CN104639835B - Image tracking system and image tracking method - Google Patents

Abstract

The present invention discloses an image tracking system and an image tracking method, which are used to track a preset object and continuously capture a dynamic image containing the preset object. The image tracking system includes an image capture unit, a storage unit, a control unit, and a rotating device. The image capture unit is used to continuously capture dynamic images; the storage unit temporarily stores dynamic images; the control unit analyzes dynamic images; the rotating device is connected to the control unit for communication, and the image capture unit, the storage unit, and the control unit are mounted on the rotating device; wherein the rotating device receives a driving signal to change the direction in which the image capture unit captures the dynamic image. When the preset object is located in the dynamic image and is not in the preset area, the control unit calculates the current direction in which the image capture unit captures the dynamic image and the angle difference between the current direction in which the preset object is located in the preset area, and outputs a driving signal to drive the rotating device to rotate according to the angle difference.

Description

Image tracking system and image tracking method

technical field

The present invention relates to an image tracking system and an image tracking method, in particular to an image tracking system for automatically tracking preset objects and the executed image tracking method.

Background technique

With the vigorous development of electronic technology, electronic devices such as smart phones have become an indispensable part of people's daily life. Smart phones have camera and video functions.

However, it is inconvenient for the user to take a selfie with the equipped camera or camera device of the smart phone, because the range of taking pictures needs to be limited by the length of a person's hand and the length of people's arms. Therefore, sometimes the person being photographed cannot be fully captured. Moreover, when consumers are taking selfies and pressing the shutter, the smart phone is also prone to shaking, resulting in out-of-focus.

Contents of the invention

In view of the above problems, the main purpose of the present invention is to provide an image tracking system and an implemented image tracking method to solve the problem faced when taking a selfie without requiring the user to hold the camera device.

In order to achieve the above object, an image tracking system proposed by the present invention is used to track a preset object and continuously capture a dynamic image containing the preset object, including an image capture unit, a storage unit, a control unit and a rotating device.

The image capture unit is used to continuously capture the dynamic image; the storage unit is electrically connected to the image capture unit to temporarily store the dynamic image; the control unit is electrically connected to the storage unit, and the control unit analyzes the dynamic image to determine the Whether the preset object enters a preset area in the sampling screen of the image capture unit; the rotating device communicates with the control unit, and the image capture unit, the storage unit, and the control unit are carried on the rotating device above; wherein the rotating device receives a driving signal to change the direction in which the image capture unit captures the dynamic image.

When the preset object is located in the dynamic image and not in the preset area, the control unit calculates the angle difference between the current direction of the image capture unit to capture the dynamic image and the angle difference between the preset object located in the preset area, and outputs The driving signal is used to drive the rotating device to rotate according to the angle difference.

In a specific embodiment, when the preset object enters the preset area, the control unit continuously analyzes the dynamic image to determine whether the preset object is still in the preset area within a predetermined time; if the preset If the object is continuously located in the preset area within the aforementioned predetermined time, the control unit drives the image capture unit to perform a photographing operation.

In a specific embodiment, the control unit calculates and finds the center point of the preset object as a reference point, and judges whether the preset object enters the preset area based on whether the reference point is located in the preset area. in the setting area.

In a specific embodiment, when there are multiple preset objects in the dynamic image, the control unit calculates and finds a common center point of the multiple preset objects as a reference point.

In a specific embodiment, when there are two preset objects in the dynamic image, the control unit calculates and finds the common center point of the two preset objects as a reference point, and the control unit analyzes the two preset objects Whether the distance between the objects is less than a threshold distance; if the control unit judges that the distance between the two preset objects is not less than the threshold distance, the control unit sends a prompt message to prompt the distance between the two preset objects The distance between them is shortened.

