TW201339986A - Method and device for capturing image - Google Patents

Abstract

A method for capturing images and an image capture device using the method are provided. The method includes: getting a preview image of a photoed person; executing a face detection on the preview image to get a face part of the photoed person; and, obtaining a plurality of features of the face part. The method also includes: determining if the face part fits a standard pose according to the features; if the face part does not fit the standard, a suggestion message is outputted from the image capture device, the suggestion message is configured to indicate how the photoed person fits the standard pose; and, if the face part fits the standard pose, capturing an image of the photoed person. Therefore, the photoed person is actively indicated that how to adjust his pose during photoing.

Description

Image capturing method and image capturing device

The present invention relates to a technique for image capture, and more particularly to an image capture method and apparatus that can actively remind a photographer how to adjust his or her posture.

With advances in the development of photosensitive elements and lenses, many devices on the market are capable of capturing images or video. For example, in addition to cameras, notebooks, cell phones, and tablets are often equipped with devices that capture images. When using these devices that can capture images to capture people, the average user would like to take a better picture of the person. However, how to determine which posture is a good-looking character photo requires a certain degree of photography skills, and the average user does not necessarily have enough photography skills.

On the other hand, with the advancement of computer vision and machine learning technology. A general image capturing device can be combined with an arithmetic unit to detect objects in the captured image. For example, with the haar-like feature extraction algorithm and the adaptive learning (AdaBoost) machine learning algorithm, the face part of the image can be accurately detected. However, such a technique can only detect a part of the face and does not help the user to take a picture of a good-looking person. Therefore, how to propose a method for a user who does not have a certain degree of photography skills can take a picture of a good person, a topic of interest to those skilled in the field.

The invention provides an image capturing method and an image capturing device, which can automatically remind the photographer how to adjust its posture so that the captured image conforms to a standard posture.

An embodiment of the invention provides an image capture method for an image capture device. The image capturing method includes: obtaining a preview image of the subject; performing a face detection on the preview image to obtain a face portion of the subject; and obtaining a plurality of features of the face portion value. The image capturing method further includes: determining, according to the feature values, whether the obtained face portion conforms to a standard posture; if the face portion does not conform to the standard posture, the image capturing device outputs a prompt message, the prompt The message is used to indicate how the subject conforms to the standard posture; and if the face portion conforms to the standard posture, an image of the subject is captured.

In an embodiment of the invention, the image capturing device includes a sound output unit. The step of outputting the prompt message by the image capturing device comprises: implementing the prompt message as an audio signal, and thereby outputting the actual sound signal by the sound output unit.

In an embodiment of the invention, the feature value includes a nose position, a left face edge position, and a right face edge position. The image capturing method further includes: determining whether a horizontal orientation of the face portion conforms to a standard posture according to a horizontal distance between a nose position and a left face edge position and a horizontal distance between the nose position and the right face edge position. Horizontally facing the angle; and, if the horizontal face of the face portion does not conform to the horizontal orientation of the standard posture, a prompt message is set to indicate that the subject turns the subject's head to the right or to the left.

In an embodiment of the invention, the feature value includes an upper eyelid height. The image capturing method further includes: determining, according to the height of the upper eyelid, whether a vertical facing angle of the face portion conforms to a vertical facing angle of the standard posture; and if the vertical facing angle of the human face portion does not meet the standard The vertical direction of the posture is set, and a prompt message is set to instruct the subject to turn the subject's head up or down.

In an embodiment of the invention, the feature value includes a pupil position and an eye contour. The image capturing method further includes: determining whether a gaze direction of the face portion conforms to a gaze direction of the standard posture according to a relative position of the pupil position relative to the contour of the eye; and if the gaze direction of the face portion does not match In the gaze direction of the standard posture, a setting prompt message is used to instruct the subject to adjust the gaze direction of the subject.

In an embodiment of the invention, the feature value includes an eye contour. The image capturing method described above further includes: determining whether an eye closed state of the face portion conforms to a standard posture eye closed state according to an eye height of the eye contour; and if the face portion of the eye closed state does not conform to the standard posture The eyes are closed, and a prompt message is set to indicate that the subject has opened or closed the subject's eyes.

