TW201410016A - Linking-up photographing system and control method for cameras thereof - Google Patents

Abstract

A linking-up photographing system and a control method for cameras thereof are provided. The linking-up photographing executes the control method for cameras, and the control method for cameras includes steps below. A first image is obtained by a first camera. A second image is obtained by a second camera, the field of view of the first image covering part of the field of view of the second image. A control look-up table is established according to the first image and the second image. An object indicating instruction which indicates a ROI (region of interest) of the first image is received. The field of view of the second camera is adjusted according to the ROI and the control look-up table, so that the second camera photographs to the field of view and outputs a third image. Therefore, the second camera links up to the first camera easily.

Description

Linkage photography system and control method of multiple cameras

The present invention relates to a linked photographic system and a multi-camera control method thereof, and more particularly to a linked photographic system capable of quickly interlocking a plurality of cameras and a multi-camera control method thereof.

With the popularization of monitoring equipment, more and more people will install monitoring equipment to protect their property and personal safety. In order to be able to monitor without a blind spot, there are manufacturers that propose a camera that can be panned. The implementation of the panoramic camera can be achieved by joining multiple sets of lenses, or by using a single fisheye lens. The advantage of panoramic shooting is that the captured digital image can be viewed in all directions and without dead ends.

Although the panoramic camera has the advantage of wide-angle shooting, the panoramic camera cannot perform optical zoom processing. Therefore, the viewer cannot perform scaling of the image from any of the viewing targets in the panoramic image. Although the panoramic camera can zoom in and out of the viewing target by digital zoom. However, such a processing method is only using interpolation processing of pixels, so that the image resolution after digital zooming is greatly reduced, which causes a problem that the digital image has poor image quality. If something is happening far from the panoramic camera Therefore, it is difficult for the viewer to view a clear viewing target from the panoramic image.

Compared to a panoramic camera, a pan-tilt-zoom (PTZ) can perform optical zoom processing. Therefore, the swivel zoom camera can zoom in the shooting field so that the distant viewing target can be enlarged. However, the shooting field of the rotary zoom camera is smaller than that of the panoramic camera. So someone tried to use a panoramic camera with another rotary zoom camera to capture the details of the same scene.

However, since the angles of the cameras at different locations may not be exactly the same, if the other rotary zoom camera in the environment is driven from the viewpoint of a panoramic camera in the environment, the field of view may have an error. Moreover, the existing matching method needs to calculate the corresponding position of each pixel in the image captured by the two cameras in real time, so it requires a lot of computing resources and computing time. Even if the calculation of each pixel is replaced by sampling, there is still a need for a short reaction time. Especially when the rotary zoom camera needs to perform rotation and zooming on different axes at the same time, more complicated calculations are required to obtain the parameters required to control the rotary zoom camera.

In view of the above problems, the present invention provides a linked photography system and a control method for the same. The linked photography system includes a first camera, a second camera, a calibration unit, and a control unit; and the linked photography system can perform a multi-camera control method.

The multi-camera control method can include the following steps. Use one The first camera gets a first image. A second image is obtained by using a second camera, wherein the field of view of the first image at least partially covers the field of view of the second image. A control correspondence table is established according to the first image and the second image. A specified instruction is received, wherein the specified instruction specifies a region of interest (ROI) in the first image. Adjusting the shooting field of the second camera according to the region of interest and the control correspondence table, so that the second camera captures the region of interest and outputs a third image.

In the linked camera system, the first camera is used to obtain the first image, and the second camera is used to obtain the second image, wherein the field of view of the first image at least partially covers the field of view of the second image. The calibration unit is configured to establish a control correspondence table according to the first image and the second image. The control unit is configured to receive the specified instruction, wherein the specified instruction specifies the region of interest in the first image; and adjust the shooting field of the second camera according to the region of interest and the control correspondence table, so that the second camera captures the region of interest and outputs the third image.

In summary, the linkage photography system and the multi-camera control method can use the first image and the second image to establish a control correspondence table, and directly obtain the parameters corresponding to the region of interest in a table lookup manner and obtain a third image. Therefore, the linked photography system and its multi-camera control method can quickly link multiple cameras and reduce the reaction time that the user has to wait.

22‧‧‧First camera

24‧‧‧Second camera

26‧‧‧ calibration unit

28‧‧‧Control unit

30‧‧‧Fourth image

31‧‧‧Reference axis

32‧‧‧ Fifth image

33‧‧‧Reference axis

34‧‧‧Partial panorama

35‧‧‧Reference axis

40‧‧‧ first image

41‧‧‧ first central point

42‧‧‧second second point

43‧‧‧Location points

44, 44a, 44b, 44c, 44d‧‧‧ punctuation

45‧‧‧First reference circle

50‧‧‧second image

51‧‧‧ First point

52‧‧‧ second central point

53‧‧‧Location points

55‧‧‧Second reference circle

56‧‧‧Zoom angle

60‧‧‧interest area

61‧‧‧ third central point

62, 62a, 62b‧‧‧ boundary points

71‧‧‧ Third point

72, 72a, 72b‧‧‧

1 is a block diagram of a linked photography system according to an embodiment of the present invention. Figure.

