JP2013012848A - Camera device and control method therefor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To enable settings until a specific object is detected and a photographing direction is changed to its direction to be configured in a short time.SOLUTION: When a target object is detected while origin positions in pan and tilt directions are detected as initialization processing, information on the detection position is stored in a RAM 113. Then, an origin in a rotation direction is detected to complete initialization, and thereafter a photographing direction is changed to a direction in which the target object is photographed, thereby enabling the photographing direction to be set in a short time, so that a time for installing and setting a camera can be shortened.

Description

  The present invention relates to a camera device, a camera device control method, and a program, and more particularly to a technique suitable for use in detecting a specific image pattern from a captured image.

  Some network cameras include a pan, tilt, and rotation drive mechanism that can freely set the shooting direction of the camera based on user instructions. This user instruction is performed via a network by an operation from a client such as a personal computer (PC), and the camera shooting direction and angle of view can be set remotely.

  When changing the shooting direction and angle of view by remote operation from a remote location, the user performs pan, tilt, and rotation movement operations while viewing the shot image displayed on the monitor using a dedicated viewer. For this reason, the movement direction may not be intuitively understood. For example, when the camera is installed on a wall, the image may be rotated by pan operation depending on the tilt angle, or the image may be moved vertically or horizontally by tilt operation depending on the pan angle. I'll be relaxed.

  In order to eliminate such inconvenience, for example, Patent Document 1 stores a plurality of composition patterns for setting the shooting direction, analyzes an image shot by the camera, and indicates a composition pattern instructed by the user. A technique is disclosed in which direction control is automatically performed at the position. Furthermore, Patent Document 2 discloses a technique for recognizing a camera posture shift and automatically correcting the position as an application example of image analysis.

JP 2000-244800 A JP 2008-259056 A

  In the prior arts disclosed in Patent Documents 1 and 2, fine adjustment is performed to obtain a desired composition when the target subject is within the shooting angle of view in advance. However, when the subject is not within the angle of view, it is necessary to search for the subject by panning and tilting operations, so it is necessary to scan the entire space of the shootable area, and it takes a lot of time to search for the subject. .

  Immediately after turning on the power of the camera, it is necessary to perform the operation of detecting the origin position by the initializing operation of pan, tilt, and rotation.

  In view of the above-described problems, an object of the present invention is to enable setting in a short time until a specific subject is detected and the shooting direction is changed to that direction.

  The camera device of the present invention includes an imaging unit, an initialization unit that initializes a setting of a driving mechanism of the imaging unit while capturing an image by the imaging unit, and a change that changes a shooting direction by the driving mechanism of the imaging unit. Means for analyzing the photographed image to detect a predetermined image pattern, and storage means for storing information relating to the predetermined image pattern detected by the image analysis means. The means analyzes the captured image while the initialization by the initialization means is being performed, and the predetermined image is analyzed by the image analysis means while the initialization is being performed by the initialization means. When the pattern is detected, the changing unit displays the predetermined image pattern based on the information stored in the storage unit after the initialization by the initialization unit is completed. And changes a shooting direction to direction including.

  According to the present invention, the setting of the shooting direction can be performed in a short time, and the time for camera installation and setting work can be shortened.

It is a block diagram which shows the example of whole structure of the network camera in embodiment. It is a figure which shows an example of the connection form of the network camera system in embodiment. It is a perspective view which shows the example of an external appearance structure of the network camera in embodiment. It is a flowchart which shows an example of the process sequence which controls the imaging | photography direction by the network camera which concerns on 1st Embodiment. FIG. 10 is a diagram illustrating an example of a screen displayed on the video viewer device side when the detection of the origin in the pan direction and the tilt direction is completed in the second embodiment. It is a figure which shows an example of the mark used as the object to detect. It is a conceptual diagram which shows the example of an external appearance structure seen from the side of the network camera in embodiment.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings.
(First embodiment)
FIG. 2 is a diagram illustrating an example of a connection form of the network camera system in the present embodiment.
In FIG. 2, each of the network cameras 201 to 204 is connected to the Internet network 206 through a LAN line 205. Video data captured by the network cameras 201 to 204 is distributed from the Internet network 206 to a client described later via the LAN line 205.