The present invention also proposes an image tracking method, which is suitable for a camera device and a rotating device. The camera device has an image capture unit, a storage unit and a control unit, and the image capture unit, the storage unit, the The control unit is mounted on the rotating device, and the rotating device receives a driving signal to change the direction in which the image capturing unit captures the dynamic image; the method includes the following steps:

The image capture unit is used to continuously capture dynamic images, and the storage unit is used to temporarily store the dynamic images;

Analyzing the dynamic image with the control unit to determine whether the preset object enters a preset area in the sampling frame of the image capture unit; and

When the preset object is located in the dynamic image and not in the preset area, the control unit calculates the angle difference between the current direction of the image capture unit to capture the dynamic image and the angle difference between the preset object located in the preset area, and outputs The driving signal is used to drive the rotating device to rotate according to the angle difference.

In a specific embodiment, when the preset object enters the preset area, it is judged whether the preset object is still in the preset area within a predetermined time; if the preset object is within the aforementioned predetermined time, Continuously located in the preset area, the image capture unit is driven to perform the photographing operation.

In a specific embodiment, the image tracking method further includes calculating and finding the center point of the preset object as a reference point, and judging whether the preset object is Enter the preset area.

In a specific embodiment, the image tracking method further includes, when there are multiple preset objects in the dynamic image, calculating and finding a common center point of the multiple preset objects as a reference point.

In a specific embodiment, the image tracking method further includes, when there are two preset objects in the dynamic image, calculating and finding the common center point of the two preset objects as a reference point, and analyzing the two preset objects Whether the distance between the objects is less than a threshold distance; if the distance between the two preset objects is not less than the threshold distance, a prompt message is sent to remind the distance between the two preset objects to be shortened.

By virtue of the image capture system for automatically tracking objects provided by the present invention, when a user wants to take a selfie, the range of the picture can be affected without being affected by the length of the arm. And the mobile app tracks the face or any other object by itself, and calculates and analyzes the focusing action at the same time, which can effectively achieve the selfie effect.

Description of drawings

FIG. 1 is a system block diagram of an image tracking system according to an embodiment of the present invention.

FIG. 2 is a flow chart of the image tracking method in the first embodiment of the present invention.

Fig. 3 is a flow chart of a method in a specific embodiment in the first embodiment of the present invention.

FIG. 4 is a flow chart of the image tracking method in the second embodiment of the present invention.

FIG. 5 is a flow chart of the image tracking method in the third embodiment of the present invention.

FIG. 6 is a flow chart of the image tracking method in the fourth embodiment of the present invention.

FIG. 7 is a flow chart of automatically searching for preset objects in the present invention.

Explanation of reference signs

100 image tracking system

100a Camera device

110 image capture unit

120 storage units

130 control unit

140 Turn device.

Detailed ways

Specific embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings. The same symbols represent components or devices with the same or similar functions.

Please refer to FIG. 1 , which is an image tracking system 100 disclosed in the first embodiment of the present invention, including an image capture unit 110 , a storage unit 120 , a control unit 130 and a rotating device 140 . The image capture unit 110 , the storage unit 120 , and the control unit 130 are integrated into a camera device, such as a smart phone or a tablet computer with a camera function. The image capture unit 110 , the storage unit 120 , and the control unit 130 are carried on the rotating device 140 along with the smart phone 100 a. The rotating device 140 has a motor, which can be a stepping motor or other types of motors, and is used to drive the platform carrying the smart phone 100a to rotate according to one or more axial directions, so as to change the image capturing direction of the image capturing unit 110 . A mobile application is stored in the storage unit 120 for the control unit 130 to load and execute to track preset objects, analyze images and calculate image reference points, so that the image tracking system 100 executes an image tracking method.

Please refer to FIG. 1 and FIG. 2 , the image tracking system 100 is used to execute the image tracking method for tracking a preset object, so that the default object can continuously appear in the stream of dynamic images captured by the image tracking system 100 , so as to continuously execute the video recording operation or photographing operation of the preset object.

The preset object can be a human face, but does not exclude other objects, such as a human body, a car, or other objects that can be photographed.

Analyzing a human face or a specific object in an image, and treating it as the preset object described in the present invention, is the common sense of those skilled in the art, and can be found in existing known technical documents (such as the invention of Taiwan, China) Patent I469063), so the technical details of analyzing human faces or specific objects in images will not be repeated below.

