CN104683692A - Continuous shooting method and device - Google Patents

Abstract

The embodiment of the invention discloses a continuous shooting method and a continuous shooting device. The continuous shooting method comprises the following steps of receiving a continuous shooting start instruction, acquiring an initial image frame which is collected by a camera and includes a facial image, and regarding the initial image frame as a reference image and storing the initial image frame; acquiring a collected image frame which is collected by the camera and includes the facial image; detecting a difference value between facial feature information in the collected image frame and the facial feature information of the reference image; when the difference value is more than or equal to a threshold value, setting the collected image frame as the reference image and storing the collected image frame, and carrying out the step of acquiring the collected image frame collected by the camera and including the facial image. By adopting the continuous shooting method and the continuous shooting device, a terminal resource can be saved, and the accuracy of capturing a useful image frame in a continuous shooting process is improved.

Description

Continuous shooting method and device

technical field

The present invention relates to the field of terminals, in particular to a continuous shooting method and device.

Background technique

With the continuous development of terminal technology, more and more functions are integrated in the terminal, gradually becoming an indispensable part of people's life. Among them, the GIF (Graphic Interchange Format, Image Interchange Format) dynamic graphic mode provided by the terminal is very interesting and is deeply loved by users. The principle of this GIF dynamic image is: store multiple images in a GIF file, and read out and display the multiple image data stored in a GIF file one by one to form the simplest animation. In actual use, users usually use continuous shooting mode to capture image frames with continuous actions or expressions as the original material for GIF dynamic image synthesis. In the prior art, continuous shooting mode usually uses fixed time intervals to collect image frames as continuous shooting images , in order to make the captured images continuous, the fixed time interval is set to be short, but the speed of the user's facial expressions or actions often cannot keep up with the speed of the continuous shooting of the camera, so some blurred images will be captured during the continuous shooting Or change useless image frames in the process of moving expressions, wasting storage resources of the terminal.

Contents of the invention

Embodiments of the present invention provide a continuous shooting method and device, which can save storage resources of a terminal and improve the accuracy of capturing useful image frames during continuous shooting.

An embodiment of the present invention provides a continuous shooting method, and the continuous shooting method includes:

Receive the start command of continuous shooting, obtain the initial image frame that contains the face image collected by the camera, and store the initial image frame as a reference image;

Obtaining the collected image frame containing the facial image collected by the camera;

Detecting a difference value between the face feature information in the captured image frame and the face feature information in the reference image;

When the difference value is greater than or equal to a threshold, the captured image frame is set as a reference image and stored, and the step of acquiring the captured image frame containing the facial image captured by the camera is performed.

Correspondingly, an embodiment of the present invention provides a continuous shooting device, which includes:

The image frame acquisition unit is used to receive the continuous shooting start instruction, and acquire the initial image frame containing the facial image collected by the camera;

a reference image setting unit, configured to use the initial image frame as a reference image and store it;

The image frame acquiring unit is also used to acquire the captured image frame containing the facial image captured by the camera;

A face feature detection unit, configured to detect a difference value between the face feature information in the captured image frame and the face feature information of the reference image;

The reference image setting unit is further configured to set and store the acquired image frame as a reference image when the difference value is greater than or equal to a threshold.

In the embodiment of the present invention, a continuous shooting start instruction can be received, the initial image frame containing the facial image collected by the camera can be obtained, and the initial image frame can be used as a reference image and stored; Collecting an image frame; detecting the difference value between the face feature information in the collected image frame and the face feature information of the reference image; and then setting the collected image frame as a reference image when the difference value is greater than a threshold and storing, and performing the step of acquiring the captured image frame containing the facial image captured by the camera can save storage resources of the terminal, improve the accuracy of capturing useful image frames in the continuous shooting process, and improve user experience.

Description of drawings

In order to illustrate the technical solutions of the embodiments of the present invention more clearly, the drawings that need to be used in the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are some embodiments of the present invention. Ordinary technicians can also obtain other drawings based on these drawings on the premise of not paying creative work.

