage unit 26 in advance, and gives an instruction to the photographing apparatus 10 to the photographed image Is determined, that is, whether or not registration information 38 registered in advance in the photographed image data 34 exists. In this embodiment, the registration information 38 is feature amount data indicating the feature amount of the face image 42 of the owner of the photographing apparatus 10 as shown in FIG. 2, for example, and the storage unit 26 in a format comparable to the face image information 36. Is remembered. The owner's face image 42 is registered as a face image of a person who gives an instruction to the photographing apparatus 10 to release the shutter, that is, to perform photographing.

  In this example, the owner of the photographing apparatus 10 is registered as a person who gives an instruction to the photographing apparatus 10 to start photographing, that is, as a photographing start instructor, but the present invention is not limited to this and is arbitrary. Can be designated as an instruction to start photographing. In addition, the present invention is not limited to registering as an instruction to start shooting, and it is also possible to register a face image as another instruction, for example, a person who gives an instruction to the shooting apparatus 10 to end shooting. It is also possible to register a face image as a person who gives all instructions to the photographing apparatus 10, for example, both at the start and end of photographing, and who gives which instruction to the photographing apparatus 10 is arbitrary It is possible to decide on.

  The number of persons registered as instructors is one in this embodiment, but a plurality of persons may be registered. For example, by registering multiple persons, it is possible to issue an instruction to the photographing apparatus 10 if at least one of them is present in the photographed image, so that other instructors do not want to appear in the photograph. Even if it exists, it becomes possible to give an instruction to the photographing apparatus 10. The present invention is not limited to this.

  The registration information 38 can be registered in the storage unit 26 of the photographing apparatus 10 at an arbitrary timing and in an arbitrary manner. For example, in this embodiment, before taking a picture in the automatic shooting mode, the owner took a picture of his / her face in advance with the photographing apparatus 10 by so-called selfie, or photographed by a third party, and created by this photography. The registration is performed by extracting the feature data from the face image 42 of the owner by the face extraction unit 18 and storing it in the storage unit 26. However, the present invention is not limited to this, and any method can be adopted.

  Using such registration information 38, the face identification unit 20 adopts a known method, and the face image information 36 sent from the face extraction unit 18 and the registration information stored in the storage unit 26 in advance. 38 is checked to determine whether or not a person registered in the photographed image created by the photographing unit 16 is shown. Specifically, in this embodiment, the feature amount data of the face image shown in the photographed image created by photographing and the feature amount of the face image 42 of the owner of the photographing device 10 registered as the registration information 38 The degree of coincidence is calculated by comparing with the data. If the degree of coincidence is greater than or equal to a threshold value, it is determined that the image is captured. Conversely, if the degree of coincidence is lower than the threshold value, it is determined that the image is not captured. Note that the present invention is not limited to this, and any method can be adopted for the determination. The determination result 46 is sent to the range setting unit 22.

  The range setting unit 22 is registered in advance in order to give an instruction to the imaging device 10 when it is determined that a person registered in the captured image is captured as a result of the collation performed by the face identification unit 20. A pattern detection range for detecting a pattern is set. Such a pattern detection range can be set in an arbitrary manner. For example, in the present embodiment, the storage unit 20 stores information 48 such as the shape and size of the pattern detection range as a reference, and the relative position with respect to the face image, and the range setting unit 22 stores the information. 48, a rectangular pattern detection range in which the length in the vertical direction and the horizontal direction of the face image extracted by the face extraction unit 18 is the length of the vertical side and the horizontal side, respectively, It is set in the vicinity of the determined face image. The present invention is not limited to this. Information 50 relating to the set pattern detection range, for example, in this embodiment, a two-dimensional coordinate group indicating a region in the captured image is supplied to the pattern identification unit 24.

