JP2005277640A - Device and method for setting image area - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image area setting device and an image area setting method capable of simply setting a motion detection area even if a photography range is changed by the travel of a photography means. <P>SOLUTION: The image area setting device sets the motion detection area in an image obtained by allowing a photographing part 301 to change the photography range for photography. The image area setting device has a detection area production section 108, which sets a first motion detection area in a first image photographed by the photographing part 301, and sets a second motion detection area corresponding to the first motion detection area in a second image based on the relationship between the second image photographed by the photographing part 301 and the first image. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an image region setting apparatus and an image region setting method for setting a motion detection region in an image obtained mainly by use in a monitoring system that detects an intruder or the like by using an image.

  2. Description of the Related Art Conventionally, a monitoring system that enables discovery of an intruder or the like transmits an image obtained by a camera to other places. A monitor in another place looks at the transmitted image and finds an intruder or the like. However, since visual inspection by a supervisor requires human labor, a function for automatically detecting motion in an image has been proposed in order to save the human labor.

  For motion detection, a comparison is made between the current image and the image obtained by the previous shooting, and when there is a difference, it is determined that there is a motion, and an image with no abnormality in advance (hereinafter referred to as “reference image”). Is recorded), the current image and the reference image are compared, and if there is a difference, it is determined that there is a movement.

  By the way, in motion detection, there is a demand to detect only motion in a partial area, not necessarily motion of the entire image. For example, in the case of monitoring an intruder from a gate, only a person entering from the gate needs to be detected, and it is desirable that other passersby are not detected even if there is movement. For this reason, a motion detection function has been proposed in which an operator can set a region (motion detection region) in which motion should be detected in an image (see, for example, Patent Document 1).

  An example of the motion detection area in the real space is shown in FIG. 16, and an example of the motion detection area in the image obtained by photographing is shown in FIG. When the image 404 shown in FIG. 17 is obtained by photographing with the camera 402 shown in FIG. 16, when the operator sets the rectangular area 405 in the image 404 as the motion detection area, the rectangular area 405 is not displayed in the real space. Is associated with a cone-shaped area 403 having a vertex as a vertex.

  In addition, in conventional surveillance systems, it is common to use a camera with a fixed shooting range, so there are many blind spots. To eliminate this blind spot, it is necessary to use many cameras and the cost increases. There was a problem to do. For this reason, an image monitoring function using a moving body using a recent wireless communication technique and a moving body technique represented by a robot has been proposed (for example, see Patent Document 2). By mounting the camera on the moving body, the shooting range is widened, and the blind spot can be eliminated.

However, it is difficult to use the above-described motion detection method in such a monitoring system in which the shooting range of the camera changes. Specifically, in the method of comparing the current image with the image obtained by the previous shooting, the change in the image is also caused by the movement of the camera, and therefore the motion cannot be detected by comparing the two. On the other hand, in the method of comparing the current image with the reference image, a plurality of reference images are set according to the change in the shooting range (see, for example, Patent Document 3).
JP 2000-3480 A JP-A-8-7186 JP 2002-158999 A

  However, in the conventional motion detection method, when the shooting range changes, it is necessary to set a motion detection region for each reference image having a different shooting range, and there is a problem that the burden on the operator is heavy.

  The present invention has been made to solve the conventional problems, and an image area setting device and an image area setting capable of easily setting a motion detection area even when an imaging range changes due to movement of an imaging unit. It aims to provide a method.

  An image area setting device according to the present invention sets a motion detection area in an image obtained by photographing while the photographing means changes the photographing range, and is obtained by photographing by the photographing means. A first motion detection region setting means for setting a first motion detection region in one image, and a relationship between the second image obtained by photographing by the photographing means and the first image; And a second motion detection region setting means for setting a second motion detection region corresponding to the first motion detection region in the second image.

  With this configuration, any image obtained by photographing by the photographing unit is set as the first image, and the second image is based on the relationship between the first image and the second image other than the first image. A second motion detection area corresponding to the first motion detection area set in the first image is set in the first image. For this reason, even when the shooting range changes, the second motion detection region can be set in the second image without manual intervention.

  In the image region setting device of the present invention, the second motion detection region setting unit determines whether or not a region corresponding to the first motion detection region is included in the second image, If included, the region corresponding to the first motion detection region may be set as the second motion detection region.

  With this configuration, it is possible to appropriately set the second motion detection area as necessary.

  In the image region setting device according to the present invention, the second motion detection region setting means may be configured such that the second motion detection region is based on a correspondence relationship between the first image and the first motion detection region. Can be configured.

  With this configuration, it is possible to appropriately determine the position and range of the first motion detection area from the correspondence between the first image and the first motion detection area, and to appropriately set the second motion detection area. It becomes.