2A is a schematic view showing the erection of the first camera and the second camera according to an embodiment of the present invention.

2B is a schematic view showing the erection of the first camera and the second camera according to another embodiment of the present invention.

Fig. 3 is a flow chart showing a method of controlling a multi-camera according to an embodiment of the present invention.

Figure 4 is a flow chart of step S200 of an embodiment of the present invention.

Figure 5 is a schematic diagram of a fourth image of an embodiment of the present invention.

Figure 6 is a flow chart of step S300 of an embodiment of the present invention.

Figure 7 is a schematic diagram of an anchor point according to an embodiment of the present invention.

Figure 8 is a flow chart of step S340 of an embodiment of the present invention.

Figure 9 is a schematic view of a rounded corner coordinate of an embodiment of the present invention.

Figure 10 is a schematic diagram of an interpolation method according to an embodiment of the present invention.

Figure 11 is a flow chart showing the step S500 of an embodiment of the present invention.

Figure 12A is a flow chart of step S510 of an embodiment of the present invention.

Figure 12B is a flow chart of step S510 of another embodiment of the present invention.

Figure 13 is a schematic view of an edge map of an embodiment of the present invention.

Figure 14 is a schematic view of a second reference circle in accordance with an embodiment of the present invention.

The detailed features and advantages of the present invention are described in detail below in the Detailed Description of the invention. The technical content is embodied in accordance with the present disclosure, and the objects and advantages of the present invention can be readily understood by those skilled in the art in light of the disclosure. The following examples are intended to describe the present invention in further detail, but are not intended to limit the scope of the invention.

The invention provides a linked photographic system and a multi-camera control method thereof, wherein the linked photographic system can perform a multi-camera control method. First, please refer to "FIG. 1", which is a block diagram of a linked photography system according to an embodiment of the present invention. The linked photography system includes a first camera 22, a second camera 24, a calibration unit 26, and a control unit 28. The first camera 22, the second camera 24, and the calibration unit 26 are connected to the control unit 28, respectively. The first camera 22 and the second camera 24 can respectively capture images from their field of view, and the control unit 28 can obtain a first image by using the first camera 22 and a second image by using the second camera 24. image.

The first camera 22 and the second camera 24 may be, for example, a set of lenses plus a charge-coupled device (CCD), a set of lenses plus a complementary metal-oxide-semiconductor (CMOS). Or an Internet Protocol camera (IP camera). Furthermore, the type of the first camera 22 may be a fisheye camera, a panoramic camera or a wide-angle camera. The following is a fisheye camera, so the first image is a panoramic image (or a fisheye image). ). In general, panoramic images can cover 360 degrees. Shoot the field of view. The second camera 24 can be a pan-tilt-zoom camera (PTZ camera) or a digital zoom camera.

The field of view of the first camera 22 at least partially covers a portion of the field of view taken by the second camera 24. To this end, the first camera 22 and the second camera 24 may be erected adjacently, and the first camera 22 and the second camera 24 may be erected at a predetermined distance. For example, in the same room, the first camera 22 can be placed adjacent to the second camera 24 in an adjacent position, as shown in "Fig. 2A". Further, the first camera 22 and the second camera 24 may be disposed at different positions in the room as shown in "FIG. 2B".

The calibration unit 26 and the control unit 28 can be implemented by a personal computer, a network video recorder (DVR), an embedded system, or an electronic device with computing power. In addition to the above components, the linked photography system may also include a storage unit to temporarily store or store the captured image, and the storage unit may be connected to the control unit 28. The storage unit may be, for example, a random access memory (RAM), a flash memory, or a hard disk.

The control unit 28 can also be coupled to a display to display images captured by the first camera 22 or the second camera 24 or to display images processed by the control unit 28. And the linked photography system can be connected to the network through a communication unit and connected to a server or a remote display through the network to increase the flexibility of the linked photography system. According to an embodiment, the calibration unit 26 And the control unit 28 can be configured on the same computer or server, and then connected to the first camera 22 and the second camera 24 via the network.

Next, please refer to "FIG. 3", which is a flowchart of a method for controlling a multi-camera according to an embodiment of the present invention. Through the multi-camera control method, the linked camera system can cause the second camera 24 to perform panning, tilting or zooming according to the first camera 22 and a specified command from the user. The magnification, in turn, causes the second camera 24 to capture a third image.