In an embodiment of the invention, the feature value includes a plurality of mouth corner positions, an upper lip center position, and a lower lip center position. The image capturing method further includes: determining, according to the position of the corner of the mouth, the relative position between the center position of the upper lip and the center position of the lower lip, whether the smile state of the face portion conforms to the smile state of the standard posture; and, if the face portion The smile state of the share does not conform to the smile state of the standard posture, and the setting prompt message is used to instruct the subject to adjust the smile degree of the subject.

In an embodiment of the invention, the feature value includes an eye contour. The image capturing method further includes: determining whether the face portion is obscured by the object according to the integrity of the contour of the eye; and, if the face portion has been obscured, setting a prompt message to notify the subject that the subject has been obscured .

In another aspect, an embodiment of the present invention further provides an image capturing device. The image capturing device comprises an image capturing unit and a control unit. The image capturing unit is used to obtain a preview image of the subject. The control unit is coupled to the image capturing unit for performing a face detection on the preview image to obtain a face portion of the subject, and also obtaining a plurality of feature values of the face portion. Based on these characteristic values, the control unit determines whether the obtained face portion conforms to a standard posture. If the face part does not conform to the standard posture, the control unit will generate a prompt message to indicate how the subject conforms to the standard posture. When the face part conforms to the standard posture, an image of the subject is captured by the image capturing unit.

In an embodiment of the invention, the image capturing device further includes a sound output unit. The control unit is further configured to implement the prompt message as an audio signal, and the sound output unit sends the sound signal.

In an embodiment of the invention, the feature value includes a nose position, a left face edge position, and a right face edge position. The control unit is further configured to determine whether a horizontal orientation of the face portion conforms to a horizontal orientation of the standard posture according to a horizontal distance between the nose position and the left face edge position and a horizontal distance between the nose position and the right face edge position. . If the horizontal face of the face portion does not conform to the horizontal orientation of the standard posture, the control unit sets a prompt message to instruct the subject to turn the subject's head to the right or to the left.

In an embodiment of the invention, the feature value includes an upper eyelid height. The control unit is further configured to determine whether a vertical facing angle of the face portion conforms to a vertical facing angle of the standard posture according to the height of the upper eyelid height. If the vertical facing angle of the face portion does not conform to the vertical facing angle of the standard posture, the control unit sets a prompt message for instructing the subject to turn the subject's head up or down.

In an embodiment of the invention, the feature value includes a pupil position and an eye contour. The control unit is further configured to determine whether a gaze direction of the face portion conforms to a gaze direction of the standard posture according to a relative position of the pupil position relative to the contour of the eye. If the gaze direction of the face portion does not conform to the gaze direction of the standard posture, the control unit sets a prompt message for instructing the subject to adjust the gaze direction of the subject.

In an embodiment of the invention, the feature value includes an eye contour. The control unit is further configured to determine whether the eye closed state of the face portion conforms to the eye closed state of the standard posture according to the eye height of the eye contour. If the eye closed state of the face portion does not conform to the standard eye closed state, the control unit sets a prompt message to instruct the subject to open or close the subject's eyes.

In an embodiment of the invention, the feature value includes a plurality of mouth corner positions, an upper lip center position, and a lower lip center position. The control unit is further configured to determine whether a smile state of the face portion conforms to a smile state of the standard posture according to the relative positions between the corner position of the mouth, the center position of the upper lip, and the center position of the lower lip. If the smile state of the face portion does not conform to the smile state of the standard posture, the control unit sets a prompt message for instructing the subject to adjust the smile degree of the subject.

In an embodiment of the invention, the feature value includes an eye contour. The control unit is further configured to determine whether the face portion is obscured by an object according to the integrity of the contour of the eye. If the face portion has been obscured, the control unit sets a prompt message to notify the subject that the subject has been obscured.

Based on the above, the image capturing device and the image capturing method provided by the embodiments of the present invention can actively remind the photographer how to adjust the posture thereof through voice. Therefore, even if the photographer does not have professional photography skills, it is possible to take an image that is close to the standard posture.

The above described features and advantages of the present invention will be more apparent from the following description.