FIG. 1 is a schematic flow chart of a continuous shooting method provided by an embodiment of the present invention;

Fig. 2 is a schematic flow chart of another continuous shooting method provided by an embodiment of the present invention;

Fig. 3 is a schematic structural diagram of a continuous shooting device provided by an embodiment of the present invention;

4 is a schematic structural diagram of an image frame acquisition unit provided by an embodiment of the present invention;

Fig. 5 is a schematic structural diagram of a face feature detection unit provided by an embodiment of the present invention.

detailed description

The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some of the embodiments of the present invention, but not all of them. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.

In specific implementation, the terminals mentioned in the embodiments of the present invention include, but are not limited to: smart phones (such as Android phones, IOS phones), tablet computers, notebook computers, palmtop computers, digital cameras and other electronic devices. The method described in the embodiment of the present invention is executed by a set of codes stored in the memory of the terminal, and the codes are executed in a computer system based on the Von Neumann architecture.

A continuous shooting method and device provided in the embodiments of the present invention will be described in detail below with reference to FIG. 1 to FIG. 5 .

Referring to FIG. 1 , it is a schematic flow chart of a continuous shooting method provided by an embodiment of the present invention. The continuous shooting method as shown in the figure may include the following steps:

S101. Receive a continuous shooting start instruction, acquire a starting image frame including a facial image collected by a camera, and store the starting image frame as a reference image.

In the specific implementation, you can first start the camera and select the continuous shooting mode in the setting menu of the terminal. After the camera enters the viewfinder mode, it will automatically focus. When the user enters the camera command when the posture is ready, it will be regarded as the continuous shooting start command. Among them, There are many ways to input the camera command. It can be triggered by the user clicking the camera button, or it can be triggered when the terminal is in the delayed camera mode and the camera delay arrives. In the continuous shooting method in this embodiment, its main application scenario is to collect continuous shooting images containing facial images of people. Therefore, when the terminal receives a continuous shooting instruction, the camera collects the initial image frame containing facial images, and the initial frame The image frame is used as a reference image and stored as the first image captured in the continuous shooting process.

S102. Acquire a captured image frame including a facial image captured by the camera.

In step S101, after the initial image frame collected by the camera is used as a reference image, the camera continues to find a view and perform auto-focus to collect an image frame containing a facial image, waiting for other images in the continuous shooting process.

S103. Detect a difference value between the face feature information in the captured image frame and the face feature information in the reference image.

In the specific implementation, this step can be based on the face recognition technology. Firstly, face detection is performed on the collected image frame and the reference image (judging whether there is a facial image in a complex background and segmenting it from the background) to obtain the information contained in it. Facial image, then sample the facial image contained in the captured image frame and the reference image to extract facial feature information, and then detect the difference between the facial feature information in the captured image frame and the facial feature information of the reference image difference value. Face detection methods include but are not limited to: reference template method, face rule method, sample learning method, skin color model method or feature sub-face method, etc. The extraction of face feature information usually includes: feature vector method and face pattern template method. Among them, the eigenvector method of face recognition refers to: determine the size, position, distance and other attributes of the contours of the facial features such as the iris, nose, and corners of the mouth, and then calculate the geometric features of each feature point in the contours of the facial features (European The distance, curvature, angle) or algebraic feature quantity (matrix feature vector) are used to describe the facial features of the facial image in the image frame.

In this embodiment, the coordinate information of facial feature points (such as mouth corners, nose wings, eye corners, eye contours, nose tip, brow head, brow tail) in the image frame can be obtained, and then detected respectively. Collect the coordinate difference between each facial feature point in the image frame and each corresponding facial feature point in the reference image to judge whether the facial image in the collected image frame has an expression or a position change, for example, the change in the distance between the brows can judge the human Whether the face has a frowning expression, the distance between the tip of the nose and the corner of the eye can be used to judge whether there is a wrinkled nose expression on the face.

