JP4605039B2 - Automatic photographing method, automatic photographing apparatus and automatic photographing program - Google Patents

Description

  The present invention relates to an automatic photographing method, an automatic photographing apparatus, and an automatic photographing program, and more particularly to an automatic photographing method, an automatic photographing apparatus, and an automatic photographing program that automatically perform photographing when a specific state of a subject is changed.

  In a photographing apparatus such as a digital camera or a film camera, it is important how to photograph a subject as intended by the photographer. Therefore, if the subject is a person, the direction of the person's face is recognized and recorded when the face turns to a predetermined direction or when the facial expression of the subject is recognized and the face becomes a predetermined expression. 2. Description of the Related Art Conventionally, a photographing apparatus that performs an operation is known (see, for example, Patent Document 1). That is, in the conventional photographing apparatus described in Patent Document 1, the image information of the object scene is detected, and the facial expression recognition information obtained by recognizing the facial expression of the subject based on the image information is used to obtain a predetermined facial expression. The subject recording operation is performed when the result of the determination is a predetermined facial expression. Thereby, the person of the subject can be photographed with a natural expression as much as possible.

  There is also known an imaging apparatus that identifies and captures a desired person (for example, see Patent Document 2). The conventional photographing apparatus includes an extraction unit that extracts predetermined feature points from image data, a reception unit that receives instructions from a user, and an instruction received by the reception unit when a plurality of feature points are extracted. It is a digital camera having a configuration including selection means for selecting feature parts in a predetermined order and display means for displaying feature part information for specifying the feature parts selected by the selection means.

  Furthermore, an automatic photographing apparatus that automatically photographs a scene image of a specific subject with a specific posture is also known (see, for example, Patent Document 3). In this conventional automatic photographing apparatus, a specific posture of a specific subject is used as a photographing condition, subject information including position information and posture-related information of the specific subject, photographing position information and photographing of a photographing means for photographing the specific subject. Shooting information including direction information is acquired, and based on the shooting information and subject information, posture information of a specific subject with respect to a shooting scene range in a real space that can be shot by the shooting unit is detected, and according to the detected posture information Thus, the photographing operation of the photographing means is controlled to cause the photographing means to photograph a scene image corresponding to the photographing condition.

Japanese Patent Laid-Open No. 2001-051338 JP 2004-320287 A JP 2005-286394 A

  However, in the conventional photographing apparatuses described in Patent Documents 1 and 2 described above, it is possible to recognize the feature points of the face of the person who is the subject and to capture an image of the moment when the person smiles. Since the shooting device accumulates still images in a time series like a movie, saves the image after recognizing it as a smile, and discards other images, the flash cannot be turned on at all times. In some cases, the image of the moment you smiled is not lit, and the background may remain dark.

  In addition, in a system in which only a characteristic image is stored after being photographed several times, an irreversible reaction photographing apparatus in which the film cannot be reused once the film is exposed, such as a film camera. In this case, the best photographed image is selected from the images, and in this case, the film is considerably wasted.

  Further, in the conventional photographing apparatuses described in Patent Documents 1 and 2, since the motion of the subject and the movement of the face cannot be predicted, for example, when a family takes a souvenir picture around the infant, Even if you try to take a picture that matches the direction of the camera, you cannot predict when the infant will face the camera. In many cases, it is difficult for the entire family, including infants, to face the camera.

  On the other hand, in the conventional automatic photographing apparatus described in Patent Document 3, it is possible to predict a specific posture of the subject. However, when a family takes a commemorative photo around the infant, It is difficult to shoot scenes where the whole family, including the family, is smiling, and it is difficult to capture the timing. is there.

  The present invention has been made in view of the above points, and predicts a transition of a specific feature point such as a facial expression of a subject, and notifies a subject of shooting, thereby triggering a specific state change of the specific subject. It is an object of the present invention to provide an automatic photographing method, an automatic photographing apparatus, and an automatic photographing program that can automatically perform photographing intended by a photographer.