1 is a block diagram illustrating an image capture device in accordance with an embodiment.

Referring to FIG. 1, the image capturing device 100 is configured to capture an image of a subject. In particular, the image capturing device 100 can remind the photographer how to adjust its posture to make the captured image look better. In this embodiment, the image capturing device 100 is a digital camera. However, in other embodiments, the image capturing device may also be a smart phone or a notebook computer with a photographic lens. The present invention is not limit. Specifically, the image capturing device 100 includes an image capturing unit 120, a control unit 140, a preview display unit 160, and a sound output unit 180.

The image capturing unit 120 is configured to capture an image. For example, the image capturing unit 120 is a device including a lens, an aperture, a shutter, and a photosensitive element.

The control unit 140 is coupled to the image capturing unit 120, the preview display unit 160, and the sound output unit. The control unit 140 is used to control the entire image capturing device 100 and perform image processing related algorithms. For example, control unit 140 can be used to perform denoising, white balance, sharpness enhancement, or other image processing. In this embodiment, the control unit 140 is a central processing unit (CPU). However, in other embodiments, the control unit 140 may also be a microprocessor (mircoprocessor) or a special application integrated circuit (Application). -specific integrated circuit (ASIC), the invention is not limited thereto.

The preview display unit 160 is configured to display a live view, and the display of the live view allows the user to preview the content of the captured image before capturing the image. For example, the preview display unit 160 may be a liquid crystal display (LCD). However, the preview display unit 160 may also be other types of display panels, and the present invention is not limited thereto.

The sound output unit 180 is for emitting sound. For example, the sound output unit 180 is a speaker for outputting a voice signal or a music signal of the image processing apparatus 100.

The image capturing unit 120 obtains a preview image of at least one subject, and the control unit 140 displays the preview image on the preview display unit 160.

2 is a schematic diagram illustrating photographing a subject in accordance with an embodiment.

Referring to FIG. 2, in this embodiment, the subject 210 has only one bit. However, in other embodiments, the photographer may have more bits, and the present invention does not limit the number of photographers. After acquiring the preview image 202, the image capturing unit 120 displays the preview image 202 on the preview display unit 160. Moreover, the control unit 140 performs face detection on the preview image 202 to obtain a face portion 220 corresponding to the subject 210. For example, the control unit 140 can perform face detection according to the object detection algorithm proposed by Viola and Jones. However, the control unit can also use a type of neural network for face detection, and the present invention does not limit the algorithm for face detection. In this embodiment, the control unit 140 can determine whether the face portion conforms to a standard posture according to the obtained face portion, and then actively inform the photographer how to adjust the posture thereof, so that the captured image is better viewed. . Specifically, the control unit 140 first obtains a plurality of feature values of the face portion 220, and determines whether the face portion 220 conforms to a standard posture based on the feature values.

3 is a schematic diagram illustrating facial feature values in accordance with an embodiment.

Please refer to Figure 3. The control unit 140 can obtain a plurality of feature points 300a to 300g from the face portion 220. The location of these feature points is more able to describe the position of the face portion 220. For example, the information on the eyes and the eyebrows in the face can express a person's emotions, while the information on the cheeks is less able to describe the person's emotions. The number of feature points on the cheeks will be less than the number of feature points on the eyes and eyebrows. Each feature point may include feature values representative of color, position, brightness, or edge sharpness, which may be used to describe different information for the face portion 220. For example, the feature values of the feature points 300a, 300b, 300g can be used to represent the outline of the face portion 220. The feature value of the feature point 300c may represent the outline of the eye. The feature value of the feature point 300d may indicate the position of the nose. The feature values of the feature points 300e, 300f can be used to indicate whether there is a smile. In addition to this, the control unit 140 can also calculate a more semantic feature value according to the feature value of the feature point on the face portion 220. For example, the control unit 140 may calculate the feature values of the height of the nose, the position of the left face, the size of the eye, and the position of the corner of the mouth based on the feature values such as the color, position, and brightness described above. Alternatively, the control unit 140 may calculate a feature value that can represent an outline of the eye, a contour of the mouth, or the like, based on the distance or continuity between the feature points. The control unit 140 determines whether the face portion 220 is based on the feature values (eg, color, position, brightness) included in the feature points and the calculated more semantic feature values (eg, nose height, eye contour). Meets a standard posture. The present invention does not limit the content of the feature values captured and calculated by the control unit 140.