  The pattern identification unit 24 is supplied with information 50 on the pattern detection range determined by the range setting unit 22 and the captured image data 34 created by the imaging unit 16, and the range setting unit 22 in the captured image data 34 It is determined whether or not the same pattern as the pattern registered in advance in the storage unit 26 is detected within the set pattern detection range. It is possible to adopt a known method for detecting the pattern. For example, in this embodiment, the comparison evaluation value is calculated by comparing the image feature amount of the pattern registered in the storage unit 26 in advance with the image feature amount appearing in the pattern detection range, and the calculated comparison evaluation value has a threshold value. If it is above, it is determined that they match. The present invention is not limited to this. When it is determined that they match, that is, when the same pattern is detected, the pattern identification unit 24 supplies a signal 54 informing the control unit 14 of that fact.

  The storage unit 26 is a storage circuit that stores registration information 38 used in the collation performed by the face identification unit 20 and the pattern identification unit 24, and information related to the pattern, that is, the pattern information 52. In this embodiment, the registration information 38 is feature amount data indicating the feature amount of the face image 42 of the owner of the photographing apparatus 10 as shown in FIG. 2, and the pattern information 52 is the right hand of the owner of the photographing apparatus 10 as shown in FIG. The image feature amount data of the image 60 is stored. In FIG. 3, since it is a photographed image, the left and right are reversed. In addition to these data, any data other than these data can be stored in the storage unit 26. For example, in this embodiment, the feature amount data of the face image of the instructor and the image registered as the pattern are stored. In addition to the image feature amount data, the shape and size of the reference pattern detection range, and the relative position to the face image are stored, and are used when setting the pattern detection range by the range setting unit 22 Yes.

  The display unit 28 is controlled by the control unit 14 and includes, for example, a display device such as a display. The display unit 28 outputs information stored in the storage unit 26, a photographed image created by the photographing unit 16, a setting screen, and the like to the display. To do.

  The imaging device 10 having the above-described configuration corresponds to a device including the imaging unit 16 and capable of imaging a subject. For example, a digital camera, an image input device, a movie camera, or a mobile phone provided with a camera However, the present invention is not limited to these. In the present invention, the configuration of each unit is not limited to the present embodiment, and an arbitrary configuration can be adopted according to the photographing apparatus 10.

  The pattern registered in the storage unit will be described with reference to FIGS. In the drawings, the same reference numerals indicate similar components. In the present invention, a face image of the instructor is detected from a photographed image created by photographing, and a pattern detection range for detecting a pattern is set with reference to the position of the detected face image. Therefore, the pattern is preferably a pattern that can be reproduced in the vicinity of the instructor's body, and more preferably a pattern that can be reproduced in the vicinity of the instructor's face. By creating a pattern that can be reproduced in the vicinity of the instructor's body or face in this way, for example, a person other than the instructor who is shown together has the same thing registered as the pattern, or is registered as a pattern. Even when the same action is performed, it is possible to prevent detection of this as a pattern, and it is possible to prevent malfunction.

  The vicinity of the body means, for example, a horizontal length 86 from the fingertip of the right hand 82 to the fingertip of the left hand 84 when the person 80 extends the arm in the horizontal direction as shown in FIG. When considering a rectangle 96 with the vertical length 92 from the fingertip 88 to the foot 90 when stretched in the direction, the length of the vertical side and the horizontal side, respectively, this rectangle 96 and the center 98 are the same and this rectangle It is preferably within the range of the rectangle 100 having a size 1.5 times that of 96, and more preferably within the range of the rectangle 96. Further, the vicinity of the face is, for example, as shown in FIG. 5, when a rectangle 114 having a vertical length 110 and a horizontal length 112 as the vertical side and horizontal side lengths, respectively, is considered. 114 and the center 116 are the same, and it is preferable to be within the range of a rectangle 118 that is three times the size of the rectangle 114, and further within the range of a rectangle 120 that is twice the size of the rectangle 114. More preferred. However, the present invention is not limited to these ranges, and the pattern may be a pattern that can be reproduced at an arbitrary position.

  Further, the pattern is not limited to a buoy sign made with the right hand as in this embodiment, and any pattern can be used. For example, an arbitrary shape made using a part of the body like a hand may be used as a pattern, and an arbitrary object or a state where the object is held or worn by a hand may be used as a pattern. Moreover, for example, an arbitrary operation may be used as a pattern.