  In order to achieve the above object, an automatic photographing method of the present invention includes a first step of designating a desired photographing pattern of a subject and a photographing timing, and an analysis image of the subject from the subject image before photographing the main image. A second step of acquiring, a third step of extracting a feature point image from the analysis image, and a time-series change of the feature point image based on the feature point motion analysis information from the feature point motion analysis information knowledge A fourth step of predicting a time point when the image of the feature point substantially matches the desired shooting pattern, and a fifth step of starting the main image shooting according to the shooting timing after the prediction time point in the fourth step It is characterized by including.

  In the present invention, an image of a feature point is extracted from an analysis image of a subject before photographing a main image, and a time-series change of the image of the feature point is based on feature point motion analysis information from feature point motion analysis information knowledge. Analyzing and predicting the time point when the image of the feature point substantially matches the desired shooting pattern, and starting the main image shooting according to the shooting timing after the prediction time point, the image at the moment when the desired shooting pattern is obtained Can be taken. In other words, according to the present invention, when the movement of the feature point of the subject is recognized and the state is changed to a state intended for photographing exceeding a certain threshold, the main image photographing can be activated.

  In order to achieve the above object, the automatic photographing method of the present invention includes a sixth step of storing the analysis image, and reproducing the stored analysis image to obtain a desired one from the analysis image. A seventh step of selecting an analysis image indicating a photographing pattern, and a feature of changing the feature point motion analysis information stored in the feature point motion analysis information knowledge based on the analysis image selected in the seventh step And further comprising 8 steps.

  In the present invention, since the analysis image showing the desired shooting pattern is selected from the analysis images and the content of the feature point motion analysis information knowledge is changed, the feature point motion analysis of the fourth step is performed. Can be corrected.

  In order to achieve the above object, in the fourth step, the automatic photographing method of the present invention, when the point of time when the feature point image substantially matches the desired photographing pattern cannot be predicted within the set time in the fourth step. A ninth step of designating a new image as a desired photographing pattern, and a tenth step of changing the new image designated in the ninth step as an image of a desired photographing pattern in the feature point motion analysis information knowledge The method further includes a step.

  In this invention, when the prediction of the point in time when the feature point image substantially matches the desired shooting pattern cannot be made within the set time, a new image is designated as the desired shooting pattern, and the feature point motion analysis information knowledge Therefore, based on the image of the shooting pattern intended by the photographer, the feature point motion analysis of the image for analyzing the subject can be accurately performed.

  In order to achieve the above object, in the automatic photographing method of the present invention, the fifth step includes a function of notifying the subject of the start of photographing immediately before the photographing timing. According to the present invention, since the shooting start notice is performed immediately before the shooting timing, intended shooting can be performed on a person who is conscious of shooting.

  Here, when the subject is a person, the shooting pattern includes at least one of a specific facial expression and a face direction of the person.

  In order to achieve the above object, the automatic photographing apparatus of the present invention includes a specifying unit for designating a desired photographing pattern and photographing timing of a subject, and an analysis image of the subject from the subject image before photographing the main image. Analysis image acquisition means for acquiring, feature point extraction means for extracting an image of a feature point from the analysis image, and feature point motion analysis information for analyzing the motion of the image based on the image of the feature point are stored in advance. Analyzes feature point motion analysis information knowledge and time-series changes of feature point images extracted by feature point extraction means based on feature point motion analysis information from feature point motion analysis information knowledge, and images of the feature points Feature point motion analysis means for predicting a time point at which the image substantially matches a desired shooting pattern, and main image shooting start means for starting main image shooting according to the shooting timing after the prediction time point by the feature point motion analysis means Characterized in that it has a.

  In the present invention, an image of a feature point is extracted from an analysis image of a subject before photographing a main image, and a time-series change of the image of the feature point is based on feature point motion analysis information from feature point motion analysis information knowledge. Analyzing and predicting the time point when the image of the feature point substantially matches the desired shooting pattern, and starting the main image shooting according to the shooting timing after the prediction time point, the image at the moment when the desired shooting pattern is obtained Can be taken.