  Video viewer devices 207 and 208 as clients are connected to the Internet network 206 through another LAN line 210. The video viewer devices 207 and 208 are, for example, an application that operates on a personal computer (PC), a home digital TV, and a mobile terminal. In addition, the video viewer devices 207 and 208 have a function of reproducing and displaying video data distributed from the network cameras 201 to 204 in real time. Furthermore, it has a UI (user interface) for performing camera control such as setting of shooting conditions and pan / tilt / zoom operations.

  Further, a recording server device 209 is connected to the LAN line 210, and video data distributed from the network cameras 201 to 204 is constantly recorded on a built-in hard disk. The recording server device 209 may be configured by a dedicated recording application that operates on the server PC, or may be dedicated hardware. In addition, a network camera system that can distribute video data recorded by the recording server device 209 to a network and reproduce and display the video data on the video viewer devices 207 and 208 described above may be used.

  In this network camera system, when the shooting direction of the camera is controlled, an operation command is transmitted to the Internet network 206 through the LAN line 210 by an operation on the video viewer devices 207 and 208 as clients. Then, an operation command is sent to the network cameras 201 to 204 via the LAN line 205, and pan, tilt, and rotation operations are executed according to the operation command.

  FIG. 1 is a block diagram illustrating an example of the overall configuration of a network camera 100 according to the present embodiment. Assume that all of the network cameras 201 to 204 shown in FIG. 2 have the configuration shown in FIG. Hereinafter, each configuration illustrated in FIG. 1 will be described together with processing until video data captured by the network camera 201 is distributed to the Internet network 206.

  The optical image captured from the lens unit 101 is converted into RGB digital data by the CCD unit 102 and sent to the signal processing unit 103. The signal processing unit 103 converts the RGB digital data into YCbCr4: 2: 0 format digital data (video data) in the YCbCr4: 2: 2 format. Further, the signal processing unit 103 performs a conversion process of the requested transmission video to an image size, various filter processes, and the like. Then, the processed video data is sent to the image analysis unit 104 and the encoding unit 105.

  The encoding unit 105 encodes and compresses to MPEG4 format when the input video data is moving image data, and encodes and compresses to JPEG format when it is still image data. The communication processing unit 106 converts the MPEG4 moving image data encoded by the encoding unit 105 or the JPEG still image data to the Internet network via the LAN line 205 according to a network protocol such as TCP / IP, HTTP, or RTP. Delivered to 206.

  On the other hand, the image analysis unit 104 analyzes the input video data, and detects whether the target subject and the image pattern of the specified condition are included in the image. Note that the signal processing unit 103, the image analysis unit 104, the encoding unit 105, and the communication processing unit 106 are connected to a CPU 111 described later, and the above-described processing is executed under the control of the CPU 111.

  The camera control unit 107 is connected to the motor driving unit 108 and the lens driving unit 110. The camera control unit 107 outputs control signals for panning, tilting, and rotation operations and control signals for zooming and AF (autofocus) operations in accordance with instructions from the CPU 111 described later.

  The motor drive unit 108 includes a motor drive circuit and the like, and drives a pan / tilt / rotation motor 109 as a drive mechanism in accordance with a control signal from the camera control unit 107, and can change a shooting direction by rotating the motor. . The lens driving unit 110 includes a motor and a motor driving circuit for performing various controls such as zoom and AF, and is controlled according to a control signal from the camera control unit 107.

  A CPU (central processing unit) 111 controls the overall operation of the network camera 100 by executing control program codes stored in a ROM 112 described later. The CPU 111 is connected to the signal processing unit 103, the image analysis unit 104, the encoding unit 105, and the communication processing unit 106, and starts / stops operations, sets operation conditions, and sets operation results for each component. Execute acquisition and control each configuration. Further, a ROM (read only memory) 112 and a RAM (random access memory) 113 are connected to the CPU 111. The operation of the CPU 111 will be described later with reference to a flowchart.