  However, if too large objects or too large actions are registered as patterns, it will be necessary to set a large pattern detection range.For example, objects or actions that are not registered as patterns are detected as registered patterns. It is preferable that the pattern is not too large. In the present invention, since the face image is detected first, the size of the face image is detected together with the detection of the face image. Setting the pattern detection range using the size makes it possible to set the pattern detection range more easily. Therefore, the pattern can be reproduced in the vicinity of the body or face as described above, and the pattern detection range is not less than 1/2 and not more than 3 times the size of the face image, more preferably the pattern detection range is substantially In addition, it is preferable to use an object or an action that is the same size as the size of the face image or twice the size of the face image.

  Specifically, for example, those that are substantially the same size as the face, or slightly larger / smaller than the face, or substantially larger than the face, but worn in the vicinity of the face or body Patterns that can be held, or holding or wearing these items in the vicinity of the face or body, or moving these items in the vicinity of the face or body, or by these It is preferable that the pattern is an arbitrary shape or a sign made in the vicinity of the body.

  More specifically, for example, if it is a hand or an object, an arbitrary shape made by hand, or an article that can be held by hand such as a handkerchief, a ball, a fan, a fan, or the like can be given. In addition, for example, there are states such as a state where these items are held by hand, or a state where both hands are raised above the head and an OK sign is made. In addition, if it is a movement, a movement that shakes the hand, a movement that closes or opens the index finger and the middle finger in the case of a shape made by hand such as a buoy sign, a movement that shakes a ball, fan, fan, etc. Fried. Examples of things that can be held or worn near the face or body include a hat, a muffler, and an umbrella. Note that the present invention is not limited to these examples, and the pattern can be arbitrarily determined, and a pattern having a pattern detection range larger than the above-described size can be employed.

  The number of patterns to be registered does not have to be one. For example, it is possible to register a plurality of patterns. Further, a pattern for starting shooting and a pattern for ending shooting are output to the shooting apparatus 10. It is also possible to register patterns separately according to instructions. Note that the instruction to be issued to the photographing apparatus 10 is not limited to the instruction to start photographing and the instruction to end photographing, but other instructions, for example, an instruction to turn on the flash, and to improve the image quality of the created captured image. It may be a simple instruction, and can be arbitrarily set according to the photographing apparatus 10.

  When there are a plurality of instructions to be registered in the photographing apparatus 10, at least one of the instructor and the instruction pattern may be different for each instruction. Thus, for example, when registering a shooting start instruction pattern and a shooting end instruction pattern, one person can register two patterns for start and end, and for example, for start and end It is also possible to register the same pattern for the purpose and to make the person giving the start instruction different from the person giving the end instruction. Note that the present invention is not limited to this.

  Similar to the registration of the registration information 38, such a pattern registration can be performed at an arbitrary timing and using a known method. For example, in the case of registering a buoy sign made with the right hand of the owner of the photographing apparatus 10 as a pattern as in this embodiment, the face image is registered and the buoy sign made with his right hand is photographed to obtain FIG. It is possible to create a shot image that only contains the pattern shown in Fig. 5 and register it as a pattern for starting shooting. For example, you can show your face when you are performing a buoy sign. 6 to create a photographed image 130 as shown in FIG. 6 by taking a self-portrait, that is, by photographing himself or by a third party, registering a face image, and The image 130 is displayed on the display unit 28, and a part to be patterned with a cursor or a touch panel method, that is, a buoy sign part in this embodiment is selected in a rectangular 132 shape. It is also possible to register by. The present invention is not limited to these methods.

  When registering a pattern, it is also possible to register the shape, position, size, etc. of the pattern detection range required by the range setting unit 22 using an arbitrary method. For example, when registering a pattern, the size of the registered pattern compared to the size of the face and how far away from the center of the face is reproduced to the instructor via the operation unit 12. For example, in the case of registering a pattern using a photographed image 130 in which the pattern and the face of the instructor are shown together as shown in FIG. 6, the cursor is to be registered as a pattern. The size and position of the range specified by the above and the size and position of the face image of the instructor are compared, and the shape, relative position, and relative size of the pattern detection range are calculated. Alternatively, for example, the shape, position, size, and the like of the pattern detection range may be registered in advance in the storage unit 26, and the patterns registered by the instructor may be limited. The present invention is not limited to these methods.