If the control unit 140 determines that the face portion 220 does not conform to the standard posture described above, the control unit 140 generates a prompt message and outputs it by the image capturing device 100. This prompt message is used to indicate how the subject 210 adjusts his or her posture to conform to the standard posture. For example, the content of this prompt message is "open your eyes", "head to the right" or "look at the lens". Moreover, the prompt message can be implemented as an audio signal, and the control unit 140 can transmit the voice signal implemented by the prompt message to the sound output unit 180, and the sound output unit 180 sends the sound signal to notify the subject 210. When the subject 210 hears the sound signal, the posture of the sound signal can be adjusted according to the indication of the sound signal.

Specifically, the control unit 140 determines whether the horizontal orientation angle, the vertical orientation angle, the gaze direction, the eye closed state, or the smile state of the face portion 220 conforms to the standard posture according to the feature value described above, thereby determining the prompt message. content. In addition, the control unit 140 may also determine whether the face portion 220 is obscured by an object according to the feature value described above. If the face portion 220 has been obscured, the control unit 140 may notify that the image is captured by a prompt message. The person 210 has been obscured.

4 is a schematic diagram illustrating a standard gesture in accordance with an embodiment. Referring to FIG. 4, in the present embodiment, the standard posture 400 is set to open the eyes, and the gaze direction is toward the lens or the like. However, the standard gesture can be set to other styles. For example, the gaze direction of the standard gesture 400 can be toward the far side, and the present invention does not limit the content of the standard gesture 400.

FIG. 5 is a schematic diagram illustrating determining a horizontal orientation angle according to an embodiment.

Please refer to Figure 5. The face portion 500 is a preview image taken from the above. The feature values obtained by the control unit 140 include a nose position 502, a left face edge position 504, and a right face edge position 506. The nose position 502 is the position of the vertical center line of the nose of the face portion 500. The left face edge position 504 is a position indicating the leftmost edge of the face portion 500. The right face edge position 506 is the position indicating the rightmost edge of the face portion 500. Control unit 140 calculates a horizontal distance 540 between left face edge position 504 and nose position 502, and a horizontal distance 520 between right face edge position 506 and nose position 502. Next, the control unit 140 can determine whether the horizontal orientation angle of the face portion 500 conforms to the horizontal orientation angle of the standard posture 400 according to the horizontal distances 520, 540. For example, when the horizontal distance 540 is greater than the horizontal distance 520, it indicates that the horizontal facing angle of the face portion is toward the right. When the control unit 140 determines that the horizontal facing angle of the face portion 500 does not conform to the horizontal orientation angle of the standard posture, the control unit 140 sets a prompt message for instructing the subject 210 to turn the subject's head to the left or to the right. . For example, when the control unit 140 determines that the horizontal facing angle of the face portion 500 is toward the right side and the horizontal facing angle of the standard posture 400 is toward the middle, the control unit 140 sets the prompt message to "Please turn the head to the right". When the subject 210 hears the sound signal of the prompt message, the head of the subject 210 can be turned to the right with respect to the subject 210, so that the preview image obtained by the image capturing device 100 is in the preview image. The horizontal facing angle of the face portion 500 is adjusted to the middle, whereby the horizontal facing angle of the face portion 500 conforms to the horizontal facing angle of the standard posture 400.

6A and 6B are schematic diagrams illustrating the determination of a vertical orientation angle, according to an embodiment.