  In order to achieve the above object, the automatic photographing apparatus of the present invention includes an analysis image storage unit that stores an analysis image in a storage region, a playback unit that plays back the analysis image stored in the storage region, Selection means for selecting an analysis image showing a desired shooting pattern from among the analysis images reproduced by the reproduction means, and feature point motion analysis information knowledge storage feature points based on the selected analysis image It further has characteristic operation correction means for changing the operation analysis information. In this invention, since the analysis image showing the desired shooting pattern is selected from the analysis images and the content of the feature point motion analysis information knowledge is changed, the feature point motion analysis by the feature point motion analysis means is performed. Can be corrected.

  In order to achieve the above object, the automatic photographing apparatus according to the present invention is such that the feature point motion analysis means cannot predict the point in time when the feature point image substantially matches the desired photographing pattern within the set time. Includes an image designating unit for designating a new image as a desired shooting pattern, and a feature point motion analysis for changing the new image designated by the image designating unit as an image of a desired shooting pattern in the feature point motion analysis information knowledge And an information changing means.

  In this invention, when the prediction of the point in time when the feature point image substantially matches the desired shooting pattern cannot be made within the set time, a new image is designated as the desired shooting pattern, and the feature point motion analysis information knowledge Therefore, based on the image of the shooting pattern intended by the photographer, the feature point motion analysis of the image for analyzing the subject can be accurately performed.

  In order to achieve the above object, the automatic photographing apparatus of the present invention further includes a queue output means for notifying the subject of the start of photographing immediately before starting the main image photographing by the main image photographing starting means. . In the present invention, the subject can be notified of the start of shooting. Here, when the subject is a person, the shooting pattern includes at least one of a specific facial expression and a face direction of the person.

  In order to achieve the above object, the present invention provides an automatic photographing program for causing a computer to sequentially execute the first to fifth steps described in claim 1, and the computer performs the sixth to eighth steps described in claim 2. In addition, an automatic photographing program is also included.

  According to the present invention, an image of a feature point is extracted from an analysis image of a subject before photographing the main image, and the time-series change of the image of the feature point is converted into feature point motion analysis information from the feature point motion analysis information knowledge. Based on the analysis, predicting the time point when the image of the feature point substantially matches the desired shooting pattern, and starting the main image shooting according to the shooting timing after the prediction time point, the moment when the desired shooting pattern is obtained Images can be taken, so the infant who doesn't understand the language or understands it but doesn't move as expected, takes a picture of the desired shooting pattern (for example, a smile) and is conscious of being shot Effective filming, such as filming the desired shooting pattern of a person who has not been photographed, and the pose that a person who is aware of the surrounding area wants to shoot at the same time, is performed without wasting film. That.

  Next, embodiments of the present invention will be described with reference to the drawings. FIG. 1 shows a block diagram of an embodiment of an automatic photographing apparatus according to the present invention. As shown in the figure, in the present embodiment, the operation unit 10 is provided with a photographing pattern designation unit 11, a photographing reservation instruction unit 12, and a feature point designation unit 13. The shooting pattern specifying means 11 specifies shooting conditions by the photographer. The shooting reservation instruction means 12 issues a reservation instruction for shooting conditions and the like. The feature point designating unit 13 designates a feature point (image) desired by the photographer.

  As shown in FIG. 2, for example, as shown in FIG. 2, the operation unit 10 that constitutes the shooting pattern specifying unit 11, the shooting reservation specifying unit 12, and the feature point specifying unit 13 is a shooting unit such as a digital camera. From the preview window 31 of the apparatus main body, the menu key 32, the enter key 33, the cancel key 34, and the cross key 35 for specifying the menu items displayed on the preview window 31 by moving the cursor up / down / left / right. Composed.

  Referring back to FIG. 1 again, the present embodiment is based on an analysis image acquisition unit 15, a feature point extraction unit 16 that extracts a feature point of a subject, and an acquired feature point for actual image shooting. When the feature point motion analysis means 17 for analyzing the motion, the feature point motion analysis information knowledge 18 which is a database storing knowledge for the feature point motion analysis means 17 to analyze the feature point motion, and the conditions are met The image capturing start unit 19 for starting image capturing, the analysis image storing unit 23 for storing the analysis image acquired by the analysis image acquiring unit 15, and the analysis for storing the analysis image by the analysis image storing unit 23. And an image storage area 24.