  When registering the shape, position, size, etc. of the pattern detection range, the shape, position, and size used as the reference of the pattern detection range are registered in the pattern registered in the storage unit 26 in order to prevent malfunction. It is preferable to determine accordingly. For example, in the case of registering a buoy sign made with the right hand as a pattern, as shown in FIG. 6, the pattern detection range is set to have a length 110 that is substantially the same as the length 110 in the vertical direction and the length 112 in the horizontal direction. It is preferable to register so that the upper right vertex of the rectangle 132 is positioned at the center 116 of the face image in the region of the rectangle 132 having the vertical side and the horizontal side. At this time, for example, it is preferable to display the rectangle 114 or the center 116 superimposed on the display unit 28 because it is easy to grasp the position of the pattern detection range to be registered, but the present invention is not limited to this. Absent.

  Similarly, for example, when an OK sign in which a large circle is formed on the head using arms and hands as shown in FIG. 7 is registered as a pattern, the position, shape, and size serving as a reference for the pattern detection range are set. An area of a rectangle 136 having a vertical side and a horizontal side that are substantially twice the vertical length 110 and the horizontal length 112 of the face image, and the center of the rectangle 136 is the center of the face image Although it is preferable to register so as to be located at 116, the present invention is not limited to this.

  FIG. 8 is a flowchart showing an example of a processing procedure for automatically taking an image in the photographing apparatus 10 shown in FIG. Note that, for example, even when giving an instruction to the photographing apparatus 10 to end photographing or when giving an instruction to strike the flash, for example, processing similar to that shown in FIG. Omitted. For example, when the photographing apparatus 10 is instructed to end photographing, the face of the person who instructs the photographing apparatus 10 to end photographing is detected, and the pattern that instructs the photographing apparatus 10 to end photographing is detected. It becomes like this.

  In the example shown in FIG. 8, the photographing apparatus 10 receives an instruction to drive in the automatic photographing mode from the operation unit 12, photographs a subject from an arbitrary place placed by the operator, and creates an image created by photographing. Determine whether or not the face of the instructor previously stored in the storage unit 26 is captured, and if so, instruct the photographing apparatus 10 to start photographing at a predetermined position near the face of the instructor. A pattern detection range is set, and when a pattern for giving a pre-registered shooting start instruction to the shooting apparatus 10, that is, a pattern for starting shooting is detected in this range, it is recognized as a trigger for shooting and shooting is performed. To start. This will be described in detail below.

  In FIG. 8, when an operation signal 30 indicating that driving in the automatic photographing mode is performed is input to the photographing apparatus 10 via the operation unit 12 by the operator, the control unit 14 of the photographing apparatus 10 performs photographing of the subject. A control signal 32 is supplied to the photographing unit 16, and the photographing unit 16 performs photographing to create photographed image data (step S10).

  At this time, the photographing apparatus is placed at an arbitrary position by the operator, and photographing is started from there to create photographed image data. For example, in this embodiment, the photographing device is placed at an arbitrary place several meters away from the operator 200 and the friend 202 as the subject, and photographs the place where the operator 200 and the friend 202 are lined up, for example, A captured image 206 as shown in FIG. 9 is created. FIG. 9 conceptually shows a photographed image created by photographing. Note that this photographing may be preliminary photographing that is not stored as one image in the storage unit 26, and the photographed image 206 is continuously created by the photographing unit 16 by driving in the preliminary photographing mode. Any one of the photographed image data 34 may be used. Further, the photographing apparatus 10 may display, for example, a through image of the image obtained by the preliminary photographing on the display unit 28.