For example, set the threshold of the difference between the horizontal and vertical coordinates of the mouth corner feature points A and B in the face features to 0.5, after performing face detection and face feature information extraction on the collected image frame and reference image, the obtained image frame The coordinates of point A and point B of the corner of the mouth are (0.5,1.5) and (2.5,1.5) respectively, and the coordinates of point A and point B of the corner of the mouth in the reference image are (1,1) and (2,1) respectively, so it can be detected The difference values of the horizontal and vertical coordinates of point A and point B in the captured image frame and the reference image are respectively 0.5, which is equal to the preset threshold value. At this time, step S104 is performed, and the captured image frame is set as the reference image and stored. Wherein, the difference between the abscissa of the corners of the mouth A and B can represent the length of the mouth, the difference between the ordinates of the corners of the mouth A and B can represent the height difference of the corners of the mouth, it can be seen that relative to the mouth of the facial image in the reference image, The length of the mouth in the facial image in the collected image frame has increased and the corners of the mouth have risen. Therefore, it can be judged that compared with the mouth in the facial image in the reference image, the mouth in the collected image frame has changed into a smiling state. Compared with the mouth in the reference image collected The facial images in the image frames have expression changes.

S104. When the difference value is greater than or equal to a threshold, set the captured image frame as a reference image and store it, and perform a step of acquiring a captured image frame including a facial image captured by the camera.

In this embodiment, after the captured image frame is set as a reference image and stored, if the captured images for this continuous shooting have not been completed, steps S102-S104 may be executed in a loop until the current continuous shooting is completed.

In another optional embodiment, after performing the step of setting the collected image frames as reference images and storing them, the method further includes: judging whether the number of image frames stored as reference images is greater than or equal to the first a threshold, if so, end the continuous shooting; otherwise, execute the step of acquiring the captured image frame containing the facial image captured by the camera.

In the specific implementation, the number of images required for this continuous shooting can be specified as the first threshold when starting the camera to select the continuous shooting mode. Usually, the terminal provides 4 continuous shooting, 6 continuous shooting, 8 continuous shooting or 16 continuous shooting mode options It is for users to choose, and the number of images required for continuous shooting can also be customized after selecting the continuous shooting mode. Each time the terminal captures an image during the continuous shooting process, it judges whether the image captured by this continuous shooting reaches the preset first threshold value, and if so, ends this continuous shooting; if not, continues to execute steps S102-S104 in a loop until the continuous shooting is completed.

In yet another optional embodiment, after performing the step of receiving the continuous shooting start instruction, the method further includes: acquiring a start time stamp of the continuous shooting start instruction.

Correspondingly, the step of acquiring the captured image frame containing the facial image captured by the camera further includes: acquiring the current time stamp, calculating the continuous shooting duration according to the current time stamp and the starting time stamp, and judging the Whether the shooting duration is greater than or equal to the second threshold, if so, end the continuous shooting; otherwise, execute the step of acquiring the captured image frame containing the facial image captured by the camera.

In a specific implementation, when the continuous shooting mode is selected in advance, the continuous shooting duration required for this continuous shooting can be set as the second threshold. After receiving the continuous shooting start command from the terminal, start timing. Each time an image is taken, it is judged whether the duration of the continuous shooting reaches the preset second threshold, and if so, the continuous shooting is ended; if not, steps S102-S104 are continuously executed in a loop until the duration of the continuous shooting reaches the preset value Threshold to complete this continuous shooting.

In yet another optional embodiment, after performing the step of setting the acquired image frame as a reference image and storing it, the method further includes: receiving a dynamic image generation instruction, acquiring an image frame that is used as a reference image and stored, A dynamic image is generated according to the image frame stored as a reference image.

In a specific implementation, after the continuous shooting is finished, the terminal can directly or according to the needs of the user make the image captured by the continuous shooting into a GIF dynamic image, so as to improve the fun of the continuous shooting mode.

In yet another optional embodiment, step S102 may include the steps of:

11), obtaining multiple cached image frames containing facial images collected by the camera;

12), calculating the focus evaluation value of each buffered image frame in the plurality of buffered image frames;

13) Selecting the cached image frame with the largest focus evaluation value among the plurality of cached image frames as the acquired image frame.

After the terminal acquires the first continuous shooting image, the terminal collects multiple cached image frames containing facial images through the camera, and simultaneously calculates the focus evaluation value of the focus area in each image frame, where the focus evaluation value is the sharpness of the focus area Reference value (such as sharpness value), and then select the cached image frame with the largest focus evaluation value as the captured image frame, which can obtain a relatively clear captured image frame to increase the clarity of the continuous shooting image, and has a certain anti-shake effect .