  Further, as a means for starting shooting, for example, a cue output means 20 for outputting a cue for recognizing that a subject is shot, a flash lighting means 21 for lighting a flash as necessary for shooting, An image capturing unit 22 for capturing a target image, and an image reproducing unit for analysis 25 and an image reproducing unit for analysis as correction units for increasing accuracy for recognizing a match with the image capturing condition. And feature operation correction means 26 for changing the stored information of the feature point motion analysis information knowledge 18 to correct the feature motion based on the playback image selected from the playback images of the means 25. In the means 26, for example, the feature operation correction condition is set by the feature point specifying means 13 and the knowledge changing means 14.

  This embodiment is provided as one function of a digital camera, or is provided as an apparatus for controlling shutter timing of a digital camera or a film camera. When provided as one function of a digital camera, the mechanism of the digital camera is used as it is for lens and image acquisition, and the analysis image is analyzed from the image displayed in the preview window 31 as shown in FIG. The image acquisition means 15 acquires and processes. In addition, the photographing activation unit 19 activates a digital shutter function, and the digital camera body stores the photographed image.

  Further, when this embodiment is incorporated in an optical camera or the like as a device for controlling the shutter timing, it is used as a device for pressing the shutter of the optical camera, for example. In this case, an image photographed by the camera body is taken in by the analysis image acquisition means 15 and triggers a mechanism by which the photographing activation means 19 presses the shutter.

  Next, an automatic photographing method according to the present invention will be described. FIG. 3 shows a flowchart of an embodiment of the automatic photographing method according to the present invention. The case where a smiling girl is photographed by a self-timer together with other persons including the photographer will be described as an example. First, the photographer uses the photographing pattern designating means 11 on the condition that the girl in the subject is smiling. Designate (step S1).

  Specifically, the photographer presses the menu key 32 shown in FIG. 2, and thereby the menu item (this is selected from the plurality of selectable menu items displayed in the preview window 31). In this specification, “smile shutter” is selected using the up / down key 35 and then the enter key 33 is pressed.

  Then, the subject image shown in FIG. 4A is displayed in the preview window 31 as a subject image to be photographed, and a portion having a possibility of a face (a portion having an eye in a round shape) ) 41, 42 and 43 are highlighted. Note that the authentication method for a potential face can be applied by applying the existing face authentication / expression authentication technology, and since autofocusing is possible, the distance to the subject can be estimated. It is possible, and the face is roughly elliptical, and pattern matching is performed with the information that the eyes are on the left and right in the middle, and the size of the human face corresponding to the distance to the subject is imaged Is compared with a threshold value set in advance as a device, it is possible to select which ellipse is highly likely to be a face.

  Subsequently, the photographer selects the highlight portion 41 as shown in FIG. 4B by using the cross key 35 shown in FIG. 2 among the subject image and the highlight portions 41, 42 and 43 described above. In addition, the highlight portions 42 and 43 are deselected as indicated by 42 ′ and 43 ′, and the enter key 33 is pressed. When the menu key 32 is pressed, as shown in FIG. 4C, only the highlight portion 41 of the face of the smiling girl desired by the photographer is selected as the target.

  Next, the photographer instructs the photographing reservation instructing means 12 to, for example, take a picture when conditions (the girl becomes smiling) between 10 seconds and 30 seconds later, Get to the side of the girl within 10 seconds. Specifically, the photographer presses the menu key 32 shown in FIG. 2, and thereby, from among a plurality of selectable menu items displayed in the preview window 31, the time setting menu item (in this embodiment) ( When this is selected using the cross key 35 and the enter key 33 is pressed, the smile shutter time setting is set as shown in FIG. A screen is displayed in the preview window 31. Therefore, move the cursor with the left and right keys of the cross key 35, set a number with the up and down keys of the cross key 35 (disclose time is 10 and end time is 30), and finally "set" at the bottom of the screen By placing the cursor on "End" and pressing the enter key 33, the shooting reservation instruction in step S2 is performed.