  The photographed image data 34 created by photographing is sent to the face extraction unit 18. The face extraction unit 18 extracts face image information 36, that is, feature amount data indicating the feature amounts of the face images of the operator 200 and the friend 202 in the present embodiment from the captured image data 34 (step S12). In the present embodiment, the face extraction unit 18 also extracts the positions of the centers 208 and 210 of the face images in the captured image 206 and the sizes of the face images in the vertical and horizontal directions. In this embodiment, feature amount data of all face images captured in the photographed image 206 is extracted. However, the present invention is not limited to this, and the face for extracting feature amount data is identified. It is also possible to add conditions. At this time, the photographing apparatus 10 creates frames 212 and 214 as shown in FIG. 9 based on the size of the extracted face image in the center, vertical direction, and horizontal direction, and displays the frames 212 and 214 in a superimposed manner. May be.

  Returning to FIG. 8, thereafter, the face extraction unit 18 determines how many pieces of extracted face image information 36 are present (step S14). As a result, when the extracted face image information 36 is zero, that is, when a face image is not extracted from the photographed image 206, the photographing apparatus 10 uses at least one face image information using another photographed image obtained by photographing. Repeat the process until 36 is obtained. On the other hand, if the result of determination in step S14 is that there is one or more extracted face image information 36, the face extraction unit 18 sends all the extracted face image information 36 to the face identification unit 20.

  Note that the operator 200 can be set to repeat the processing until, for example, a certain number of faces or more are detected in the photographing apparatus 10. By setting in this way, for example, when taking a group photo, it is possible to prevent a situation in which most subjects are not suitable for the camera. The present invention is not limited to this.

  The face identification unit 20 collates the face image information 36 sent from the face extraction unit 18 with the registration information 38 registered in advance in the storage unit 26 (step S16), and stores in the captured image 206 in advance the storage unit It is determined whether or not the face image of the person registered in 26 exists (step S18). In the present embodiment, the face image 42 of the operator 200 who is the owner of the imaging device 10 is registered in advance in the storage unit 26 as an instructor that causes the photographing device to start photographing. Any method can be adopted for this determination. For example, in this embodiment, the feature amount data stored in the storage unit 26 matches the feature amount data obtained by the face extraction unit 18. If the degree is lower than a certain threshold, the face identification unit 20 determines that there is no person's face registered in the photographed image 206. Conversely, if the degree of coincidence is a certain threshold or more, the registered person's face is It is determined that the captured image 206 exists. The present invention is not limited to this.

  As a result of the determination in the face identification unit 20, when it is determined that there is no person's face registered in the photographed image 206, the photographing apparatus 10 returns to step S10, and the registered person's face is detected from another photographed image. The same processing is repeated until it is done. On the other hand, when it is determined that the registered person's face is in the captured image 206, the face identifying unit 20 notifies the range setting unit 22 of the fact. When the face identifying unit 20 determines that there is a person's face registered in the captured image 206, the range setting unit 22 determines the position and size of the facial image in the captured image 206 extracted by the face extracting unit 18. Utilizing this, a pattern detection range, that is, a pattern detection range is set in the vicinity of the face image determined to be an instructor (step S20).

  In the present embodiment, the pattern detection range is a rectangle whose vertical and horizontal sizes are the same size as the vertical and horizontal directions of the instructor's face image extracted by the face extraction unit 18, that is, a face image. The storage unit 26 stores that the rectangle has the same size as the surrounding frame. Further, as a relative position of the position of the pattern detection range with respect to the face image, the storage unit 26 stores that the center that can be calculated from the vertical and horizontal lengths of the face image is the upper right vertex of the pattern detection range. Yes. Therefore, the range setting unit 22 has a pattern detection range in which the area of the rectangle 220 having the same size as the frame 212 surrounding the face image and the upper right vertex passing through the center 208 of the face image is shown in FIG. Set as. At this time, the imaging apparatus 10 may display the set pattern detection range so as to be superimposed on the captured image 206 as shown in FIG.

  In this embodiment, the pattern detection range is set as a rectangular area 220 having the same size as the frame 212 and the upper right vertex passing through the center 208 of the face image. However, the present invention is not limited to this. The range setting unit 22 can set a pattern detection range in an arbitrary size and shape at an arbitrary position by adopting an arbitrary method. For example, the size, shape, relative position, and the like serving as a reference for the pattern detection range may be stored in the storage unit 26 in advance.