In the embodiment of the present invention, a continuous shooting start instruction can be received, the initial image frame containing the facial image collected by the camera can be obtained, and the initial image frame can be used as a reference image and stored; Collecting an image frame; detecting the difference value between the face feature information in the collected image frame and the face feature information of the reference image; and then setting the collected image frame as a reference image when the difference value is greater than a threshold and storing, and performing the step of acquiring the captured image frame containing the facial image captured by the camera can save storage resources of the terminal, improve the accuracy of capturing useful image frames in the continuous shooting process, and improve user experience.

Referring to FIG. 2 , it is a schematic flowchart of another continuous shooting method provided by an embodiment of the present invention. The continuous shooting method as shown in the figure may include the following steps:

S201. Receive a continuous shooting start instruction, acquire a starting image frame including a facial image collected by a camera, and store the starting image frame as a reference image.

S202. Acquire a captured image frame including a facial image captured by the camera.

S203. Obtain coordinate values of facial features in the collected image frame and the reference image respectively.

In the specific implementation, this step can be based on the face recognition technology. Firstly, face detection is performed on the collected image frame and the reference image (judging whether there is a facial image in a complex background and segmenting it from the background) to obtain the information contained in it. The facial image, and then sample the facial image contained in the acquisition image frame and the reference image to obtain the coordinate values of the facial feature points.

S204. Determine whether the difference between the coordinate value of the facial feature in the reference image and the coordinate value of the facial feature in the captured image frame is greater than or equal to a threshold.

If the difference between the coordinates of the facial features in the reference image and the coordinates of the facial features in the captured image frame is greater than or equal to the threshold, continue to step S205. For example, set the threshold of the difference between the horizontal and vertical coordinates of the mouth corner feature points A and B in the face features to 0.5, after performing face detection and face feature information extraction on the collected image frame and reference image, the obtained image frame The coordinates of point A and point B of the corner of the mouth are (0.5,1.5) and (2.5,1.5) respectively, and the coordinates of point A and point B of the corner of the mouth in the reference image are (1,1) and (2,1) respectively, so it can be detected The difference values of the horizontal and vertical coordinates of point A and point B in the captured image frame and the reference image are respectively 0.5 and equal to the preset threshold, and step S205 is executed at this time. Optionally, if the difference between the coordinates of the facial features in the reference image and the coordinates of the facial features in the captured image frame is smaller than the threshold, go to step S202.

S205. Set and store the captured image frame as a reference image.

S206. Determine whether the number of image frames stored as reference images is greater than or equal to a first threshold.

In the specific implementation, the number of images required for this continuous shooting can be designated as the first threshold while starting the camera to select the continuous shooting mode. Usually, the terminal provides 4 continuous shooting, 6 continuous shooting, 8 continuous shooting or 16 continuous shooting mode options for selection. The user can choose, and can also customize the number of images required for continuous shooting after selecting the continuous shooting mode. Each time the terminal captures an image during the continuous shooting process, it judges whether the image captured by this continuous shooting reaches the preset first threshold, and if so, ends this continuous shooting; if not, proceeds to steps S202-S206 until the continuous shooting is completed.

S207, end the continuous shooting.

In the embodiment of the present invention, the initial image frame containing the facial image collected by the camera can be obtained when receiving the continuous shooting start instruction, and the initial image frame can be used as a reference image and stored, and the image containing the facial image collected by the camera can also be obtained The collected image frames, and then obtain the coordinate values of the facial features in the collected image frame and the reference image respectively, and determine the coordinate values of the facial features in the reference image and the coordinate values of the facial features in the collected image frame When the difference between them is greater than or equal to a threshold value, set the captured image frame as a reference image and store it, then judge whether the number of image frames stored as a reference image is greater than or equal to a first threshold value, and judge The result is that it is time to end the continuous shooting, which can save the storage resources of the terminal, improve the accuracy of capturing useful image frames during the continuous shooting, and improve user experience.

Referring to FIG. 3 , a schematic structural diagram of a continuous shooting device provided by an embodiment of the present invention, as shown in the figure, the continuous shooting device may at least include: an image frame acquisition unit 301 , a reference image setting unit 302 and a facial feature detection unit 303 .