  Subsequently, the analysis image acquisition unit 15 acquires an analysis image of the face of the girl designated by the shooting pattern designation unit 11 and stores it in the analysis image storage area 24 of the memory via the analysis image storage unit 23. At the same time, the image for analysis is supplied to the feature point extraction unit 16, and the feature point extraction unit 16 extracts feature points such as the eyes, corners of the eyes, and the corners of the mouth from the analysis image (step S3). Next, the feature point motion analysis unit 17 analyzes the change of the feature point image extracted by the feature point extraction unit 16 based on the feature point motion analysis information input from the feature point motion analysis information knowledge 18, Predictive analysis is performed as to whether or not a smile corresponding to the designated photographing pattern is obtained (step S4).

  Subsequently, the feature point motion analysis means 17 monitors whether or not the feature point motion analysis result is an analysis result indicating that the specified shooting pattern is likely to be a smile (step S5), and the feature point motion analysis unit 17 is likely to smile. When a feature point operation analysis result is obtained, the imaging activation means 19 is called. As a result, for example, the shooting activation unit 19 controls the cue output unit 20 to notify the photographer or the subject that shooting is about to be performed, or the flash lighting unit 21 flashes as necessary. The image capturing unit 22 is caused to capture a subject image (step S6).

  As described above, in this embodiment, since the desired subject's face is not automatically photographed after being smiled, but is automatically photographed by predicting the time point at which the face will be smiling, It is possible to photograph a subject image with a natural expression that is not conscious of the photographing apparatus immediately before or just after the smile.

  Therefore, for example, when a family takes a commemorative photo around an infant, it is impossible to predict when the infant will face the camera. On the other hand, in order to make the infant laugh, According to the present embodiment, an infant who is not conscious of photographing can be directed to the pattern of facing the camera as a photographing pattern and the photographing target is an infant. Since the shooting can be performed in accordance with the timing when the camera is directed, and immediately before the shooting timing, the cue output means 20 notifies the family who is conscious of shooting that shooting will be performed soon. Everyone can instantly smile and take pictures with the infant facing the camera. At the time of shooting, the flash is turned on as necessary, so that the shot image can be shot clearly even when the surroundings are dark. Furthermore, in the case of a film camera, since continuous shooting is unnecessary, the film is not wasted.

  Next, another embodiment of the present invention will be described. In the present embodiment, the accuracy is improved by the characteristic operation correcting means 26 of FIG. In the above embodiment, it is specified that the image is taken when the girl smiles within 10 to 30 seconds, but the feature point motion analysis means 17 does not recognize the smile within the time, but actually In this embodiment, correction is performed when the girl is smiling.

  The operation of this embodiment will be described with reference to FIGS. In this embodiment, the feature point motion analysis processing in step S4 of FIG. 3 is performed according to the flowchart shown in FIG. That is, the analysis image storage unit 23 in FIG. 1 stores the analysis image acquired by the analysis image acquisition unit 15 in the analysis image storage region 24 (step S41), and the analysis image reproduction unit 25 performs analysis. The analysis image stored in the image storage area 24 is reproduced and displayed in the preview window (step S42).

  The photographer designates a desired image (for example, a smiling image) of a desired subject from the analysis images displayed in the preview window by the feature point designating unit 13 in FIG. 1 (step S43). In response to this, the feature motion correcting unit 26 uses the knowledge changing unit 27 to change the feature point motion analysis information stored in the feature point motion analysis information knowledge 18. Thereby, according to the present embodiment, the feature point motion analysis unit 17 changes the image of the feature point image extracted by the feature point extraction unit 16 after the change input from the feature point motion analysis information knowledge 18. Analyzing based on the feature point motion analysis information, it is possible to predict and analyze the smiling face of the girl that is the designated shooting pattern with higher accuracy.

  Next, still another embodiment of the present invention will be described. In the embodiment of FIG. 3, the processing of steps S4 and S5 is repeated until the feature point motion analysis result is likely to become the designated shooting pattern. In this embodiment, for example, a certain time is exceeded. If the specified shooting pattern is not obtained, the feature point motion analysis in step S4 is stopped, and the feature motion correction means 26 is used to feed back to the photographer that the current analysis was not effective. It is a thing.