As shown in FIG. 2 , the image tracking system 100 of the first embodiment executes the steps of the image tracking method as follows:

First, the image tracking system 100 establishes a communication link between the control unit 130 and the rotating device 140 , as shown in step S110 . The communication link can be achieved by wired communication means or wireless communication means. For example, taking the smart phone 100a as an example, the smart phone 100a and the rotating device 140 are respectively equipped with Bluetooth modules, and after a pairing procedure, the communication link between the control unit 130 and the rotating device 140 is established with the Bluetooth communication protocol; The connection does not exclude the establishment of the communication link performed by other wireless communication protocols, USB wired connection, IEEE 1394 wired connection.

Next, the control unit 130 turns on a preset object tracking mode, such as a face tracking mode, as shown in step S110a, to activate the image capture unit 110, so that the image capture unit 110 continuously captures a dynamic image, but does not execute it temporarily. Photo assignment.

The main difference between capturing dynamic images and photographing operations is the difference in sampling resolution; capturing dynamic images is to execute multiple static image captures per second and sort them according to time sequence to form dynamic images, and the photographing operation only captures a single static image, so , the photographing operation is usually performed at high resolution, while the capture of dynamic images is performed at relatively low resolution, and the video recording operation is to continuously record the stream of dynamic images in a file. In addition, because the image capture unit 110 may not be in focus, the photographing operation is not performed temporarily, and out-of-focus photos can also be avoided.

The storage unit 120 is electrically connected to the image capture unit 110 . The aforementioned moving image is transmitted to the storage unit 120 to temporarily store the moving image, but it does not rule out that the streaming of the moving image is continuously recorded in a file and becomes a recording operation.

The control unit 130 is electrically connected to the storage unit 120 to receive the dynamic image for analysis, as shown in step S120.

As shown in step S130, the control unit 130 first analyzes whether the preset object is in the dynamic image.

As shown in step S140, then, if the preset object is located in the dynamic image, the control unit 130 further calculates and finds the center point of the preset object as a reference point.

As shown in step S150 , the control unit 130 analyzes whether the preset object enters a preset area in the sampling frame of the image capture unit 110 . A preset area is a custom area that can change its position or shape arbitrarily. In a specific embodiment, judging whether the preset object enters a preset area in the sampling frame of the image capture unit 110 is to determine whether the aforementioned reference point is located in the preset area.

As shown in step S160, if the preset object enters the preset area in the sampling frame of the image capture unit 110, the control unit 130 completes the tracking operation and waits for the next trigger command.

As shown in step S170, if the preset object is located in the dynamic image, but not in the preset area in the sampling frame of the image capture unit 110, the control unit 130 calculates the current direction of the image capture unit 110 to capture the dynamic image, and The angle difference that causes the preset object to be located in the preset area, and output a driving signal. At this moment, the rotating device 140 receives the driving signal and rotates according to the angle difference, so as to change the direction in which the image capture unit 110 captures the dynamic image, so that the preset object enters the preset area in the sampling frame of the image capture unit 110 .

In step S130, if the control unit 130 determines that the preset object is not in the dynamic image after analyzing the dynamic image, then the control unit 130 determines whether the rotating device 140 is rotating, as in step S180.

In step S180, if the rotating device 140 has stopped rotating, return to step S120. If the rotating device 140 is rotating, the control unit 130 continues to output the driving signal to control the rotating device 140 to continue rotating along the original rotating direction, so as to continuously change the direction in which the image capturing unit 110 captures the dynamic image, as in step S190 shown.

Referring to Fig. 3, a specific application of the first embodiment is automatic tracking video recording, which mainly changes the step of turning on the preset object tracking mode in step S110a, such as the step of face tracking mode, into the face tracking mode, And start the video recording function step, such as step S110c. The preset object tracking mode may be tracking of a specific object, or tracking of a human face.

As shown in FIG. 4 , it is the image tracking method of the second embodiment. The first steps of the second embodiment are roughly the same as those of the first embodiment, and the difference is after step S150 .

In step S150, if the preset object enters the preset area of the sampling frame of the image capture unit 110, the control unit 130 completes the tracking operation and triggers a countdown mechanism to wait for a predetermined time, as shown in step S200. When the countdown mechanism is triggered, the control unit 130 can output a display signal at the same time to drive a display unit to display the countdown time, so that the user can understand to avoid moving the default object (such as a human face) during the countdown time, Complete the photoshoot to ensure the tracking results.