First, the first image is obtained by the first camera (step S100), and the second image is obtained by the second camera, wherein the captured field of view of the first image at least partially covers the captured field of view of the second image (step S200). Please refer to FIG. 4 together, which is a flowchart of step S200 according to an embodiment of the present invention. In step S200, the calibration unit 26 or the control unit 28 may perform the following steps S210 and S220 to obtain a second image.

The calibration unit 26 first captures a plurality of fourth images for different shooting fields by using the second camera 24 (step S210), and the fourth images may not be panoramic images but planar images. Then, the calibration unit 26 performs panorama stitching on the fourth image to generate a panoramic image, and uses the panoramic image as the second image (step S220), as shown in FIG.

The panoramic view is used to splicing a plurality of fourth images 30 captured by the second camera 24. As shown in Figure 5, a number of different perspectives The four images 30 are respectively captured, and the viewing angles covered by the fourth images 30 are adjacent backgrounds, and part of the image content may overlap in the fourth images 30. Then, the fourth images 30 are subjected to a distortion process, and a corresponding plurality of fifth images 32 are obtained. Since the first image 22 is a panoramic image taken by a fisheye lens or a wide-angle lens, the second image obtained by twisting the fourth image 30 into the fifth image 32 and then performing the panoramic image processing will be The first image is similar.

In general, the panoramic view may be joined according to a plurality of sets of features in the plurality of warped fifth images 32, and the features may be specific objects or scenes in the image. Therefore, the two fifth images 32 can be docked by scene or object when merging. After the aforementioned bonding process, a partial panoramic view 34 can be obtained. Similarly, for other areas of the captured field of view in the first image, the second camera 24 can take a picture and perform corresponding processing. All the panoramic views 34 are also subjected to the aforementioned bonding process to obtain a complete panoramic image, and the combined panoramic image is taken as the second image.

Since the fourth image 30 is twisted and deformed to become the fifth image 32, the horizontal line on each of the fourth images 30 is converted into an arc. For example, in the "figure 5", a reference axis 31 penetrating the fourth image 30 in the horizontal direction is horizontal, but after twisting, the reference axis 33 of the fifth image 32 has become a circular arc shape, and the panoramic portion of the joined portion is obtained. The reference axis 35 in Fig. 34 naturally also maintains the same circular arc shape as the reference axis 33.

After obtaining the first image and the second image, the calibration unit 26 A control correspondence table is established according to the first image and the second image (step S300). The calibration unit 26 can receive the first image or the fourth image directly through the control unit 28; and if the second image is generated by the control unit 28, the calibration unit 26 can also receive the second image directly from the control unit 28.

Please refer to FIG. 6 and FIG. 7 together, which are respectively a flowchart of step S300 and a schematic diagram of the positioning point according to an embodiment of the present invention.

The first image 40 has a first center point 41, and the second image 50 has a second center point 52. The first center point 41 corresponds to a first center point 51 of the second image 50, and the second center point 52 corresponds to a second intermediate point 42 in the first image 40. The first midpoint 51 can be viewed as a mapping point that maps the first center point 41 into the second image 50; the second center point 42 can be viewed as mapping the second center point 52 into the first image 40. Projection point. Since the first camera 22 and the second camera 24 are necessarily disposed at different positions, the first image 40 and the second image 50 must have the same field of view. In the second image 50, the positions of the first center point 51 and the second center point 52 are also different. And as the configuration spacing is further apart, the distance between the first center point 51 and the second center point 52 in the second image 50 is greater. Similarly, in the first image 40, the positions of the second midpoint 42 and the second center point 52 must also be different.

The calibration unit 26 can respectively set the M positioning points 43 and 53 corresponding to the positions in the first image 40 and the second image 50. Where M is greater than Or a positive integer equal to 3 (step S310), for example, M may be 50 or 100. The positioning points 43 and 53 are reference points corresponding to the same position in the first image 40 and the second image 50. As described above, since the arrangement positions of the first camera 40 and the second camera 50 are different, the same object in the shooting field of view may also fall in different positions in the first image 40 and the second image 50. For example, the coordinates of the upper right corner of the same picture in the first image 40 are (160, 180), but the coordinates of the same point in the first image 40 are (270, 80).

The method of selecting the reference point can be selected by the user, and the corresponding point 43 and 53 in the first image 40 and the panoramic image 50 can also be found through the feature point extraction and matching technique. s position.

The calibration unit 26 selects three positioning points 43 from the first image 40 as the positioning point group from the M positioning points 43 of the first image 40. A total of N sets of anchor points are generated, and N is a positive integer (step S320). More specifically, N is the number of combinations of three positioning points 43 taken from M positioning points 43 at a time. . Assuming M is 4, the calibration unit 26 can sequentially use the first, second, and third positioning points 43 as the first positioning point group from the four positioning points 43; and the first, second, and fourth positioning points 43 As the second positioning point group; the first, third, and fourth positioning points 43 are used as the third positioning point group; the second, third, and fourth positioning points 43 are used as the fourth positioning point group; = 4 anchor point groups.