Please refer to FIG. 6A. The face portion 600a is a pre-view image taken from the above, wherein the face portion 600a is slightly raised upward. The feature values obtained by the control unit 140 include an upper eyelid height 602a which is a vertical height indicating the upper eyelid of the eye in the face portion 600a. The control unit 140 can determine the vertical facing angle of the face portion 600a according to the size of the upper eyelid height 602a, which is used to indicate that the face portion 600a is facing upward or downward. For example, when the upper eyelid height 602a is larger, the degree to which the face portion 600a is directed upward is increased. On the contrary, when the upper eyelid height 602b is smaller, it means that the face portion 600b is directed downward (as shown in Fig. 6B). If the vertical orientation angle of the face portion 600a does not conform to the vertical orientation angle of the standard posture 400, the control unit 140 sets a prompt message for instructing the photographer 210 to turn the head of the photographer 210 up or down. For example, the vertical face angle of the face portion 600a is upward, and the vertical face angle of the standard posture 400 is toward the middle, so the control unit 140 can set the prompt message to "Please turn the head down". When the subject 210 hears the sound signal implemented by the prompt message, the head of the subject 210 can be turned down, so that the captured face portion 600a more conforms to the standard posture 400.

Figure 7 is a schematic diagram illustrating the determination of the direction of the eye according to an embodiment.

Please refer to Figure 7. The face portion 700 is captured from the preview image. The feature values obtained by control unit 140 also include pupil positions 704a, 704b and eye contours 702a, 702b. The pupil position 704a is the position of the pupil representing the right eye in the face portion 700. The eye contour 702a represents the contour of the right eye in the face portion 700, for example, the eye contour 702a can be described as an ellipse. The control unit 140 can determine the gaze direction of the face portion 700 based on the relative position of the pupil position 704a relative to the eye contour 702a, which indicates the direction in which the eyes of the face portion 700 are looking. For example, when the pupil position 704a is off-center and relatively right with respect to the eye contour 702a, the gaze pattern of the face portion 700 is toward the right. However, the control unit 140 can also calculate the relative position of the pupil position 704a relative to the eye contour 702a and the relative position of the pupil position 704b relative to the eye contour 702b to calculate the gaze direction of the face portion 700. The present invention is not This limit. If the gaze direction of the face portion 700 does not conform to the gaze direction of the standard posture 400, the control unit 140 sets a prompt message for instructing the subject 210 to adjust the gaze direction of the subject 210. For example, the gaze direction of the face portion 700 is toward the right side, and the gaze direction of the standard posture is toward the image capturing device 100, so the control unit can set the prompt message to "see the lens". When the subject 210 hears the sound signal of the prompt message, the direction of the eye can be adjusted so that the captured face portion 700 conforms to the standard posture 400.

FIG. 8 is a schematic diagram illustrating the determination of an eye closure device, in accordance with an embodiment.

Please refer to Figure 8. The face portion 800 is captured from the preview image. The feature values obtained by control unit 140 also include eye contours 802a, 802b. The eye contour 802a represents the contour of the left eye in the face portion 800, for example, the eye contour 802a can be described as an ellipse. Control unit 140 may also calculate eye heights 804a, 804b based on eye contours 802a, 802b. The control unit 140 can determine the eye closed state of the face portion according to the size of the eye height 804a of the eye contour 802a, and the eye closed state indicates whether the eye portion of the face portion 800 is closed or opened. For example, when the eye height 804a is very small, it indicates that the eye contour 802a is in a closed state. Conversely, when the eye height 804a is large, it indicates that the eye contour 802a is in a state of being split. In addition, the control unit 140 can also consider the eye height 804a and the eye height 804b to determine the eye closure state of the face portion, which is not limited by the present invention. If the eye closed state of the face portion 800 does not conform to the eye closed state of the standard posture 400, the control unit 140 sets a prompt message for instructing the subject 210 to open or close the eyes of the subject 210. For example, the control unit 140 determines that the eye closing state of the face portion 800 is closed, and the eye closing device of the standard posture 400 belongs to the opening, so the control unit 140 can set the prompt message to "Please Open your eyes." When the subject 210 hears the sound signal of the prompt message, the eyes can be opened, so that the captured face portion 800 conforms to the standard posture 400.

9 is a schematic diagram illustrating the determination of a smile state, in accordance with an embodiment.