The image frame acquiring unit 301 is configured to receive a continuous shooting start instruction, and acquire an initial image frame including a facial image captured by the camera.

In the specific implementation, you can first start the camera and select the continuous shooting mode in the setting menu of the terminal. After the camera enters the viewfinder mode, it will automatically focus. When the user enters the camera command when the posture is ready, it will be regarded as the continuous shooting start command. Among them, There are many ways to input the camera command, which may be triggered by the user clicking the camera button, or triggered when the terminal is in the delayed camera mode and the terminal's camera delay arrives. In the continuous shooting method in this embodiment, its main application scenario is to collect continuous shooting images that include facial images of people. Therefore, when the image frame acquisition unit 301 receives a continuous shooting instruction, the image frame acquisition unit 301 collects images containing facial images through the camera. The starting image frame, the reference image setting unit 302 stores the starting image frame as a reference image as the first image captured in the continuous shooting process.

The reference image setting unit 302 is configured to use the starting image frame as a reference image and store it.

The image frame acquiring unit 301 is further configured to acquire the captured image frame including the facial image captured by the camera.

After the reference image setting unit 302 uses the initial image frame as a reference image, the camera continues to find a view and perform auto-focus to collect image frames containing facial images, waiting for other images in the continuous shooting process.

A face feature detection unit 303, configured to detect a difference value between the face feature information in the captured image frame and the face feature information in the reference image.

In the specific implementation, the face feature detection unit 303 can be based on the face recognition technology, and first perform face detection on the captured image frame and the reference image (judging whether there is a facial image in a complex background and segmenting it from the background) to Obtain the facial image contained in it, then sample the facial image contained in the collected image frame and the reference image to extract the facial feature information, and then detect the facial feature information in the collected image frame and the human face of the reference image Difference value of face feature information. Face detection methods include but are not limited to: reference template method, face rule method, sample learning method, skin color model method or feature sub-face method, etc. The extraction of face feature information usually includes: feature vector method and face pattern template method. Among them, the eigenvector method of face recognition refers to: determine the size, position, distance and other attributes of the facial features contours such as iris, nose wing, and mouth corners, and then calculate the geometric feature quantity (Euclidean distance) of each feature point in the facial features contours , curvature, angle) or algebraic feature quantity (matrix feature vector) is used to describe the facial features of the facial image in the image frame.

In this embodiment, the coordinate information of facial feature points (such as mouth corners, nose wings, eye corners, eye contours, nose tip, brow head, brow tail) in the image frame can be obtained, and then detected respectively. Collect the coordinate difference between each facial feature point in the image frame and each corresponding facial feature point in the reference image to judge whether the facial image in the collected image frame has an expression or a position change, for example, the change in the distance between the brows can judge the human Whether the face has a frowning expression, the distance between the tip of the nose and the corner of the eye can be used to judge whether there is a wrinkled nose expression on the face.

For example, set the threshold of the difference between the horizontal and vertical coordinates of the mouth corner feature points A and B in the face features to 0.5, after performing face detection and face feature information extraction on the collected image frame and reference image, the obtained image frame The coordinates of point A and point B of the corner of the mouth are (0.5,1.5) and (2.5,1.5) respectively, and the coordinates of point A and point B of the corner of the mouth in the reference image are (1,1) and (2,1) respectively, so it can be detected The difference values of the horizontal and vertical coordinates of point A and point B in the captured image frame and the reference image are respectively 0.5, which is equal to the preset threshold, and the reference image setting unit 302 sets the captured image frame as a reference image and stores it. Wherein, the difference between the abscissa of the corners of the mouth A and B can represent the length of the mouth, the difference between the ordinates of the corners of the mouth A and B can represent the height difference of the corners of the mouth, it can be seen that relative to the mouth of the facial image in the reference image, The length of the mouth in the facial image in the collected image frame has increased and the corners of the mouth have risen. Therefore, it can be judged that compared with the mouth in the facial image in the reference image, the mouth in the collected image frame has changed into a smiling state. Compared with the mouth in the reference image collected The facial images in the image frames have expression changes.