As shown in step S210, the control unit 130 continues to analyze the dynamic image to determine whether the preset object is still located in the preset area within the predetermined time.

As shown in step S220, if the preset object continues to be located in the preset area within the aforementioned predetermined time, the control unit 130 drives the image capture unit 110 to perform a photographing operation to capture a relatively high-resolution still image.

As shown in step S230, after the photographing operation is completed, the control unit 130 outputs a control signal to control the rotating device 140 to rotate toward a rotation direction by a preset angle, and then returns to step S130.

In step S210, if the preset object leaves the preset area within the aforementioned predetermined time, the countdown mechanism is interrupted and reset, such as step S240, and returns to step S170 to change the image captured by the image capture unit 110 again. The direction of the dynamic image is taken so that the preset object enters into a preset area in the sampling frame of the image capture unit 110 .

The above-mentioned image tracking methods of the first embodiment and the second embodiment are used to track a single preset object, but the image tracking method of the present invention can track multiple preset objects after a small adjustment.

As shown in FIG. 5 , it is an image tracking method according to a third embodiment of the present invention, which is used to track a plurality of preset objects. The steps are as follows.

First, the image tracking system 100 establishes a communication link between the control unit 130 and the rotating device 140 , as shown in step S110 .

Next, the control unit 130 activates a multi-preset object tracking mode, such as a multi-person tracking mode, as shown in step S110b, to activate the image capture unit 110, so that the image capture unit 110 continuously captures a dynamic image.

The control unit 130 receives the dynamic image for analysis, as shown in step S120.

As shown in step S130, the control unit 130 first analyzes whether the preset object is in the dynamic image.

As shown in step S140a, then, if there are multiple preset objects located in the dynamic image, the control unit 130 further calculates and finds a common center point of the multiple preset objects as a reference point.

As shown in step S150, the control unit 130 determines whether the aforementioned reference point is located in a preset area.

In step S150, if the reference point is located in the preset area, the control unit 130 completes the tracking operation and triggers a countdown mechanism to wait for a predetermined time, as shown in step S200.

As shown in step S210 , the control unit 130 continues to analyze the dynamic image to determine whether the reference point is still located in the preset area within the aforementioned predetermined time.

As shown in step S220, if the reference point is continuously located in the preset area within the aforementioned predetermined time, the control unit 130 drives the image capture unit 110 to perform a photographing operation to capture a relatively high-resolution still image.

As shown in step S230 , after the photographing operation is completed, the control unit 130 outputs a control signal to control the rotation device 140 to rotate toward a rotation direction by a preset angle, and then returns to step S130 .

In step S210, if the reference point leaves the preset area within the aforementioned predetermined time, the countdown mechanism is interrupted and reset, as in step S240, and returns to step S170 to change the image captured by the image capture unit 110 again. The direction of the dynamic image is such that the reference point enters a preset area in the sampling frame of the image capturing unit 110 .

As shown in FIG. 6, it is the image tracking method of the fourth embodiment of the present invention, which is used to track two preset objects, such as two faces, and execute the tracking of the two preset objects, and take pictures, such as Step S110d shows. The main differences between the fourth embodiment and the third embodiment are as follows.

In step S130, if a preset object (face) is detected, the control unit 130 further analyzes whether there are two preset objects (face), as shown in step S132. If there are no two faces, stop the rotation of the rotating device 140, as shown in step S134, and return to step S120.

If two preset objects are tracked in step S132, step S140b is further executed to find the center point of the two preset objects as a reference point.

Next, the control unit 130 also analyzes whether the reference point enters the preset area, as shown in step S150.

As shown in step S152, if the reference point is located in the preset area, the control unit 130 further analyzes whether the distance between the two preset objects is less than a threshold distance.

If the control unit 130 judges that the distance between the two preset objects is not less than the threshold distance, the control unit 130 sends a prompt message, as shown in step S224, to prompt to shorten the distance between the two preset objects, Then return to step S140b.