Please refer to Figure 9. The face portion 900 is captured from the preview image. The feature values obtained by control unit 140 include mouth corner positions 902a, 902b, upper lip center position 904, and lower lip center position 906. The corner position 902a, 902b represents the position of the two corners of the face portion 900. The upper lip center position 904 represents the position of the horizontal center point of the upper lip of the face portion 900. The lower lip center position indicates the position of the horizontal center point of the lower lip of the face portion 900. The control unit 140 can obtain the smiling state of the face portion 900 according to the relative position between the corner position 902a, 904b, the upper lip center position 904, and the lower lip center position 906. The smile state indicates that the face portion 900 is smiling. degree. For example, when the mouth angular position 902a, 904b is closer to the lower lip center position 906 relative to the upper lip center position 904, the face portion 900 is not smiling. Conversely, when the mouth angular position 902a, 904b is closer to the upper lip center position 904 relative to the lower lip center position 906, the face portion 900 is more likely to smile. If the smile state of the face portion 900 does not meet the smile state of the standard gesture 400, the control unit 140 may set a prompt message for instructing the photographer 210 to adjust the smile degree of the subject 210. For example, if the control unit 140 determines that the face portion 900 does not smile and the standard gesture 400 has a smile, the control unit 140 may set the prompt message to "please smile". When the subject 210 hears the voice signal of the prompt message, the degree of smile can be increased, so that the captured face portion 900 more conforms to the standard posture 400.

FIG. 10 is a schematic diagram illustrating determining whether a face portion is obscured, according to an embodiment.

Please refer to Figure 10. The face portion 1000 is taken from a pre-view image. The feature values obtained by control unit 140 also include eye contours 1002a, 1002b. The eye contours 1002a, 1002b are outlines of the eyes representing the face portion 1000, for example, the eye contours can be described as an ellipse. The control unit 140 determines whether the face portion is obscured by an object 1004 based on the integrity of the eye contours 1002a, 1002b. For example, when the control unit 140 determines that the eye contour 1002b is not complete, for example, the eye contour 1002b is not a complete ellipse, the control unit 140 determines that the face portion 1000 has been obscured by the object 1004. If the face portion 1000 is obscured by the object 1004, the control unit sets a prompt message to notify the subject 210 that it has been obscured by the object 1004. For example, the control unit 140 can set the prompt message to "You are blocked." When the photographer 210 hears the sound signal of the prompt message, he can know that he has been blocked by an object.

However, it is worth noting that the content of the above prompt message is only an example. In other embodiments, the control unit 140 may set the prompt message to include different text content, and the present invention is not limited thereto. In addition, when there is more than one face part included in the preview image, the control unit may separately generate different prompt messages according to the condition of each face part, and the present invention is not limited thereto. In another embodiment, the control unit 140 may also display the prompt message on the preview display unit 160 by means of a pattern or a text, so that the photographer (not shown) can guide the viewer after reading the prompt message. The photographer 210 adjusts its posture to make it more conform to the standard posture 400.

The image processing apparatus 100 can automatically capture images in addition to prompting the photographer 210 how to adjust his or her posture. Referring back to FIG. 2, after the control unit 140 determines that the face portion 220 conforms to the standard posture 400, the image capturing device automatically captures the image of the subject 210. In this way, it is possible to prevent the eyes from being closed, or to have a situation such as a smile that does not conform to the standard posture 400.

In another aspect, the present invention also provides an image capture method. 11 is a flow chart illustrating an image capture method in accordance with an embodiment.

Please refer to Figure 11. In step S1102, a pre-view image of the subject is acquired. In step S1104, face detection is performed on the preview image, and at least one face portion of the subject is obtained. In step S1106, a plurality of feature values of the face portion are obtained. In step S1108, based on these feature values, it is determined whether the face portion conforms to a standard posture. In step S1110, if the face portion does not conform to the standard posture, the image capturing device outputs a prompt message for indicating how the subject conforms to the standard posture. In step S1112, if the face portion conforms to the standard posture, the image of the subject is captured. However, the steps in FIG. 11 are as described above in detail, and will not be described again here.

In summary, the image capturing device and the image capturing method provided by the embodiments of the present invention can automatically generate a prompt message to remind the photographer how to adjust the posture thereof to make the captured image look better. And determining whether the horizontal orientation of the face portion, the vertical orientation angle, the gaze direction, the eye closure state, or the smile state conform to the standard posture according to different feature values, thereby determining the content of the prompt message.