Next, the calibration unit 26 calculates N coordinate conversion parameter groups respectively corresponding to the N positioning point groups (step S330). For N sets of positioning points, the calibration unit 26 can perform affine transformation according to the corresponding positioning points 43 and 53 one by one to calculate a coordinate conversion parameter set. The position of the M positioning points 43 in the first image 40 and the position of the positioning point 53 in the second image 50 are regarded as known numbers. The affine transformation is based on the following equation:

Where (X _i , Y _i ) is the coordinate of the anchor point 43 of the first image 40, and (X _o , Y _o ) is the coordinate of the anchor point 53 of the second image 50, (a1, a2, a3, b1, b2) , b3) is the coordinate conversion parameter group corresponding to the set of positioning points 43 and 53, that is, the value of each element of the conversion matrix in "Formula 1". According to the N positioning point groups, the corresponding N coordinate conversion parameter groups can be respectively calculated. In a subsequent processing step, the control unit 28 may map any point of the first image 40 to the second image 50 according to the calculated coordinate conversion parameter set. In other words, according to the coordinate conversion parameter group, the corresponding position in the second image 50 obtained by the bonding can be calculated for any point in the first image 40 obtained directly.

Calibration unit 26 can then set a plurality of coordinate points 44 in first image 40. According to an embodiment, all pixels other than the anchor point 43 can be set as the coordinate point 44. According to another embodiment, in order to save the amount of calculation and the operation time, only a part of the pixels may be set as the coordinate point 44 in a sampling manner. For example, one pixel can be taken as a coordinate every 8×8 or 16×16 pixels. Point 44, as shown in Figure 7.

For each of the coordinate points 44 in the first image 40, the calibration unit 26 may perform a conversion process according to one of the N coordinate conversion parameter sets to obtain a shooting position parameter corresponding to the second camera 24 of each coordinate point 44 ( Step S340), the shooting position parameter is stored in the control correspondence table (step S350). In other words, the calibration unit 26 may sequentially select a coordinate point 44 to perform the conversion process in step S340, and may confirm whether all the coordinate points 44 have been processed after the conversion program is executed. If there are still unprocessed coordinate points 44, the conversion process continues to be performed among the unprocessed coordinate points 44 until the conversion procedure has been performed for all of the coordinate points 44.

Please refer to FIG. 8 , which is a flowchart of step S340 according to an embodiment of the present invention. According to the coordinate point 44 currently being targeted, one of the N anchor point groups is selected from the first image 40 (step S341). Furthermore, the selected set of anchor points may be formed by three anchor points 43 in the first image 40 that are closest in line to the coordinate point 44 currently being targeted. The calibration unit 26 can calculate the sum of the distances between the three anchor points 43 in each of the set of anchor points and the currently-targeted coordinate points 44, and select the set of anchor points corresponding to the smallest sum.

Then, according to the coordinate conversion parameter group corresponding to the selected positioning point group, the mapping point of the currently-targeted coordinate point 44 in the second image can be calculated (step S342), and the shooting position parameter corresponding to the mapping point is obtained (step S343). As described above, the coordinate conversion parameter set may represent a conversion matrix; the calibration unit 26 may multiply the coordinate of the coordinate point 44 currently being targeted by the conversion matrix to obtain a pair. The coordinates of the mapped point.

According to various embodiments, the coordinate points 44 and the coordinates of the mapped points are in the form of rectangular coordinates (x, y) or in the form of rounded corners (pan, tilt). In the form of a rounded corner coordinate, the shooting position parameter may include a tilt angle and a pan angle of the second camera, and the tilt angle and the pan angle correspond to the rounded coordinates of the mapped point respectively. ,tilt) value. For the right angle coordinates and the rounded corner coordinates, please refer to "9th drawing", which is a schematic diagram of the rounded coordinates of an embodiment of the present invention. Cartesian coordinates and rounded coordinates can be interchanged by the formula of a trigonometric function. For example, a rectangular coordinate (x, y) can be converted into a rounded coordinate (pan, tilt) by the following formula:

In short, the calibration unit 26 calculates the mapping points corresponding to each coordinate point 44 in step S340, and can store the rounded coordinates of these mapping points as shooting position parameters in the control correspondence table. By controlling the correspondence table, the control unit 28 can quickly look up the table to the mapping point of each coordinate point 44 corresponding to the second image 50.

After establishing the control correspondence table, the control unit 28 can continually confirm whether a specified instruction received by the user has been received. If so, the control unit 28 receives the specified instruction; wherein the specified instruction specifies a Region of Interest (ROI) in the first image (step S400). According to an embodiment, The linked camera system provides a graphical user interface (GUI) to allow the user to intuitively enter specified commands. For example, the region of interest may be circular, and the user may drag a fixed-size circle to set the region of interest. The graphical user interface also allows the user to set the center of the area of interest and the length of the radius. However, the region of interest may also be a rectangular rectangle, a rounded rectangle or an ellipse, which is not limited herein.