If the control unit 130 determines that the distance between the two preset objects is less than the threshold distance, the control unit 130 executes S200 to trigger the countdown mechanism and wait for a predetermined time, as shown in step S200 .

As shown in step S210 , the control unit 130 continues to analyze the dynamic image to determine whether the reference point is still located in the preset area within the aforementioned predetermined time.

As shown in step S212, the control unit 130 continues to analyze the dynamic image to determine whether the distance between the two preset objects is less than a threshold distance within the aforementioned predetermined time.

As shown in step S220, if the distance between the two preset objects continues to be smaller than the threshold distance within the aforementioned predetermined time, the control unit 130 drives the image capture unit 110 to perform a photographing operation to capture relatively high-resolution images. still image.

If the distance between the two preset objects is not less than the threshold distance within the aforementioned predetermined time, that is, the distance is too far, then the control unit 130 stops the countdown mechanism, as shown in step S222; then the control unit 130 A prompt message is issued, as shown in step S224, to remind the two default objects that they must be close together, so as to ensure that the two default objects are captured into the desired image. Then the control unit 130 returns to step S140b.

In this way, according to the image tracking method of the fourth embodiment, photos of two preset objects can be taken at the same time, while ensuring that the distance between the two preset objects is not too large.

Referring to FIG. 7 , it is a process of automatically searching for preset objects, which can be arranged after step S110a, S110b, S110c or S110d, and can also be used to replace step S180 and step S190. Alternatively, another independent judging loop is added to the control unit 130 to execute the process of automatically searching for a preset object after a predetermined time (for example, 8 seconds) has elapsed in the moving image.

The trigger condition of this process is mainly that no preset object is detected in the dynamic image, as shown in step S310.

Next, the control unit 130 drives the rotation device 140 to rotate in a first rotation direction in a first axis to an upper limit, as shown in step S320 .

Then, the control unit 130 drives the rotating device 140 to rotate 360 degrees clockwise or counterclockwise on the second axis, as shown in step S330 .

The control unit 130 drives the rotating device 140 to rotate a predetermined angle along a second rotating direction opposite to the first rotating direction in the first axis, as in step S340.

The control unit 130 determines whether the rotation device 140 has rotated to the lower limit of the second rotation direction, as shown in step S350; if not, returns to step S330, and rotates 360 degrees clockwise or counterclockwise on the second axis.

If the rotation device 140 rotates to the lower limit of the second rotation direction, first execute the procedure of rotating 360 degrees clockwise or counterclockwise in the second axis, as shown in step S360; The rotation direction is rotated by the predetermined angle, as in step S370.

Finally, the control unit 130 judges whether the rotating device 140 has reached the upper limit of the first rotating direction of the first axis, as shown in step S380. If it has not entered the upper limit, return to step S360. If it has entered the upper limit, return to step S330.

Before at least one predetermined object is detected, the rotating device 140 repeatedly moves between the upper limit and the lower limit of the rotating track, and stops until a predetermined object is detected.

The above descriptions are only preferred embodiments of the present invention. The scope of the present invention is not limited to the above-mentioned embodiments. All equivalent modifications or changes made by those skilled in the art according to the spirit of the present invention shall be included in the scope of the claims.

Claims (8)