Although the present invention has been disclosed in the above embodiments, it is not intended to limit the invention, and any one of ordinary skill in the art can make some modifications and refinements without departing from the spirit and scope of the invention. The scope of the invention is defined by the scope of the appended claims.

100. . . Image capture device

120. . . Image capture unit

140. . . control unit

160. . . Preview display unit

180. . . Sound output unit

210. . . Photographer

220, 500, 600a, 600b, 700, 800, 900, 1000. . . Face part

202. . . Preview image

300a~300g. . . Feature points

400. . . Standard posture

502. . . Nose position

504. . . Left face edge position

506. . . Right face edge position

520, 540. . . Horizontal distance

602a, 602b. . . Upper eyelid height

702a, 702b, 802a, 802b, 1002a, 1002b. . . Eye contour

704a, 704b. . . Pupil position

804a, 804b. . . Eye height

902a, 902b. . . Mouth corner position

904. . . Upper lip center position

906. . . Lower lip center position

1004. . . object

S1102, S1104, S1106, S1108, S1110, S1112. . . Steps of the image capture method

1 is a block diagram illustrating an image capture device in accordance with an embodiment.

2 is a schematic diagram illustrating photographing a subject in accordance with an embodiment.

3 is a schematic diagram illustrating facial feature values in accordance with an embodiment.

4 is a schematic diagram illustrating a standard gesture in accordance with an embodiment.

FIG. 5 is a schematic diagram illustrating determining a horizontal orientation angle according to an embodiment.

6A and 6B are schematic diagrams illustrating the determination of a vertical orientation angle, according to an embodiment.

Figure 7 is a schematic diagram illustrating the determination of the direction of the eye according to an embodiment.

FIG. 8 is a schematic diagram illustrating the determination of an eye closure device, in accordance with an embodiment.

9 is a schematic diagram illustrating the determination of a smile state, in accordance with an embodiment.

FIG. 10 is a schematic diagram illustrating determining whether a face portion is obscured, according to an embodiment.

11 is a flow chart illustrating an image capture method in accordance with an embodiment.

S1102, S1104, S1106, S1108, S1110, S1112. . . Steps of the image capture method

Claims (16)