  The ROM 112 stores program codes and data for the CPU 111 to control application processing and the like. The RAM 113 is a memory for writing and reading data when the CPU 111 executes a program stored in the ROM 112. The RAM 113 is provided with a work area, a temporary save area, and the like that are used by the CPU 111 to execute a program for device control.

FIG. 3 is a perspective view showing an external configuration example of the network cameras 201 to 204.
In FIG. 3, a lens unit 301 that is an imaging unit on the camera base unit 303 includes a pan, tilt, and rotation mechanism.

  Arrows 304 to 306 indicate respective operation directions. An arrow 304 indicates a panning operation direction, and the lens unit 301 itself rotates as the upper part of the camera base unit 303 rotates. An arrow 305 indicates a tilting operation direction, and the lens unit 301 can be moved from the horizontal direction to the vertical direction by a rotation mechanism 307 provided at portions supported at both ends of the lens unit 301. An arrow 306 indicates the direction of rotation, and the image can be rotated by a rotation mechanism inside the lens. The dome portion 302 is a transparent dome cover for protecting the lens portion 301.

FIG. 4 is a flowchart illustrating an example of a processing procedure for controlling the shooting direction in the network camera 100. 4 is performed under the control of the CPU 111. In the following description, an example in which video data is distributed from the network camera 201 to the video viewer device 207 will be described.
In FIG. 4, when the network camera 201 is powered on, the processing is started. In step S401, image distribution processing is started. As described with reference to FIG. 1, the image distribution process distributes video data captured via the lens unit 101 and the CCD unit 102 from the communication processing unit 106 via the signal processing unit 103 and the encoding unit 105. It is processing.

  In step S <b> 402, the process waits until an initialization instruction is received from the communication processing unit 106. Here, the initialization instruction is an instruction transmitted from the video viewer device 207 when an initialization button for automatically setting the shooting direction of the network camera 201 in the video viewer device 207 is operated.

  When the initialization instruction is received, processing for enabling the image analysis function is executed in S403. Specifically, an object to be detected by image analysis is set for the image analysis unit 104, and processing for starting analysis processing is executed. Here, image data of an object that can be a detection object as a predetermined image pattern, and feature information such as a keyword, specific brightness, and color are stored in advance, and conditions in these feature information can be set, Register image patterns in advance. As a result, processing such as image analysis, image recognition, and image comparison is performed on the captured image, and it is detected whether the condition is met or matched. At this time, the zoom is set to the wide angle position as an initial setting.

  In step S <b> 404, the CPU 111 functions as an initialization unit, and executes an origin position detection process as a pan position initialization operation. In this process, the pan driving unit in the motor driving unit 108 is scanned to detect the origin sensor, and the detected position is registered as the origin position. At this time, the pan / tilt / rotation motor 109 is driven by the motor driving unit 108 to move the pan direction. Note that the amount of movement in the pan direction in response to a movement instruction from the client by the motor drive unit 108 is calculated based on this origin position.

  Next, in S405, it is determined whether or not the origin sensor position (origin) has been detected. If the origin of the pan direction is detected as a result of this determination, the process proceeds to S408, and if the origin of the pan direction is not detected, the process proceeds to S406.

  In S406, it is determined whether or not the object has been detected by the image analysis started in S403. If the object is detected as a result of this determination, in S407, information on the detected position is stored in the RAM 113, and the process returns to the origin detection operation in S404. Moreover, also when an object is not detected as a result of determination of S407, it returns to S404. As described above, the processing from S404 to S407 is repeated until the origin position in the pan direction is detected.

  If the origin of the pan direction is detected as a result of the determination in S405, the origin position detection process in the pan direction is terminated. Next, in S408, the origin position detection process is executed as the tilt position initialization operation. In this process, as in the case of panning, the tilt driving unit in the motor driving unit 108 is scanned to detect the origin sensor, and the detected position is registered as the origin position. At this time, the pan / tilt / rotation motor 109 is driven by the motor drive unit 108 to move the tilt direction.

  Next, in S409, it is determined whether or not the origin sensor position (origin) has been detected. If the origin in the tilt direction is detected as a result of this determination, the process proceeds to S412. If the origin in the tilt direction is not detected, the process proceeds to S410.