  For example, when there are a plurality of registered patterns, it is also possible to set a plurality of pattern detection ranges according to the registered patterns as shown in FIG. In the example shown in FIG. 11, when a buoy sign made with the right hand and an OK sign in which a large circle is formed on the head by the arm and the hand are registered as the pattern for starting shooting, the pattern detection ranges are shown. Rectangles 220 and 230 are set. For example, although not shown, when the friend 202 has also registered a face image and a pattern, the pattern detection range can be set for the friend 202 in the same manner. The present invention is not limited to this. For example, even when a plurality of patterns are registered, a pattern detection range corresponding to one pattern can be set.

  Since the position and size of the pattern to be captured change depending on the position at which the photograph is taken, the position and size that is the reference for the pattern detection range as in this embodiment is based on the position and size of the face image. When setting the pattern detection range, compared with the position and size of the face image extracted by the face extraction unit 18, the position of the pattern detection range is changed or the size is changed, that is, It is preferable to set by enlarging or reducing. As a result, it is possible to more appropriately detect a change in the pattern in the pattern detection range, and to prevent malfunction. It should be noted that what is changed is not limited to the position and size, but can be changed arbitrarily. For example, when a rectangle is used as a pattern detection range, the angle of the rectangle can be changed and set. Is also possible.

  Returning to FIG. 8, when the pattern detection range is set by the range setting unit 22, the pattern identification unit 24 determines whether or not a pattern exists in the set pattern detection range (step S22). Specifically, in the present embodiment, the pattern identification unit 24 calculates the image feature amount in the pattern detection range set by the range setting unit 22, that is, the region indicated by the rectangle 220, and stores the calculated image feature amount in advance. The comparison evaluation value is calculated by comparing the image feature quantity stored in the unit 26. In the example shown in FIG. 9, the friend 202 also has a buoy sign, but this buoy sign is not detected as a pattern because it is outside the pattern detection range as shown in FIG.

  After calculating the comparative evaluation value, the pattern identification unit 24 compares the comparative evaluation value with a certain threshold value, and determines that the registered pattern does not exist in the pattern detection range when the comparative evaluation value is lower than the certain threshold value. . In this case, the imaging apparatus 10 returns to step S10 and performs the same processing using another captured image obtained by the imaging unit 16. On the other hand, when the comparative evaluation value is equal to or greater than a certain threshold value, the pattern identification unit 24 determines that the registered pattern exists in the pattern detection range, and notifies the control unit 14 that the pattern has been detected.

  At this time, for example, when a pattern for giving an instruction to strike the flash to the photographing apparatus 10, that is, when a pattern for flash is also registered, the instructor 200 also sends the registered flash pattern to the photographing apparatus 10. It is also possible to present the image and direct the photographing device 10 to take a flash. In this case, it is possible to arbitrarily decide whether to instruct the photographing apparatus 10 to use one pattern for flashing and to start photographing, or to use two patterns for flash and photographing start. Is possible.

  Further, for example, when a plurality of patterns are registered and a plurality of pattern detection ranges are set, the pattern identification unit 24 determines whether or not the pattern is registered in advance for each pattern detection range. In this case, it is possible to arbitrarily set the number of pattern detection ranges in which the comparative evaluation value is equal to or greater than the threshold, and the pre-registered pattern can be detected. .

  For example, in the case where all the subjects have registered their faces and patterns, the pattern identification unit 24 is set so that when the pattern identification unit 24 detects the pattern of all members, the fact is notified to the control unit 14. Is also possible. In this way, all the subjects can be designated as instructors for shooting, so that it is possible to prevent a failure such as closing the eyes because the timing is not met. The present invention is not limited to these.