  In this embodiment, if necessary, the feature point designating unit 13 is used to designate a desired feature point. Specifically, when the desired feature point is designated as a smile, the photographer presses the menu key 32 shown in FIG. 2, thereby selecting from among a plurality of selectable menu items displayed in the preview window 31. The menu item of this embodiment (this is referred to as “smile correction” in the present specification) is selected using the up / down key 35 and then pressed by pressing the enter key 33.

  Then, a smile correction screen as shown in FIG. 7 is displayed in the preview window 31. In the lower part of the smile correction screen, images 51-54 of smiles and facial expressions before and after the face are displayed in time series, and among these, the “normal playback”, “stop”, When the smile image 53 is selected by making full use of “frame advance playback” or the like, the selected smile image 53 is displayed large in the upper part of the screen.

  When the feature point is specified by the feature point specifying unit 13, the feature operation correcting unit 26 traces the previous images 52 and 51 in time series based on the smile image 53 in the case of smile correction in FIG. The feature point motion analysis information knowledge 18 is changed to such feature point motion analysis information by using the knowledge changing means 27, and the difference between the feature point motion analysis knowledge information stored in the feature point motion analysis information knowledge 18 until then is changed. Correct.

  Although the means are not explicitly described, if the reservation is 30 seconds and the conditions for shooting are not met within 30 seconds, shooting is stopped, or shooting is performed even if the conditions are not met. There are termination means such as.

  The present invention is not limited to the above embodiment. For example, when shooting under the condition that a plurality of subjects are facing the camera, the shooting pattern designating unit 11 selects a plurality of subjects. When the condition is set that the shooting is performed when the face is facing the camera or when it is sufficiently predicted that the camera is facing the camera, the shooting reservation instruction unit 12 captures the image when the condition is met after 30 seconds. In this case, it is easy to take an effective photograph when photographing a plurality of children.

  Also, the feature motion correction means 26 can be acquired and updated via a network or the like, for example, as additional information from the manufacturer, instead of a user operation for each shooting. It is also possible to take a group photo under the condition that everyone is open. In this case, the feature point extraction unit 16 recognizes the face from the analysis image by known face authentication / expression authentication, and searches for a portion corresponding to the eye in the face (exists on the left and right in the vicinity of the center of the ellipse). 1), feature points such as each of the eyes and the corners of the eyes are extracted, and the feature point motion analysis means 17 recognizes from the feature points whether a pupil exists between the eyes and the corners of the eyes (the eyes, white portions, and not white) In the pattern that cannot recognize the head and the corner of the eyes with sunglasses (such as the part (pupil), the white part, and the corner of the eye), sunglasses, etc., the face is not considered as an object whose eyes are open. When the eyes are opened (whether the eyes are visible with respect to the face that can be recognized as the face and the eyes and the corners of the eyes), the image photographing means 22 is activated via the photographing activation means 19 at that moment. .

  Furthermore, according to the present invention, image capturing at the startle of the subject who opened the surprise box, photography of the moment when it was not a movement of the face, but a beautiful pose with extended hands, such as sports, Predictive photography is possible, such as taking a photograph under the expectation that Mabataki will not occur immediately after opening the eyes with Mabataki.

  Furthermore, the present invention can be used as a photographing aid for a mobile phone with a camera, a digital camera, and an optical photographer. It is also possible to apply the above-mentioned means.

  The present invention also includes a computer program that causes a computer to execute each means in FIG. 1 and each step in the flowcharts in FIGS. 3 and 6.