The shooting field of view of the second camera 24 is adjusted in accordance with the region of interest and the control correspondence table, and the control unit 28 causes the second camera 24 to capture the region of interest and output a third image (step S500). The control unit 28 can control the second camera 24 by using the tilt angle and the pan angle, and the second camera 24 directly rotates to the set angle in the axial direction corresponding to the tilt angle and the pan angle, and then moves the shooting field in this state. Capture into a third image.

In the control method of the multi-camera, steps S100 to S300 can be regarded as a pre-calibration step, and steps S400 and S500 can control the second camera 24 in real time according to the user's instruction and the control correspondence table. Control unit 28 may repeatedly perform steps S400 and S500 in response to a plurality of designated instructions.

Please refer to FIG. 10, which is a flowchart of step S500 according to an embodiment of the present invention.

The control unit 28 firstly controls the correspondence table according to a third center point of the region of interest to obtain the shooting position parameter of the corresponding region of interest (step S510); and then causes the second camera 24 to act according to the captured shooting position parameter and extract the output. The third image (step S520). When the area of interest is circular, The third center point will be the center of the circle. If the position of the third center point is exactly at one of the coordinate points 44, the control correspondence table can be queried according to the coordinate point 44, and the shooting position parameters such as the inclination angle and the pan angle of the coordinate point 44 are obtained.

If the position of the third center point is not at any of the coordinate points 44, the coordinate point of the coordinate point 44 adjacent to the third center point can be queried in the control correspondence table to calculate the corresponding third center point by interpolation. Parameters (including tilt and pan angle). Please refer to FIG. 10, which is a schematic diagram of an interpolation method according to an embodiment of the present invention. It is assumed that the third center point 61 falls between the four coordinate points 44a, the coordinate points 44b, the coordinate points 44c, and the coordinate points 44d adjacent to each other, and two adjacent coordinate points 44 (for example, the coordinate points 44a and the coordinate points 44b) The distance between them is 1 unit. And assume that the distance between the third center point 61 on the horizontal axis and the coordinate point 44a is m, and the distance between the third center point 61 on the vertical axis and the coordinate point 44a is n, where m and n are both A positive number less than one. In this way, the shooting parameters (inclination angle and panning angle) of the third center point 61 can be calculated by the formula of the following interpolation method: P = [ A × (1 - m ) + B × m ] × (1 - n ) + [ C × (1- m ) + D × m ] × n (Formula 4). Where P is the shooting parameter (inclination angle and panning angle) of the third center point 61, and A, B, C, and D are the shooting parameters of the coordinate point 44a, the coordinate point 44b, the coordinate point 44c, and the coordinate point 44d, respectively (inclination and shaking) Angle of view). And P, A, B, C, and D can be expressed in a format of a pair (pan, tilt).

As described above, the control unit 28 first looks up the table to obtain the corresponding inclination angle and panning of the coordinate point 44a, the coordinate point 44b, the coordinate point 44c, and the coordinate point 44d. The angle and the interpolation method are used to calculate the inclination angle and the pan angle corresponding to the third center point 61. In this way, the control unit 28 can rotate the second camera 24 to the set angle in the axial direction corresponding to the tilt angle and the pan angle according to the position of the third center point, and then take the action into the captured field of view. Into the third image.

According to an embodiment, the user can change the size of the region of interest to cause the second camera 24 to perform an optical zoom or a digital zoom. That is to say, the scaling parameter of the second camera 24 can be changed depending on the region of interest. In this embodiment, in addition to the tilt angle and the pan angle, the shooting position parameter further includes a zoom angle for indicating the zoom parameter. The zoom parameter may be a zoom ratio, a focus, or a viewing angle when the image is captured by the lens. The zoom ratio is a relative magnification calculated according to the focal length range of the lens group; the focal length or the angle of view is an absolute value, which can be set and controlled for different lens groups.

Please refer to FIG. 12A and FIG. 13 together, which are respectively a flowchart of step S510 and a schematic diagram of the edge points according to an embodiment of the present invention. In the present embodiment, the region of interest 60 is circular. As shown in FIG. 13, the region of interest 60 includes a third center point 61 and a plurality of boundary points 62, wherein the boundary points 62 surround the boundary of the region of interest 60. The third center point 61 corresponds to a third center point 71 of the second image 50, and each of the boundary points 62 also corresponds to the plurality of side points 72 of the second image 50.

The control unit 28 can calculate the third midpoint 71 of the third center point 61 in the second image 50, and the plurality of boundary points 62 in the second image 50. The edge point 72 is reflected (step S511). The coordinates of the third midpoint 71 and the edge reflection point 72 can be quickly obtained according to the control correspondence table or the table matching interpolation method in the foregoing manner.