1. An image tracking system, used to track a default object, and continuously capture a dynamic image containing the default object, characterized in that the image tracking system includes: an image capture unit, used to continuously capture the dynamic image; a storage unit electrically connected to the image capture unit to temporarily store the dynamic image; A control unit, electrically connected to the storage unit, firstly enables the default object tracking mode through the control unit, and then continuously captures the dynamic image through the image capture unit, the control unit analyzes the dynamic image, first analyzes the Whether the default object is in the dynamic image, if the control unit judges that the default object is in the dynamic image, then determine whether the default object enters a preset area in the sampling frame of the image capture unit; and A rotating device communicates with the control unit, and the image capture unit, the storage unit, and the control unit are mounted on the rotating device; wherein the rotating device receives a driving signal to change the image capturing unit Taking the direction of the dynamic image, when the preset object enters the preset area, the control unit continues to analyze the dynamic image to determine whether the preset object is still in the preset area within a predetermined time; if the preset If the object continues to be located in the preset area within the aforementioned predetermined time, the control unit drives the image capture unit to perform the photographing operation, and if the preset object leaves the preset area within the aforementioned predetermined time, the image is changed again The capturing unit captures the direction of the dynamic image so that the preset object enters the preset area in the sampling frame of the image capturing unit; Wherein, when the preset object is located in the dynamic image and not in the preset area, the control unit calculates the angle difference between the current direction of the image capturing unit to capture the dynamic image and the preset object in the preset area, And output the driving signal to drive the rotating device to rotate according to the angle difference, if the control unit judges that the preset object is not in the dynamic image, then the control unit judges whether the rotating device is rotating, if the control unit judges that the The rotating device has stopped rotating, then the control unit analyzes the dynamic image, and judges whether the preset object is in the dynamic image, if the control unit judges that the rotating device is rotating, the control unit controls the rotating device to continue to turn. 2. The image tracking system according to claim 1, wherein the control unit calculates and finds the center point of the preset object as a reference point, and the condition is based on whether the reference point is located in the preset area , to determine whether the default object enters the default area. 3. The image tracking system according to claim 2, wherein when there are multiple preset objects in the dynamic image, the control unit calculates and finds a common center point of the multiple preset objects as a reference point . 4. The image tracking system according to claim 2, wherein when there are two preset objects in the dynamic image, the control unit calculates and finds a common center point of the two preset objects as a reference point , the control unit analyzes whether the distance between the two preset objects is less than a threshold distance; if the control unit judges that the distance between the two preset objects is not less than the threshold distance, the control unit sends a prompt message , as a hint to shorten the distance between the two default objects. 5. An image tracking method, applicable to a camera device and a rotating device, the camera device has an image capture unit, a storage unit and a control unit, and the image capture unit, the storage unit, the control unit Riding on the rotating device, the rotating device receives a driving signal to change the direction in which the image capture unit captures the dynamic image; it is characterized in that the method includes the following steps: Open a default object tracking mode through the control unit first; The image capture unit is used to continuously capture dynamic images, and the storage unit is used to temporarily store the dynamic images; Using the control unit to analyze the dynamic image, first analyze whether the preset object is in the dynamic image, if the control unit judges that the preset object is in the dynamic image, then judge whether the preset object enters the image capture unit a preset area in the sampled image; When the preset object enters the preset area, it is judged whether the preset object is still located in the preset area within a predetermined time; if the preset object continues to be located in the preset area within the aforementioned predetermined time, Drive the image capture unit to take pictures, if the preset object leaves the preset area within the aforementioned predetermined time, then change the direction of the image capture unit to capture the dynamic image again, so that the preset object enters In the preset area in the sampling frame of the image capture unit; When the preset object is located in the dynamic image and not in the preset area, the control unit calculates the angle difference between the current direction of the image capture unit to capture the dynamic image and the angle difference between the preset object located in the preset area, and outputs The driving signal is used to drive the rotating device to rotate according to the angle difference; If the control unit determines that the preset object is not in the dynamic image, then the control unit determines whether the rotating device is rotating; If the control unit determines that the rotating device has stopped rotating, the control unit analyzes the dynamic image and determines whether the preset object is in the dynamic image; and If the control unit determines that the rotating device is rotating, the control unit controls the rotating device to continue to rotate forward. 6. The image tracking method according to claim 5, further comprising: calculating and finding the center point of the preset object as a reference point, and taking whether the reference point is located in the preset area as a condition , to determine whether the default object enters the default area. 7. The image tracking method according to claim 6, further comprising, when there are multiple preset objects in the dynamic image, calculating and finding a common center point of the multiple preset objects as a reference point . 8. The image tracking method according to claim 6, further comprising, when there are two preset objects in the dynamic image, calculating and finding a common center point of the two preset objects as a reference point , and analyze whether the distance between the two default objects is less than a threshold distance; if the distance between the two default objects is not less than the threshold distance, a prompt message is sent to remind the distance between the two default objects distance is shortened.