  In the image region setting device according to the present invention, the second motion detection region setting unit may include the position information and posture information of the photographing unit when the first image is obtained, and the second image. The second motion detection region can be set based on the position information and posture information of the photographing unit obtained at the time.

  With this configuration, based on the position information and posture information of the photographing unit when the first image is obtained, and the position information and posture information of the photographing unit when the second image is obtained, It is possible to grasp the shooting ranges of the image and the second image and appropriately set the second motion detection area.

  In the image region setting device of the present invention, the first motion detection region setting unit sets a three-dimensional space region as the first motion detection region, and the second motion detection region setting unit includes: A three-dimensional space area as the second motion detection area can be set.

  With this configuration, it is possible to set the first motion detection region and the second motion detection region as the motion detection region as a three-dimensional space region.

  Moreover, the image area setting device of the present invention is configured such that the first motion detection area setting means sets the three-dimensional space area formed by the planar area and the depth in the first image. it can.

  With this configuration, it is possible to appropriately set the second motion detection region using information regarding the planar region and the depth in the first image.

  In the image region setting device of the present invention, the first motion detection region setting means is formed in a cone shape formed by the planar region and at least one of a front depth and a rear depth in the first image. Further, the three-dimensional space region can be set to any one of a frustum shape and a rectangular parallelepiped shape.

  In the image region setting device of the present invention, the first motion detection region setting means sets the front depth and the rear depth using a reference point whose position in the first image is known. It can be configured.

  In the image region setting device according to the present invention, the first motion detection region setting unit sets the spherical three-dimensional space region having a predetermined position as a center and a distance from the predetermined position as a diameter. can do.

  With this configuration, it is possible to appropriately set the spherical second motion detection region using the predetermined position in the first image and the distance from the predetermined position.

  In the image region setting device of the present invention, the first motion detection region setting unit sets a plurality of the first motion detection regions, and the second motion detection region setting unit includes a plurality of the first motion detection regions. A plurality of the second motion detection areas corresponding to each of the one motion detection areas can be set.

  In the image region setting device of the present invention, the first motion detection region setting unit sets a priority for each of the first motion detection regions, and the second motion detection region setting unit includes the It can be set as the structure which sets a priority to each of a 2nd motion detection area | region.

  The image area setting device of the present invention includes a first motion non-detection area setting means for setting a first motion non-detection area in which motion should not be detected in the first image, and the second image. And a second motion non-detection region in which a motion corresponding to the first motion non-detection region should not be detected is set in the second image based on the relationship between the first image and the first image. The motion detection area setting means can be configured.

  The image area setting method of the present invention is to set a motion detection area in an image obtained by photographing while changing the photographing range by the photographing means, and obtained by photographing by the photographing means. A first motion detection region setting step for setting a first motion detection region in the first image, and a relationship between the second image obtained by photographing by the photographing means and the first image. And a second motion detection region setting step for setting a second motion detection region corresponding to the first motion detection region in the second image.

  In the image region setting method of the present invention, the second motion detection region setting step determines whether or not a region corresponding to the first motion detection region is included in the second image, If included, the region corresponding to the first motion detection region may be set as the second motion detection region.

  In the image region setting method of the present invention, the second motion detection region setting step may be configured such that the second motion detection region is based on a correspondence relationship between the first image and the first motion detection region. Can be configured.

  In the image region setting method according to the present invention, the second motion detection region setting step may include the positional information and posture information of the photographing unit when the first image is obtained, and the second image. The second motion detection region can be set based on the position information and posture information of the photographing unit obtained at the time.

  In the image region setting method of the present invention, the first motion detection region setting step sets a three-dimensional space region as the first motion detection region, and the second motion detection region setting step includes: A three-dimensional space area as the second motion detection area can be set.

  The image region setting method of the present invention may be configured such that the first motion detection region setting step sets the three-dimensional space region formed by a planar region and a depth in the first image. it can.

  In the image region setting method of the present invention, the first motion detection region setting step sets the spherical three-dimensional space region having a predetermined position as a center and a distance from the predetermined position as a diameter. can do.

  In the image region setting method of the present invention, the first motion detection region setting step sets a plurality of the first motion detection regions, and the second motion detection region setting step includes a plurality of the first motion detection regions. A plurality of the second motion detection areas corresponding to each of the one motion detection areas can be set.

  In the image region setting method of the present invention, the first motion detection region setting step sets a priority for each of the first motion detection regions, and the second motion detection region setting step includes It can be set as the structure which sets a priority to each of a 2nd motion detection area | region.

  The image region setting method of the present invention includes a first motion non-detection region setting step for setting a first motion non-detection region in which motion should not be detected in the first image, and the second image. And a second motion non-detection region in which a motion corresponding to the first motion non-detection region should not be detected is set in the second image based on the relationship between the first image and the first image. And a motion detection region setting step.