  In S410, it is determined whether or not the object is detected by the image analysis started in S403. If the object is detected as a result of this determination, in S411, information on the detected position is stored in the RAM 113, and the process returns to the origin detection operation in S408. Moreover, also when a target is not detected as a result of determination of S410, it returns to S408. As described above, the processing from S408 to S411 is repeated until the origin position in the tilt direction is detected as in the case of pan.

  If the origin in the tilt direction is detected as a result of the determination in S409, next, the origin position detection process is executed as the rotation position initialization operation in S412. In this process, as in the case of pan and tilt, the rotation drive unit in the motor drive unit 108 is scanned to detect the origin sensor, and the detected position is registered as the origin position.

  Next, in S413, it is determined whether or not the origin sensor position (origin) has been detected. If the origin in the rotation direction is detected as a result of this determination, the process proceeds to S414. If the origin in the rotation direction is not detected, the process returns to S412 and the origin detection operation is repeated.

  Next, in S414, an object is found during the origin detection operation, and information on the detection position of the object is stored in the RAM 113 to determine whether or not the detection operation is complete. As a result of the determination, if the detection operation is completed, the process proceeds to S419. In step S419, the camera control unit 107 functions as a changing unit, and the pan / tilt / rotation motor 109 is driven by the motor driving unit 108 to control the pan / tilt / rotation direction. And the process which moves the direction of a camera to the position of the registered target object is performed. Thereafter, the process returns to S401.

  On the other hand, if the detection of the object is not completed as a result of the determination in S414, it is determined in S415 whether or not the scan in the entire imageable region has been completed. In the origin detection operation in the pan direction in S404 and the origin detection operation in the tilt direction in S408, the trajectory moving in the pan direction and the trajectory moving in the tilt direction are sequentially scanned. However, depending on the state of the lens unit 301 before the detection operation, the origin may be detected before moving the entire trajectory, which may cause a scan residue. In such a case, it is necessary to confirm whether or not an object exists even in an area where scanning is not performed.

  If the result of determination in S415 is that the entire area has been scanned, error processing is executed in S418. An error occurs when the object cannot be detected even after scanning the entire area. The desired object does not exist in the imageable area, or the detection failed due to some trouble. Such a case can be considered. In step S418, the communication processing unit 106 transmits information indicating an error to the video viewer device 207. On the video viewer device 207 side, error processing such as displaying an error message for the user or stopping the video distribution processing is executed.

  As a result of the determination in S415, if the entire area has not been scanned, the process proceeds to S416. In step S416, the pan / tilt / rotation motor 109 is driven by the motor driving unit 108 to control the pan direction and the tilt direction, and an object detection process is performed while scanning the remaining area. In step S417, it is determined whether an object has been detected. If the object is detected as a result of this determination, the process proceeds to S419. If the object is not detected, the process returns to S415 and the scan is repeated.

  Here, in the control of the pan and tilt directions, an internal process when detecting an object, when detecting an origin, and when moving the camera in the direction of the object will be described in detail.

  When an initialization instruction is received from the video viewer device 207, the current pan position of the lens unit 101 is first stored as a temporary origin. Here, when driving in the pan direction is started in order to detect the origin in S404, a value corresponding to the movement amount is updated as a current value with reference to the temporary origin according to the movement in the pan direction. If an object is detected at this time, the detection position is memorize | stored as a detection position of a temporary object by S407. Thereafter, when the origin is detected, the position is stored as the origin, and the difference from the temporary origin stored at the initial position is also corrected at the detection position of the object to obtain the detection position of the object. In S419, movement control is performed to the corrected position of the object.

  With respect to the tilting operation, it is possible to move the object to the position by the same processing as the panning operation described above.

  In the image analysis of S403, a specific mark may be recognized as a detection target. FIG. 6 is a diagram illustrating an example of a mark 601 to be detected. In the mark 601 shown in FIG. 6, the arrow is directed upward so that the up / down determination can be performed, and the rotation can be controlled in accordance with the direction of the arrow of the mark 601. When this mark 601 is set as a detection target, the correct photographing direction can be automatically adjusted in the vertical direction. As described above, it is possible to control an accurate shooting direction by installing an object (for example, a stand) on which the mark 601 is drawn within a shooting range of the network camera 201 intentionally. It becomes.