  When the pattern registered in the storage unit 26 in advance is detected from the pattern identification unit, the control unit 14 controls the start of shooting using that as a trigger, that is, the shooting unit 16 opens the desired shutter. The main shooting, which is shooting for storing the subject image as one image in the storage unit 26, is started (step S24). For example, in the present embodiment, the control unit 14 turns on a light emitting unit (not shown) provided in the vicinity of the optical system to notify the subject that automatic shooting has started, and after a certain period of time has elapsed, the light emitting unit blinks and the image is taken. The unit 16 is caused to release the shutter and perform the actual shooting. Note that the present invention is not limited to this, and the control unit 14 informs the subject of the timing at which the shooting unit 16 releases the shutter, for example, by voice, and causes the shooting unit 16 to perform main shooting after a certain period of time. Alternatively, for example, the pattern identifying unit 24 may cause the photographing unit 16 to release the shutter at substantially the same time as determining that the registered pattern has been recognized, and may perform actual photographing. Can be adopted.

  When the photographing unit 16 performs the main photographing, the control unit 14 continues or ends the driving in the automatic photographing mode in a known manner. For example, the control unit 14 may continuously cause the photographing unit 16 to perform main photographing until an end signal is input from the operation unit 12, or for example, every time one photographing is performed, the process returns to step S10 and the processing is performed again. The driving in the automatic photographing mode may be terminated every time one image is photographed.

  In addition, a pattern for instructing the photographing apparatus 10 to end photographing, that is, a pattern for ending photographing is registered in the storage unit 26 in advance, for example, in the same manner as reproducing the pattern for photographing start, the instructor 200 Alternatively, the friend 202 may notify the photographing apparatus 10 of the end of photographing by presenting the photographing end pattern to the photographing apparatus 10. The present invention is not limited to these examples.

  The driving in the automatic photographing mode may be parallel to the driving in the automatic focus adjustment mode for automatically focusing, the automatic exposure adjustment mode for automatically exposing, or the like. When a signal for driving in automatic shooting mode and starting automatic shooting is sent from the operation unit 12 to the control unit 14, driving is performed in an automatic focus adjustment mode, an automatic exposure adjustment mode in which exposure is automatically performed, or the like. Any method may be employed.

  As described above, this embodiment recognizes a face image, sets a pattern detection range near the recognized face image, and triggers when a pattern registered in advance in the set pattern detection range is detected. As described above, the photographing apparatus is controlled, that is, the photographing apparatus starts processing designated by the pattern. Therefore, wherever the instructor stands, it is possible to accurately specify the position and accurately set the pattern detection range, and it is possible to accurately detect the registered pattern It is possible to prevent malfunction. For example, when there are two or more subjects, when a shooting start instruction is given, even if a person who is not registered as an instructor takes a pose registered as a pattern, the shutter may be accidentally released. No.

  In addition, by making things that can appear in the vicinity of the face, patterns that can be expressed in the vicinity of the face, and things that can be worn in the vicinity of the face, It is possible to set the pattern detection range more accurately, and it is possible to prevent malfunction of the photographing apparatus. Therefore, it is possible to provide a photographing apparatus and a photographing method that are more convenient for the user.

1 is a block diagram schematically showing a photographing apparatus according to the present invention. It is the figure which showed notionally the registration information memorize | stored in the memory | storage part of the imaging device shown in FIG. It is the figure which showed notionally the pattern memorize | stored in the memory | storage part of the imaging device shown in FIG. It is the figure which showed notionally an example of the range which can register a pattern. It is the figure which showed notionally another example of the range which can register a pattern. It is the figure which showed notionally an example of the pattern and the pattern detection range. It is the figure which showed notionally another example of the pattern and the pattern detection range. It is the flowchart which showed an example of the automatic imaging | photography process sequence in the imaging device shown in FIG. It is the figure which showed notionally an example of the image produced in the imaging | photography part of the imaging device shown in FIG. It is the figure which showed notionally an example at the time of setting the pattern detection range in the imaging device shown in FIG. It is the figure which showed notionally an example at the time of setting the pattern detection range in the imaging device shown in FIG.