It is a block diagram of one embodiment of the device of the present invention. It is a block diagram of an example of the operation part in FIG. It is a flowchart of one embodiment of the method of the present invention. It is explanatory drawing of an example of the imaging | photography pattern designation | designated method by this invention. It is explanatory drawing of an example of the imaging | photography reservation instruction | indication by the imaging | photography reservation instruction | indication means in this invention. It is a flowchart of other embodiment of the method of this invention. It is a figure which shows an example of the screen of smile correction.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Operation part 11 Shooting pattern designation | designated means 12 Shooting reservation instruction | indication means 13 Feature point designation means 15 Analysis image acquisition means 16 Feature point extraction means 17 Feature point motion analysis means 18 Feature point action | operation analysis information knowledge 19 Shooting start means 20 Queue output means DESCRIPTION OF SYMBOLS 21 Flash lighting means 22 Main image photographing means 23 Analysis image storage means 24 Analysis image storage area 25 Analysis image reproduction means 26 Feature operation correction means 27 Knowledge change means 31 Preview window 32 Menu key 33 Determination key 34 Cancel key 35 Cross Key

Claims (12)

A first step of designating a desired shooting pattern of the subject and a shooting timing;
A second step of acquiring a subject analysis image from the subject image before the main image is captured;
A third step of extracting an image of feature points from the analysis image;
A time series change of the feature point image is analyzed based on the feature point motion analysis information from the feature point motion analysis information knowledge, and a time point at which the feature point image substantially matches the desired shooting pattern is predicted. 4 steps,
And a fifth step of activating main image shooting according to the shooting timing after the prediction time in the fourth step. A sixth step of storing the analysis image;
A seventh step of reproducing the stored analysis image and selecting an analysis image indicating the desired imaging pattern from the analysis image;
8. The method according to claim 1, further comprising: an eighth step of changing storage feature point motion analysis information of the feature point motion analysis information knowledge based on the analysis image selected in the seventh step. The automatic shooting method described. In the fourth step, when the prediction of the point in time when the feature point image substantially matches the desired shooting pattern cannot be made within the set time, a new image is designated as the desired shooting pattern. And the steps
The tenth step of changing the new image designated in the ninth step as an image of the desired shooting pattern in the feature point motion analysis information knowledge is further included. The automatic shooting method described.   The automatic photographing method according to claim 1, wherein the fifth step includes a function of notifying the subject of the start of photographing immediately before the photographing timing.   5. The automatic photographing method according to claim 1, wherein when the subject is a person, the photographing pattern includes at least one of a specific facial expression and a face direction of the person. A designation means for designating a desired shooting pattern of the subject and a shooting timing;
Analysis image acquisition means for acquiring an analysis image of the subject from the subject image before the main image is captured;
Feature point extraction means for extracting an image of a feature point from the analysis image;
Feature point motion analysis information knowledge in which feature point motion analysis information obtained by analyzing image motion based on feature point images is stored in advance;
The time-series change of the feature point image extracted by the feature point extraction means is analyzed based on the feature point motion analysis information from the feature point motion analysis information knowledge, and the feature point image is captured as the desired image. A feature point motion analysis means for predicting a time point substantially matching the pattern;
An automatic photographing apparatus comprising: main image photographing starting means for starting main image photographing according to the photographing timing after the prediction time point by the feature point operation analyzing means. Analysis image storage means for storing the analysis image in a storage area;
Playback means for playing back the analysis image stored in the storage area;
Selecting means for selecting an analysis image showing the desired photographing pattern from the analysis images reproduced by the reproduction means;
The automatic photographing apparatus according to claim 6, further comprising: feature motion correction means for changing the stored feature point motion analysis information of the feature point motion analysis information knowledge based on the selected analysis image. In the feature point motion analysis means, an image that designates a new image as the desired shooting pattern when the point of time when the image of the feature point substantially matches the desired shooting pattern cannot be predicted within a set time. Designation means;
7. A feature point motion analysis information changing unit for changing a new image designated by the image designating unit as an image of the desired shooting pattern in the feature point motion analysis information knowledge. Or the automatic imaging device of 7.   7. The automatic photographing apparatus according to claim 6, further comprising a cue output means for notifying the subject of the start of photographing immediately before starting the main image photographing start by the main image photographing activation means.   10. The automatic photographing apparatus according to claim 6, wherein the photographing pattern includes at least one of a specific facial expression and a face orientation of the person when the subject is a person.   An automatic photographing program which causes a computer to sequentially execute the first to fifth steps according to claim 1. An automatic photographing program which causes a computer to sequentially execute the sixth to eighth steps according to claim 2.

Images