  Further, when the origin position detection process is executed as the pan position initialization operation in S404, it is preferable that the tilt angle be set in the horizontal direction. FIG. 7 is a conceptual diagram showing an external configuration example viewed from the side of the network camera of FIG.

  FIG. 7 shows a lens unit 701 in a horizontal state in which the lens unit 301 is oriented in the horizontal direction and a lens unit 702 in a vertical state in which the lens unit 301 is oriented in the vertical direction. The shooting angle of view 703 is the angle of view of the lens unit 701 in the horizontal state, and the shooting angle of view 704 represents the angle of view of the lens unit 702 in the vertical state. As shown in FIG. 7, when the pan operation is performed in the horizontal state, it is possible to capture the entire circumference at the horizontal position, and the imaging area is widened. On the other hand, if the pan operation is performed in the vertical state, the captured image is only rotated because the lens is rotated in that position, and the captured areas are almost at the same position. By performing the pan operation in the horizontal state in this way, the coverage of the entire imaging region is increased, and the possibility that the object is detected is increased.

  As described above, when an instruction for initialization processing is received in step S402, origin detection processing in the pan, tilt, and rotation directions is performed while performing image analysis. At this time, since the lens unit 301 moves while shooting, the shutter speed becomes slow, and the shot image may be blurred. When an object detection process is performed using an image in which blurring has occurred, the accuracy of image analysis may be reduced, and the number of cases where the object cannot be detected may increase.

  Therefore, while the CPU 111 functions as a photographing control unit and simultaneously performs the origin detection and the target object detection, control is performed so that the shutter speed is higher than normal photographing and image blurring is reduced. Preferably it is done. When the shutter speed is increased, it is necessary to increase the gain when shooting in a dark place, and as a result, the noise component increases. For this reason, although it is not suitable as an ornamental image, an object can be detected without greatly degrading accuracy by performing image analysis without using a high-frequency component.

  As described above, according to the present embodiment, when the object is detected while the origin position in the pan direction and the tilt direction is detected as the initialization process, the information on the detected position is stored in the RAM 113 and the initialization is performed. After ending, the shooting direction is changed to the direction in which the object is reflected. Thereby, the shooting direction can be set in a short time, and the time for camera installation and setting work can be shortened.

(Second Embodiment)
In the first embodiment, an example has been described in which an object is detected from a captured image and the camera is moved to the position of the object after the origin detection operation is completed. On the other hand, since object recognition or image comparison technology based on image analysis is used for detecting an object, the object desired by the user may not be accurately detected depending on the detection accuracy. In this case, if the shooting direction is automatically moved to the detection position in S419 in FIG. 4, it is necessary to reset the shooting direction in the case of erroneous detection, and it takes extra time for resetting. End up. Therefore, in this embodiment, when an object is detected, a captured image at the detection position is displayed on the video viewer device side, and the camera direction is selected on the user side, so that the network camera is directed to an undesired shooting direction. An example of preventing this will be described. Note that the basic configuration and processing procedure of the present embodiment are the same as those of the first embodiment, and only differences from the first embodiment will be described.

FIG. 5 is a diagram illustrating an example of a screen 501 displayed on the video viewer device 207 side when the detection of the origin in the pan direction and the tilt direction is completed.
On the screen 501 shown in FIG. 5, images 502 to 504 photographed at positions that are candidates for the detected object are displayed, and the image 502 is in a selected state. Also, a left cursor button 505 and a right cursor button 507 are displayed on the lower side of the screen 501, and the selection state of the image moves to the left and right by selecting the left cursor button 505 or the right cursor button 507. When the determination button 506 is selected in a state where an image including the target subject (object) is selected, the shooting direction is determined.

  Note that an image captured at a position that is a candidate for an object may be selected on the video viewer device 207 side from the video data distributed from the network camera 201, and the detection stored in S407 or S411. You may display the image in a position.