The control unit 28 can form an angle according to the edge point 72a, the second center point 52 and the third center point 71 or the edge point 72b, the second center point 52 and the third center point 71. And the edge reflection point 72b is obtained by comparing all the edge reflection points 72 to form a point with the largest acute angle, and one of the left and right sides is obtained, and the larger one is multiplied by 2 to obtain a zoom angle 56 (step S512); The zoom angle cannot be directly formed by the edge map 72a, the second center point 52, and the edge map point 72b. The reason is that the edge map points are obtained by the boundary point lookup table and are composed of the edge map points generated by the corresponding lookup table. The pattern is not symmetrical (relative to the line formed by the second center point 52 and the third center point 71), so it is directly composed of the edge point 72a, the second center point 52, and the edge point 72b. The zoom range corresponding to the lens field of view will not completely cover the edge map 72a and the edge map 72b, so take the maximum angle and multiply by two, so that the center point of the third image is the third midpoint. 71 (third center point 61), the left and right sides of the third image will cover the edge point 72a (boundary point 62a) and the edge 72b (boundary point 62b). Then, the control correspondence table is queried according to the third intermediate point 71 to obtain a shooting position parameter corresponding to the interest area 60 (step S513), wherein the shooting position parameter includes a tilt angle and a pan angle. It should be noted that the zoom angle 56 is obtained by the instant calculation according to the interest area 60 as described above, instead of the query control correspondence table.

As shown in Figure 5, the horizontal lines in the fourth image 60 are Will be distorted into a circular arc. And all the fifth images 32 captured and twisted by different pan angles are twisted to form a ring image in the second image 50, and the reference axis 33 in the ring image is connected to a complete image. The circle is the second reference circle 55 shown in "Fig. 13" and "Fig. 14".

The larger the region of interest 60 is, the larger the distance between the edge points 72a and 72b intersecting the second reference circle 55 is, the larger the zoom angle 56 is, and the larger the field of view is encompassed in the region of interest 60. The second camera 24 can adjust its zooming parameters to obtain a larger field of view; or can increase its zooming parameters to obtain a narrower field of view, but can obtain more detailed information for the remote object or object details. Therefore, the zoom angle 56 can be used to indicate the zoom parameter when the third image is captured.

Please refer to FIG. 12B and FIG. 13 together, which are respectively a flowchart of step S510 and a schematic diagram of the edge points of another embodiment of the present invention. In the present embodiment, the region of interest 60 is circular.

In step S510, the control unit 28 may first calculate the second center point 42 of the second center point 52 in the first image 40, and the third center point 71 of the third center point 61 in the second image 50. (Step S515). Similarly, the coordinates of the third midpoint 71 and the second midpoint 42 can be quickly obtained according to the control correspondence table or the look-up table plus interpolation method in the foregoing manner.

The control unit 28 can calculate the first reference circle 45 with the second center point 42 as the center and the distance between the second center point 42 and the third center point 61 as a radius, wherein the first reference circle 45 and the boundary points Two of 62 Intersect (step S516). In "Fig. 13", the boundary points 62a and 62b intersect the first reference circle 45.

Next, the control unit 28 calculates the edge reflection point 72a and the edge reflection point 72b of the two boundary points 62a and 62b intersecting the first reference circle 45 in the second image 50 (step S517), and according to the edge reflection point 72a and the edge The reflection point 72b and the second center point 52 calculate the zoom angle 56 (step S518). The edge map 72a and the edge map 72b obtained in step S517 may be different from the edge map 72a and the edge map 72b obtained in step S512.

After the zoom angle 56 is obtained, the control unit 28 queries the control correspondence table according to the third center point 71 to obtain the shooting position parameter of the corresponding interest area 60 (step S519), wherein the shooting position parameter includes the tilt angle and the pan angle. The calculation of the zoom angle 56 only uses the boundary point 62a and the boundary point 62b intersecting the first reference circle 45 and the corresponding edge reflection point 72a and edge reflection point 72b, without calculating other boundary points 62 or side reflection points. 72. Therefore, the zoom angle 56 can be quickly calculated and maintained with high accuracy.

In this way, according to the control correspondence table first established by the calibration unit 26 and the interest area 60 designated by the user, the control unit 28 can instantly and quickly calculate the corresponding shooting position parameters such as the tilt angle and the pan angle, and then The second camera 24 operates to capture a third image in which the field of view corresponds to the region of interest 60. Furthermore, the field of view of the third image may be substantially the same as the region of interest 60, but with higher image resolution and more image detail.