The reference image setting unit 302 is further configured to set and store the captured image frame as a reference image when the difference value is greater than or equal to a threshold.

In this embodiment, after the reference image setting unit 302 sets the captured image frame as a reference image and stores it, if the captured image for this continuous shooting is not completed, it can continue to pass through the image frame acquiring unit 301. The reference image setting unit 302 and the face feature detection unit 303 complete the collection of continuous shooting images until the current continuous shooting ends.

In another optional embodiment, the continuous shooting device further includes: an image quantity judging unit 304, which is used for judging that the captured image frame is used as a reference image after the reference image setting unit sets and stores the captured image frame as a reference image Whether the number of image frames stored in the image is greater than or equal to the first threshold, if so, end the continuous shooting; otherwise, the image frame acquisition unit 301 acquires the captured image frame containing the facial image captured by the camera.

In the specific implementation, the number of images required for this continuous shooting can be specified as the first threshold when starting the camera to select the continuous shooting mode. Usually, the terminal provides 4 continuous shooting, 6 continuous shooting, 8 continuous shooting or 16 continuous shooting mode options It is for users to choose, and the number of images required for continuous shooting can also be customized after selecting the continuous shooting mode. Every time an image is taken during the continuous shooting process, the image quantity judging unit 304 judges whether the image taken by this continuous shooting reaches the preset first threshold, if so, then ends this continuous shooting; The image frame acquisition unit 301, the reference image setting unit 302 and the face feature detection unit 303 take continuous shooting images until the continuous shooting is completed.

In yet another optional embodiment, the continuous shooting device further includes: a time stamp acquisition unit 305, configured to acquire the start time stamp of the continuous shooting start instruction after the image frame acquisition unit receives the continuous shooting start instruction .

The timestamp acquiring unit 305 is also configured to acquire the current timestamp.

Correspondingly, the continuous shooting device further includes: a continuous shooting duration judging unit 306, configured to calculate the continuous shooting duration according to the current timestamp and the starting timestamp, and judge whether the continuous shooting duration is greater than or equal to the second threshold, if so, end the continuous shooting; otherwise, the image frame acquisition unit 301 acquires the captured image frame containing the facial image captured by the camera.

In a specific implementation, when the continuous shooting mode is selected in advance, the continuous shooting duration required for this continuous shooting can be set as the second threshold. After receiving the continuous shooting start instruction from the terminal, the timestamp acquisition unit 305 obtains Start time, start counting, and each time an image is taken during the continuous shooting process, the timestamp acquisition unit 305 obtains the current time to calculate the continuous shooting duration, and the continuous shooting duration judging unit 306 judges whether the continuous shooting duration has reached the preset second Two thresholds, if so, then end this continuous shooting; if not, continue to pass through the image frame acquisition unit 301, the reference image setting unit 302, the facial feature detection unit 303 to take continuous shooting images until this continuous shooting The continuous shooting is completed when the duration reaches the preset threshold.

In yet another optional embodiment, the continuous shooting device further includes: a dynamic image synthesis unit 307, configured to receive a dynamic image generation instruction after the reference image setting unit sets the captured image frame as a reference image and stores it , acquiring an image frame that is used as a reference image and stored, and generating a dynamic image based on the image frame that is used as a reference image and stored.

After the continuous shooting is finished, the dynamic image synthesizing unit 307 can directly or according to the needs of users make the images captured by this continuous shooting into GIF dynamic images, so as to improve the fun of the continuous shooting mode.

Referring to FIG. 4 , a schematic structural diagram of an image frame acquisition unit provided by an embodiment of the present invention, the image frame acquisition unit 301 as shown in the figure may include: a cached image frame acquisition subunit 3101, a focus evaluation value calculation subunit 3102, and an acquired image frame selection Subunit 3103.

The cached image frame acquisition subunit 3101 is configured to acquire multiple cached image frames containing facial images collected by the camera.

The focus evaluation value calculation subunit 3102 is configured to calculate the focus evaluation value of each buffered image frame in the plurality of buffered image frames.

The captured image frame selection subunit 3103 is configured to select the cached image frame with the largest focus evaluation value among the plurality of cached image frames as the captured image frame.