  According to the present invention, even when the shooting range changes, the second motion detection area as the motion detection area can be set in the second image without manual intervention.

  Hereinafter, a monitoring system according to an embodiment of the present invention will be described with reference to the drawings.

  An example of the operation of the mobile robot that captures an image used for monitoring by the monitoring system according to the embodiment of the present invention is shown in FIG. The mobile robot 300 shown in FIG. 1 moves to positions 302a, 302b, 302c, and 302d, and performs imaging using an imaging unit (not shown) built in these positions. An example of an image obtained by shooting by the mobile robot 300 is shown in FIG. In FIG. 2, an image 308a is obtained by photographing the mobile robot 300 at the position 302a, an image 308b is obtained by photographing the mobile robot 300 at the position 302b, and an image 308c is obtained by photographing the mobile robot 300 at the position 302c. The image 308d is an image obtained by photographing the mobile robot 300 at the position 302d.

  A block diagram of the monitoring system is shown in FIG. The monitoring system shown in FIG. 3 includes a sequencer 101 that manages information (sequence information) related to the operation of the mobile robot 300, an operation control unit 102 that controls the operation of the mobile robot 300 according to the sequence information, and an imaging unit in the mobile robot 300. A reference image selection unit 103 that selects a reference image from images obtained by photographing 301, a reference image information management unit 104 that acquires and manages information (reference image information) related to the selected reference image, and a reference image A reference image recording unit 105 that records a reference image, a reference image display unit 106 that displays a reference image, and an operator's operation instruction for setting a motion detection region that is a region in which motion should be detected in the reference image. A reference image area input unit 107, a detection area calculation unit 108 that digitizes the motion detection area, and a digitized motion detection area. A detection area recording unit 109 for performing comparison, an image comparison unit 110 for comparing a reference image and an image (monitoring image) obtained by performing shooting at any time by the imaging unit 301, and detection for extracting an image change in the motion detection region. An area extracting unit 111 and a warning generating unit 112 that generates a warning when an image change is extracted are included.

  Among these, the image area setting device that sets the movement area of the reference image includes a reference image selection unit 103, a reference image information management unit 104, a reference image recording unit 105, a reference image display unit 106, a detection region input unit 107, and a detection. The area calculation unit 108 and the detection area recording unit 109 are configured.

  The operation of the image area setting apparatus configured as described above will be described. A flowchart of the operation of the image area setting apparatus is shown in FIG.

  The reference image selection unit 103 determines whether or not the operator has performed an operation for selecting a reference image from images obtained by the imaging unit 301 in the mobile robot 300 being photographed as needed (S101). ). When a reference image selection operation is performed, the reference image selection unit 103 selects a reference image according to the operation and outputs the selected reference image to the reference image information management unit 104 (S102).

  The reference image information management unit 104 acquires the sequence information of the mobile robot 300 when the selected reference image is obtained from the sequencer 101 and manages it as reference image information.

  An example of the reference image information is shown in FIG. By repeating the selection of the reference image in S102 described above, the reference image information management unit 104 manages reference image information corresponding to each of a plurality of reference images as shown in FIG. In the reference image information shown in FIG. 5A, an image number for specifying each selected reference image is associated with the time when the reference image was captured. Note that the shooting time may be an actual time or a time indicating an elapsed time since the imaging unit 301 in the mobile robot 300 started shooting. On the other hand, the reference image information shown in FIG. 5B associates the image number for specifying each selected reference image with the position and posture of the mobile robot 300 when the reference image is obtained. ing. The shooting range of each reference image is specified by the position and orientation of the mobile robot 300, and the motion detection area set in the reference image can be projected onto another reference image. Note that if the imaging unit 301 can operate independently from the other parts of the mobile robot 300 and the imaging range of the reference image is not specified only by the position and orientation of the mobile robot 300, the imaging is further performed as reference image information. Information regarding the shooting direction of the unit 301 (for example, viewing angle parameters) may be included.

  Further, the reference image information management unit 104 outputs the selected reference image to the reference image recording unit 105. The reference image recording unit 105 records the input reference image and outputs it to the reference image display unit 106. The reference image display unit 106 displays the input reference image.

  Returning to FIG. 4, the description will be continued. After the selection of the reference image in S102, the reference image selection unit 103 determines whether an operation for reselecting the reference image has been performed by the operator (S103). When the reference image reselection operation is performed, the operations after the selection of the reference image (S102) by the reference image selection unit 103 are repeated.

  On the other hand, when the re-selection operation of the reference image is not performed, the detection area input unit 107 sets a first motion detection area that is an area in which motion should be detected in the reference image selected by the operator. It is determined whether or not an operation has been performed. When the operation for setting the motion detection area is not performed, the operation after the determination (S103) on whether or not the operation for reselecting the reference image by the reference image selection unit 103 is performed is repeated.