An image capturing method for an image capturing device includes: acquiring a preview image of at least one subject; performing a face detection on the preview image to obtain a face of each of the at least one subject And obtaining a plurality of feature values of the face portion; determining, according to the feature values, whether the face portion meets a standard posture; if the face portion does not conform to the standard posture, the image is The taking device outputs a prompt message for indicating how the subject conforms to the standard posture; and if the face portion conforms to the standard posture, capturing an image of the subject. The image capturing method of claim 1, wherein the image capturing device comprises a sound output unit, wherein the step of outputting the prompt message by the image capturing device comprises: implementing the prompt message as a sound a signal, and the sound signal is sent by the sound output unit. The image capturing method of claim 1, wherein the feature values include a nose position, a left face edge position, and a right face edge position, the image capturing method further comprising: according to the nose position and a horizontal distance between the left face edge position and a horizontal distance between the nose position and the right face edge position, determining whether a horizontal facing angle of the face portion conforms to a horizontal facing angle of the standard posture; The horizontal facing angle of the face portion does not conform to the horizontal facing angle of the standard posture, and the prompt message is set to instruct the subject to turn the subject's head to the right or to the left. The image capturing method of claim 1, wherein the feature values include an upper eyelid height, and the image capturing method further comprises: determining a face portion of the upper eyelid according to the height of the upper eyelid Whether the vertical facing angle conforms to a vertical facing angle of the standard posture; and if the vertical facing angle of the human face portion does not conform to the vertical facing angle of the standard posture, setting the prompt message to indicate that the subject will The subject's head turns up or down. The image capturing method of claim 1, wherein the feature values include a pupil position and an eye contour, the image capturing method further comprising: according to the relative position of the pupil position relative to the contour of the eye, Determining whether a gaze direction of the face portion conforms to a gaze direction of the standard gesture; and if the gaze direction of the face portion does not conform to the gaze direction of the standard gesture, setting the prompt message to indicate the gaze The photographer adjusts the direction of the subject's gaze. The image capturing method of claim 1, wherein the feature values include an eye contour, the image capturing method further comprising: determining a face portion of the eye based on an eye height of the eye contour Whether the eye closed state conforms to an eye closed state of the standard posture; and if the eye closed state of the face portion does not conform to the eye closed state of the standard posture, setting the prompt message to indicate that the subject is to be The subject's eyes are opened or closed. The image capturing method of claim 1, wherein the feature values include a plurality of mouth corner positions, an upper lip center position, and a lower lip center position, the image capturing method further comprising: according to the mouth corner positions a relative position between the center position of the upper lip and the center position of the lower lip, determining whether a smile state of the face portion conforms to a smile state of the standard posture; and if the smile state of the face portion is not In the smile state conforming to the standard posture, the prompt message is set to instruct the subject to adjust the smile degree of the subject. The image capturing method of claim 1, wherein the feature values include an eye contour, and the image capturing method further comprises: determining, according to a completeness of the eye contour, whether the face portion is An object is obscured; and if the face portion has been obscured, the prompt message is set to notify the subject that the subject has been obscured. An image capturing device includes: an image capturing unit for acquiring at least one preview image of a subject; and a control unit coupled to the image capturing unit for performing the preview image a face detection to obtain a face portion of each of the at least one subject, and obtain a plurality of feature values of the face portion, wherein the control unit determines whether the face portion is based on the feature values In accordance with a standard posture, if the face portion does not conform to the standard posture, the control unit generates a prompt message for indicating how the subject conforms to the standard posture, wherein the face portion When the standard posture is met, an image of the subject is captured by the image capturing unit. The image capturing device of claim 9, wherein the image capturing device further comprises a sound output unit, wherein the control unit is further configured to implement the prompt message as an audio signal and output the sound The unit sends the sound signal. The image capturing device of claim 9, wherein the feature values include a nose position, a left face edge position, and a right face edge position, and the control unit is further configured to: according to the nose position and the left a horizontal distance between the edge position of the face and a horizontal distance between the position of the nose and the edge of the right face, determining whether a horizontal orientation of the face portion conforms to a horizontal orientation of the standard posture, if the face A portion of the horizontal facing angle does not conform to the horizontal facing angle of the standard posture, and the control unit sets the prompt message to instruct the subject to turn the subject's head to the right or to the left. The image capturing device of claim 9, wherein the feature values include an upper eyelid height, and the control unit is further configured to determine a vertical face of the face portion according to the height of the upper eyelid Whether the angle conforms to a vertical orientation angle of the standard posture, and if the vertical orientation angle of the face portion does not meet the vertical orientation angle of the standard posture, the control unit sets the prompt message to indicate that the subject is to be The subject's head turns up or down. The image capturing device of claim 9, wherein the feature values include a pupil position and an eye contour, and the control unit is further configured to determine the relative position of the pupil position relative to the eye contour. Whether a gaze direction of the face portion conforms to a gaze direction of the standard posture, and if the gaze direction of the face portion does not conform to the gaze direction of the standard posture, the control unit sets the prompt message to indicate the gaze The photographer adjusts the direction of the subject's gaze. The image capturing device of claim 9, wherein the feature values include an eye contour, and the control unit is further configured to determine an eye closure of the face portion according to an eye height of the eye contour Whether the state conforms to an eye closed state of the standard posture, and if the eye closed state of the face portion does not meet the eye closed state of the standard posture, the control unit sets the prompt message to indicate that the subject will The subject's eyes are opened or closed. The image capturing device of claim 9, wherein the feature values include a plurality of mouth corner positions, an upper lip center position, and a lower lip center position, and the control unit is further configured to: according to the mouth corner positions, a relative position between the center position of the upper lip and the center position of the lower lip, determining whether a smile state of the face portion conforms to a smile state of the standard posture, and if the smile state of the face portion does not meet the standard In the smiling state of the gesture, the control unit sets the prompt message to instruct the subject to adjust the smile degree of the subject. The image capturing device of claim 9, wherein the feature values include an eye contour, and the control unit is further configured to determine whether the face portion is an object according to a completeness of the eye contour If the face portion is obscured, the control unit sets the prompt message to notify the subject that the subject has been obscured.