After acquiring the first continuous shooting image, the buffered image frame acquisition subunit 3101 collects multiple buffered image frames containing facial images through the camera, and the focus evaluation value calculation subunit 3102 calculates the focus evaluation value of the focus area in each buffered image frame , where the focus evaluation value is the sharpness reference value (such as the sharpness value) of the focus area, and then the captured image frame selection subunit 3103 selects the cached image frame with the largest focus evaluation value as the captured image frame, so that a relatively clear image can be obtained Capture image frames to increase the clarity of continuous shooting images, which has a certain anti-shake effect.

Referring to FIG. 5 , a schematic structural diagram of a face feature detection unit provided by an embodiment of the present invention, the face feature detection unit 303 as shown in the figure may include: a face feature acquisition subunit 3301 and a face feature judgment subunit 3303 .

The face feature acquisition subunit 3301 is configured to acquire the coordinate values of the face features in the captured image frame.

In a specific implementation, the face feature acquisition subunit 3301 can first perform face detection on the captured image frame (judging whether there is a face image in a complex background and segmenting it from the background) based on face recognition technology to obtain its The face image contained in the frame and the face image contained in the reference image are then sampled to obtain the coordinate values of the face feature points.

The facial feature acquiring subunit 3301 is also used to acquire the coordinate values of the facial features in the reference image.

The face feature judging subunit 3302 is used to judge whether the difference between the coordinate value of the face feature in the reference image and the coordinate value of the face feature in the captured image frame is greater than a threshold, if so, the reference image The setting unit 302 sets the captured image frame as a reference image and stores it; if not, the image frame acquiring unit 301 acquires the captured image frame containing the facial image captured by the camera.

For example, the threshold value of the horizontal and vertical coordinate difference value of the mouth corner feature point A point and B point in the face feature is set to 0.5, and the face feature acquisition subunit 3301 obtains the coordinates of the mouth corner A point and B point in the captured image frame as (0.5 , 1.5) and (2.5, 1.5), the coordinates of point A and point B of the corner of the mouth in the reference image are (1, 1) and (2, 1) respectively, so the face feature judgment subunit 3302 can detect the captured image frame and The difference values of horizontal and vertical coordinates of point A and point B in the reference image are respectively 0.5, which is equal to the preset threshold. At this time, the reference image setting unit 302 sets the captured image frame as a reference image and stores it.

A continuous shooting method and device disclosed in the embodiments of the present invention have been described in detail above. What is disclosed above is only a preferred embodiment of the present invention. Of course, the scope of rights of the present invention cannot be limited by this. Therefore, according to the rights of the present invention The equivalent changes required to be made still belong to the scope covered by the present invention.

Claims (12)

1. a continuous shooting method, is characterized in that, comprising:

Receive continuous shooting enabled instruction, obtain the start image frame containing face-image of camera collection, using described start image frame as storing with reference to image;

Obtain the acquired image frames containing face-image of described camera collection;

Detect the difference value of the face characteristic information of face characteristic information in described acquired image frames and described reference picture;

When described difference value is more than or equal to threshold value, described acquired image frames is set to reference picture and stores, performing the step of the acquired image frames containing face-image obtaining described camera collection.

2. method according to claim 1, is characterized in that, described described acquired image frames is set to reference picture and store after, also comprise:

Judge whether the quantity of the described picture frame as also storing with reference to image is more than or equal to first threshold, if so, terminates continuous shooting; Otherwise, perform the step of the acquired image frames containing face-image of the described camera collection of described acquisition.

3. method according to claim 1, is characterized in that, after described reception continuous shooting enabled instruction step, also comprises:

Obtain the initial time stamp of described continuous shooting enabled instruction;

The step of the acquired image frames containing face-image of the described camera collection of described acquisition also comprises:

Acquisition current time is stabbed, and calculates continuous shooting duration, judge whether described continuous shooting duration is more than or equal to Second Threshold, if so, terminates continuous shooting according to described current time stamp and described initial time stamp; Otherwise, perform the step of the acquired image frames containing face-image of the described camera collection of described acquisition.