Explanation of symbols

10 Shooting device
12 Operation unit
14 Memory
16 Shooting section
18 Face extractor

Claims (18)

In a photographing apparatus including photographing means for photographing a subject and creating an image, the apparatus includes:
Storage means for storing first face image information relating to a face image of a person giving an instruction to the photographing apparatus, and instruction pattern information relating to an instruction pattern giving an instruction to the photographing apparatus;
Face image extraction means for extracting second face image information relating to a face image reflected in a photographed image created by the photographing means;
Range setting means for comparing the first face image information and the second face image information, and setting a detection range for detecting the instruction pattern according to the result of the comparison;
Identification means for identifying whether or not the instruction pattern exists in the detection range,
The imaging apparatus performs the process specified by the instruction pattern when the instruction pattern is identified as being present in the detection range. The apparatus according to claim 1, wherein the instruction pattern is a pattern that gives an instruction to start photographing to the photographing apparatus,
The imaging apparatus starts imaging when it is identified that the instruction pattern exists in the detection range. The apparatus according to claim 1, wherein the instruction pattern is a pattern for giving an instruction to end photographing to the photographing apparatus,
The imaging apparatus terminates imaging when the instruction pattern is identified as being present in the detection range.   4. The photographing apparatus according to claim 1, wherein the instruction pattern is a pattern that can be reproduced in the vicinity of a person who causes the photographing apparatus to perform photographing.   4. The photographing apparatus according to claim 1, wherein the instruction pattern is a pattern that can be reproduced in the vicinity of a face of a person who causes the photographing apparatus to perform photographing.   6. The apparatus according to claim 1, wherein the storage unit further stores a position and a size of the detection range, and the range setting unit detects the detection based on the position and the size. An imaging apparatus characterized by setting a range.   The apparatus according to claim 6, wherein the position and size of the detection range are determined based on a position and a size of a face of a person who causes the imaging apparatus to perform imaging.   8. The apparatus according to claim 6, wherein the face image extraction unit further extracts a position of a face image shown in the photographed image, and the range setting unit determines the position of the detection range based on the position. An imaging apparatus characterized by adjusting and setting the camera.   9. The apparatus according to claim 6, wherein the face image extracting unit further extracts a size of a face image shown in the photographed image, and the range setting unit is based on the size. An imaging apparatus, wherein the size of a detection range is adjusted and set. A method for causing a photographing device including a photographing unit that photographs a subject to create an image to start photographing, the method comprising: a photographing step of photographing the subject with the photographing device to create a photographed image;
A face image detecting step of detecting a face image of a person who gives an instruction to the image capturing device registered in advance in the image capturing device from the captured image;
A range setting step for setting a detection range for detecting an instruction pattern for giving an instruction to the photographing apparatus according to a result of the face image detection step;
An identification step for identifying whether or not the instruction pattern exists in the detection range;
And a process execution step of causing the imaging apparatus to execute a process designated by the instruction pattern when the instruction pattern exists in the detection range. The method according to claim 10, wherein the instruction pattern is a pattern that gives an instruction to start photographing to the photographing apparatus,
In the photographing step, when the instruction pattern is present in the detection range, photographing is started on the basis of the instruction pattern. The method according to claim 10, wherein the instruction pattern is a pattern that gives an instruction to the photographing apparatus to end photographing,
In the photographing step, when the instruction pattern is present in the detection range, photographing is terminated when the instruction pattern is present.   13. The imaging method according to claim 10, wherein the instruction pattern is a pattern that can be reproduced in the vicinity of a person who gives an instruction to the imaging apparatus.   13. The imaging method according to claim 10, wherein the instruction pattern is a pattern that can be reproduced in the vicinity of a face of a person who gives an instruction to the imaging apparatus. The method according to any one of claims 10 to 14, further comprising a registration step of registering a position and a size of the detection range in the imaging apparatus.
The imaging method according to claim 1, wherein the range setting means sets the detection range based on the registered position and size.   16. The method according to claim 15, wherein the position and size of the detection range are determined and registered based on the position and size of a person giving an instruction to the imaging apparatus. Method.   The method according to claim 15 or 16, wherein the face image detection step further extracts a position of a face image shown in the photographed image, and the range setting step includes the detection range based on the position. A shooting method characterized by adjusting and setting a position.   The method according to any one of claims 15 to 17, wherein the face image detecting step further extracts a size of a face image shown in the photographed image, and the range setting step is based on the position. An imaging method, wherein the size of the detection range is adjusted and set.

Images