In summary, the linked camera system and its multi-camera control method can generate a panoramic image (ie, the first image) by using a fisheye camera, a panoramic camera or a wide-angle camera, and generate another shooting by using a rotary zoom camera. The field of view has overlapping panoramic images (ie, the aforementioned second image). The control system of the linked camera system and the multi-camera establishes a control correspondence table according to the correspondence relationship between the two panoramic images, and saves a large number of operations which may be required when the shooting position parameter is actually obtained according to the specified instruction in the manner of table lookup. .

The linkage photography system and the control method of the multi-camera can directly obtain the inclination angle and the pan angle corresponding to the region of interest in a table lookup manner. Moreover, by looking up the table and using the first reference circle, the zoom angle corresponding to the region of interest can also be quickly calculated. Therefore, the linked photography system and the multi-camera control method can quickly link a plurality of cameras, thereby reducing the reaction time that the user has to wait.

Although the present invention has been disclosed above in the foregoing embodiments, it is not intended to limit the invention. It is within the scope of the invention to be modified and modified without departing from the spirit and scope of the invention. Please refer to the attached patent application for the scope of protection defined by the present invention.

Claims (22)

A method for controlling a multi-camera includes: obtaining a first image by using a first camera; obtaining a second image by using a second camera, wherein a field of view of the first image at least partially covers a field of view of the second image; Establishing a control correspondence table according to the first image and the second image; receiving a specified instruction, wherein the specified instruction specifies an area of interest in the first image; and adjusting the second according to the area of interest and the control correspondence table The camera's field of view allows the second camera to capture the region of interest and output a third image. The method for controlling a plurality of cameras according to claim 1, wherein the step of obtaining the second image by using the second camera comprises: using the second camera to capture a plurality of fourth images for different shooting fields; The fourth images are panoramicly connected to generate a panoramic image, and the panoramic image is used as the second image. The method for controlling a plurality of cameras according to claim 1, wherein the first image includes a plurality of coordinate points, and the step of establishing the control correspondence table according to the first image and the second image comprises: at the first The M positioning points corresponding to the positions respectively set in the image and the second image, wherein M is a positive integer greater than or equal to 3; From the M positioning points of the first image, three of the positioning points are respectively selected as an positioning point group according to all combinations, wherein a total of N positioning point groups are generated, and N is a positive integer; N coordinate conversion parameter groups respectively corresponding to the positioning point groups; for each of the coordinate points in the first image, performing a conversion procedure according to one of the N coordinate conversion parameter groups, to obtain corresponding to each of the coordinate points a shooting position parameter of the second camera; and storing the shooting position parameters in the control correspondence table. According to the control method of the camera of the third application of the patent scope, wherein the conversion program performs the following steps one by one for the coordinate points: one of the N positioning point groups is selected according to the coordinate point currently targeted; according to the selected positioning The coordinate conversion parameter group corresponding to the point group calculates a mapping point of the coordinate point currently in the second image; and obtains the shooting position parameter corresponding to the mapping point. According to the control method of the camera of the fourth application of the patent scope, the selected positioning point group is formed by three positioning points which are the closest to the linear point of the coordinate point currently targeted. According to the control method of the camera of claim 3, wherein the shooting position parameter includes a tilt angle of the second camera and a pan angle. According to the control method of the camera of the first aspect of the patent application, wherein the shooting field of view of the second camera is adjusted according to the region of interest and the control correspondence table, so that the second camera captures the region of interest and outputs the third image. The step of: querying the control correspondence table according to a third center point of the interest area to obtain a shooting position parameter corresponding to the interest area; and causing the second camera to perform the shooting position parameter according to the query and extracting the output The third image. According to the control method of the multi-camera of claim 7, wherein the second image includes a second center point, the interest area includes a plurality of boundary points, and the third center point according to the interest area queries the control corresponding The step of obtaining the shooting position parameter corresponding to the region of interest includes: calculating a third center point of the third center point in the second image, and the plurality of boundary points in the second image Calculating a zoom angle according to the edge points, the second center point, and the third center point; and querying the control correspondence table according to the third center point to obtain a corresponding interest area The shooting position parameter, wherein the shooting position parameter includes a tilt angle and a pan angle. According to the control method of the multi-camera of claim 7, wherein the region of interest is a circle, and the second image includes a second center point, the interest The area includes a plurality of boundary points, and the step of querying the control correspondence table according to the third center point of the interest area to obtain the shooting position parameter corresponding to the interest area includes: calculating the second center point at the first a second midpoint in the image and a third midpoint of the third center point in the second image; the second center point is centered, and the second center point and the first point The distance between the three center points is a radius, and a first reference circle is calculated, wherein the first reference circle intersects two of the boundary points; and the second boundary point intersecting the first reference circle is calculated in the second Two edge reflection points formed in the image; calculating a zoom angle according to the two edge map points, the second center point, and the third center point; and querying the control correspondence table according to the third center point to The shooting position parameter corresponding to the region of