4. method according to claim 1, is characterized in that, described described acquired image frames is set to reference picture and store step after also comprise:

Receive Dynamic Graph and generate instruction, obtain as with reference to image the picture frame be stored, according to described as with reference to image and the picture frame be stored generates Dynamic Graph.

5. method according to claim 1, is characterized in that, the acquired image frames containing face-image of the described camera collection of described acquisition comprises:

Obtain multiple cache image frames containing face-image of described camera collection;

Calculate the focusing evaluation of estimate of each cache image frame in multiple cache image frames described;

Choose in multiple cache image frames described the maximum cache image frame of evaluation of estimate of focusing as described acquired image frames.

6. method according to claim 1, is characterized in that, the difference value of the face characteristic information of the described acquired image frames of described detection and described reference picture comprises:

Obtain the coordinate figure of face characteristic in described acquired image frames;

Obtain the coordinate figure of face characteristic in described reference picture;

Whether the difference between the coordinate figure judging face characteristic in the coordinate figure of face characteristic in described reference picture and described acquired image frames is greater than or equal to described threshold value, if so, performs and described acquired image frames is set to reference picture and storing step; If not, the acquired image frames step containing face-image obtaining described camera collection is performed.

7. a continuous shooting device, is characterized in that, comprising:

Picture frame acquiring unit, for receiving continuous shooting enabled instruction, obtains the start image frame containing face-image of camera collection;

Reference image setting unit, for using described start image frame as with reference to image storing;

Described picture frame acquiring unit, also for obtaining the acquired image frames containing face-image of described camera collection;

Facial features localization unit, for detecting the difference value of the face characteristic information of face characteristic information in described acquired image frames and described reference picture;

Described reference image setting unit, also for when described difference value is more than or equal to threshold value, is set to reference picture by described acquired image frames and stores.

8. device according to claim 7, is characterized in that, described device also comprises:

Amount of images judging unit, described acquired image frames is set to reference picture for described reference image setting unit and after storing, judge whether the quantity of the described picture frame as also storing with reference to image is more than or equal to first threshold, if so, terminates continuous shooting; Otherwise described picture frame acquiring unit obtains the acquired image frames containing face-image of described camera collection.

9. device according to claim 7, is characterized in that, described device also comprises:

Timestamp acquiring unit, after receiving continuous shooting enabled instruction, obtains the initial time stamp of described continuous shooting enabled instruction for described picture frame acquiring unit;

Described timestamp acquiring unit, also for obtaining current time stamp;

Continuous shooting duration judging unit, for calculating continuous shooting duration according to described current time stamp and described initial time stamp, judges whether described continuous shooting duration is more than or equal to Second Threshold, if so, terminates continuous shooting; Otherwise described picture frame acquiring unit obtains the acquired image frames containing face-image of described camera collection.

10. device according to claim 7, is characterized in that, described device also comprises:

Dynamic Graph synthesis unit, described acquired image frames is set to reference picture for described reference image setting unit and after storing, receive Dynamic Graph and generate instruction, obtain as with reference to image the picture frame be stored, according to described as with reference to image and the picture frame be stored generates Dynamic Graph.

11. devices according to claim 7, is characterized in that, described picture frame acquiring unit comprises:

Cache image frame obtains subelement, for obtaining multiple cache image frames containing face-image of described camera collection;

Focusing evaluation of estimate computation subunit, for calculating the focusing evaluation of estimate of each cache image frame in multiple cache image frames described;

Acquired image frames chooses subelement, for choosing in multiple cache image frames described the maximum cache image frame of evaluation of estimate of focusing as described acquired image frames.

12. devices according to claim 7, is characterized in that, described facial features localization unit comprises:

Face characteristic obtains subelement, for obtaining the coordinate figure of face characteristic in described acquired image frames;

Described face characteristic obtains subelement, also for obtaining the coordinate figure of face characteristic in described reference picture;

Face characteristic judgment sub-unit, for judge face characteristic in the coordinate figure of face characteristic in described reference picture and described acquired image frames coordinate figure between difference whether be more than or equal to described threshold value, if so, described acquired image frames is set to reference picture and stores by described reference image setting unit; If not, described picture frame acquiring unit obtains the acquired image frames containing face-image of described camera collection.