  If the network camera 201 detects the origin in the tilt direction as a result of the determination in S409, the network camera 201 transmits information to that effect to the video viewer device 207. Then, the video viewer device 207 displays the screen shown in FIG. 5 in response to receiving the information to that effect. When an image is selected by the video viewer device 207, the information is transmitted to the network camera 201, and the network camera 201 stores the pan, tilt, and rotation position information corresponding to the image in the RAM 113. In step S419, the position of the lens unit 301 is moved based on the position information of the selected image stored in the RAM 113.

  As described above, according to the present embodiment, since the user can select the shooting direction, the shooting direction can be easily set in that direction even when the target object desired by the user is not accurately detected. Can be moved.

(Other embodiments)
The present invention can also be realized by executing the following processing. That is, software (program) that realizes the functions of the above-described embodiments is supplied to a system or apparatus via a network or various storage media, and a computer (or CPU, MPU, or the like) of the system or apparatus reads the program. It is a process to be executed.

101 Lens 104 Image Analysis Unit 107 Camera Control Unit 108 Motor Drive Unit 109 Pan / Tilt / Rotation Motor 111 CPU
113 RAM

Claims (9)

Imaging means;
Initialization means for initializing settings of a drive mechanism of the imaging means while taking an image by the imaging means;
Changing means for changing the shooting direction by the drive mechanism of the imaging means;
Image analysis means for analyzing the captured image and detecting a predetermined image pattern;
Storage means for storing information relating to the predetermined image pattern detected by the image analysis means,
The image analysis means analyzes the captured image while the initialization by the initialization means is being performed,
When the predetermined image pattern is detected by the image analysis means while the initialization by the initialization means is being performed, the changing means, after the initialization by the initialization means is completed, A camera apparatus, wherein a photographing direction is changed to a direction including the predetermined image pattern in an image based on information stored in a storage unit.   The camera device according to claim 1, wherein the changing unit changes a pan direction and a tilt direction as the photographing direction.   The camera apparatus according to claim 1, wherein the image analysis unit detects whether or not the image analysis unit matches a pre-registered image pattern or feature information.   The camera device according to claim 1, wherein the initialization unit detects a position of an origin in a shooting direction by operating a driving mechanism of the imaging unit.   The camera device according to claim 4, wherein the initialization unit detects the origin in the pan direction in a state where the tilt direction is a horizontal direction when detecting the origin in the photographing direction.   A photographing control means is further provided for controlling the imaging means so that the image is taken at a higher shutter speed while the image analysis is being performed by the image analysis means than when the image is not being analyzed. The camera device according to claim 1, wherein the camera device is a camera device.   6. An imaging control means for controlling the imaging means so as to perform imaging while setting the zoom to a wide angle while the image analysis is being performed by the image analysis means. The camera device according to any one of the above. An initialization step of initializing the setting of the drive mechanism of the imaging means while taking an image by the imaging means;
A changing step of changing the shooting direction by the drive mechanism of the imaging means;
An image analysis step of analyzing the captured image and detecting a predetermined image pattern,
In the image analysis step, while the initialization in the initialization step is being performed, the captured image is analyzed,
When the predetermined image pattern is detected in the image analysis step while the initialization in the initialization step is performed, in the change step, after the initialization in the initialization step is completed, A method for controlling a camera device, comprising: changing a shooting direction to a direction including an image of the predetermined image pattern based on information on the predetermined image pattern detected in the image analysis step. An initialization step of initializing the setting of the drive mechanism of the imaging means while taking an image by the imaging means;
A changing step of changing the shooting direction by the drive mechanism of the imaging means;
Causing the computer to execute an image analysis step of analyzing the captured image and detecting a predetermined image pattern;
In the image analysis step, while the initialization in the initialization step is being performed, the captured image is analyzed,
When the predetermined image pattern is detected in the image analysis step while the initialization in the initialization step is performed, in the change step, after the initialization in the initialization step is completed, A program for changing a photographing direction to a direction including the predetermined image pattern in an image based on information on the predetermined image pattern detected in the image analysis step.

Images