interest is obtained, wherein the shooting position parameter includes a tilt angle and a pan angle. According to the control method of the multi-camera of the eighth application of the patent application, the step of calculating the zoom angle according to the edge points, the second center point, and the third center point includes: calculating each of the edge points, a plurality of angles formed by the second center point and the third center point, and finding a maximum acute angle between the angles; and multiplying the maximum acute angle by 2 as the zoom angle. According to the control method of the multi-camera of claim 9, wherein the step of calculating a zoom angle according to the two edge points, the second center point, and the third center point includes: calculating each of the side points a plurality of angles formed by the second center point and the third center point; and the angle of the largest one of the angles is multiplied by 2 as the zoom angle. A linked camera system includes: a first camera for obtaining a first image; and a second camera for obtaining a second image, wherein the first image captures at least partially the second image a field of view; a calibration unit for establishing a control correspondence table according to the first image and the second image; and a control unit for receiving a specified instruction, wherein the specified instruction specifies an interest in the first image And adjusting an imaging field of the second camera according to the region of interest and the control correspondence table, so that the second camera captures the region of interest and outputs a third image. The linked photographic system of claim 12, wherein when the second image is obtained by the second camera, the second camera is used to capture a plurality of fourth images for different shooting fields. The four images are panoramicly connected to generate a panoramic image, and the panoramic image is used as the second image. The photographic system of claim 12, wherein the first image includes a plurality of coordinate points, and the calibration unit is at the first time when the control correspondence table is established according to the first image and the second image M is an M-position corresponding to the position in the second image, and M is a positive integer greater than or equal to 3; from the M positioning points of the first image, three are selected according to all combinations The positioning point is a positioning point group, wherein a total of N positioning point groups are generated, N is a positive integer; and N coordinate conversion parameter groups respectively corresponding to the N positioning point groups are calculated; for each of the first images The coordinate point performs a conversion process according to one of the N coordinate conversion parameter groups to obtain a shooting position parameter corresponding to the second camera of each of the coordinate points; and storing the shooting position parameters in the control corresponding table. According to the linked photography system of claim 14, wherein in the conversion procedure, the control unit selects one of the N positioning point groups according to the coordinate point currently targeted; The coordinate conversion parameter group corresponding to the positioning point group calculates a mapping point of the coordinate point currently in the second image; and obtains the shooting position parameter corresponding to the mapping point. According to the fifteenth linked photography system of the patent application scope, the selected positioning point group is formed by three positioning points which are the closest to the current pointing point of the coordinate point. According to the linked photography system of claim 14, wherein the shooting position parameter includes an inclination angle of the second camera and a panning angle. According to the photographic system of claim 12, wherein when the second camera is photographed according to the region of interest and the control correspondence table, and the second camera captures the region of interest and outputs the third image, The control unit queries the control correspondence table according to a third center point of the region of interest to obtain a shooting position parameter corresponding to the region of interest; and causes the second camera to act according to the querying the shooting position parameter and extract the output. The third image. According to the linked photography system of claim 18, wherein the second image includes a second center point, the area of interest includes a plurality of boundary points, and the control corresponding to the control is performed according to a third center point of the area of interest a table, in order to obtain the shooting position parameter corresponding to the region of interest, the control unit calculates a third center point of the third center point in the second image, and the boundary points are in the second image a plurality of edge mapping points; calculating a zoom angle according to the edge points, the second center point, and the third center point; and querying the control correspondence table according to the third center point to obtain a corresponding interest area The shooting position parameter, wherein the shooting position parameter includes a tilt angle and a pan angle. According to the linked photography system of claim 18, wherein the region of interest is circular, the second image includes a second center point, the region of interest includes a plurality of boundary points, and one of the regions according to the region of interest Third center point Querying the control correspondence table to obtain the shooting position parameter corresponding to the region of interest, the control unit calculates a second center point of the second center point in the first image and the third center point in the first a third mid-point of the second image, with the second center point as a center, and a distance between the second center point and the third center point as a radius, and calculating a first reference circle, where The first reference circle intersects two of the boundary points, and then calculates two edge reflection points formed by the two boundary points intersecting the first reference circle in the second image, according to the two edge points Calculating a zoom angle according to the second center point and the third center point, and querying the control correspondence table according to the third center point to obtain the shooting position parameter corresponding to the interest area, wherein the shooting position parameter includes A dip angle and a pan angle. According to the linked photography system of claim 19, wherein the control unit calculates each of the side angles when calculating the zoom angle according to the edge points, the second center point, and the third center point a plurality of angles formed by the second center point and the third center point, and finding a maximum acute angle between the angles, and multiplying the maximum acute angle by 2 as the zoom angle. According to the linked photography system of claim 20, wherein the control unit calculates each of the edges when calculating a zoom angle according to the two edge points, the second center point, and the third center point a plurality of angles formed by the second center point and the third center point, and the angle of the largest one of the angles is multiplied by 2 as the zoom angle.