  On the other hand, when a motion detection region setting operation is performed, the detection region input unit 107 outputs information regarding the operation to the detection region calculation unit 108. The detection area calculation unit 108 sets a first motion detection area in the reference image based on the information related to this operation and the reference image information acquired from the reference image information management unit 104 (S105). Hereinafter, the reference image in which the first motion detection area is set is referred to as a first reference image.

  Specifically, the detection area calculation unit 108 recognizes the shooting range of the first reference image based on the reference image information acquired from the reference image information management unit 104, and further sets the motion detection area by the operator. Based on the information regarding the operation, the first motion detection area data (for example, coordinates) in the real space and the first motion detection area data (for example, coordinates) in the first reference image are calculated to obtain the first Set the motion detection area. The calculated data of the first motion detection area is recorded in the detection area recording unit 109 in association with the information related to the first reference image.

  An example of the first motion detection area in the real space is shown in FIG. 6, and an example of the first motion detection area in the reference image is shown in FIG. When the reference image 321 shown in FIG. 7 is the first reference image, the operator selects the point 323a, the point 323b, the point 323c, and the point 323d in the reference image 321 in order to set the first motion detection area. When the plane area 323 is designated and the front depth and the rear depth are designated in the depth input field displayed together with the reference image 321, the detection area calculation unit 108 moves in the real space as shown in FIG. A frustum-shaped region 310 formed by cutting a cone-shaped region 312 having a vertex at the shooting position of the robot 300 and a bottom surface corresponding to the plane region 323 by a front depth 315 and a rear depth 314. Is set as the first motion detection region. Here, the plane corresponding to the plane area 323 corresponds to the plane at a position away from the mobile robot 300 by the front depth 315.

  Returning to FIG. 4, the description will be continued. After the first motion detection region is set in S105, the detection region calculation unit 108 performs spatial region conversion on a reference image (second reference image) other than the first reference image (S106). It is determined whether there is a second reference image related to the one reference image (S107). Specifically, in S106 and S107, the detection area calculation unit 108 acquires reference image information from the reference image information management unit 104, and based on this reference image information, the shooting range of the first reference image and the second The imaging range of the reference image is recognized. Furthermore, the detection area calculation unit 108 uses the data of the first motion detection area calculated in S105 to determine whether or not there is a second reference image including all or part of the first motion detection area. Determine. When there is a second reference image that includes the entire or part of the first motion detection area, the detection area calculation unit 108 relates the second reference image to the first reference image. The second reference image is determined.

  If there is no second reference image related to the first reference image, the series of operations ends. On the other hand, when there is a second reference image related to the first reference image, the detection area calculation unit 108 includes a motion detection area corresponding to the first motion detection area in the second reference image. (Second motion detection area) is set.

  Examples of motion detection area data in the reference image calculated by the detection area calculation unit 108 are shown in FIGS.

  In FIG. 8, the image number for specifying each reference image, the detection region number for specifying the motion detection region appearing in the reference image, and the two-dimensional coordinates of the motion detection region in the reference image are associated with each other. It has been. For example, the reference image with the image number N + 2 is set as the first reference image, and the reference images with the image numbers N and N + 1 are set as the second reference image. The movements of the detection area numbers 1 and 2 in the first reference image. Consider the case where a detection area is set. In this case, when the motion detection area with the detection area number 1 also appears in the reference images with the image numbers N and N + 1, while the motion detection area with the detection area number 2 does not appear in the reference images with the image numbers N and N + 1, The data of the motion detection area in the reference image calculated by the detection area calculation unit 108 is as shown in FIG.

  On the other hand, in FIG. 9, in FIG. 9A, an image number for specifying each reference image is associated with a detection area number for specifying a motion detection area appearing in the reference image. In FIG. 9B, the detection area number for specifying each motion detection area is associated with the three-dimensional coordinates in the real space of the motion detection area. In this case, the detection area calculation unit 108 further converts the three-dimensional coordinates in the real space of the motion detection area into two-dimensional coordinates in each reference image.

  An example of the motion detection area in each reference image is shown in FIG. 10, the reference image 331a is an image obtained by photographing the mobile robot 300 at the position 302a in FIG. 1, the reference image 331b is an image obtained by photographing the mobile robot 300 at the position 302b in FIG. 1, and the reference image 331c is FIG. The image obtained by photographing the mobile robot 300 at the position 302c and the reference image 331d are images obtained by photographing the mobile robot 300 at the position 302d in FIG. Among these reference images, when the reference image 331b is the first reference image and the reference image 331a, the reference image 331c, and the reference image 331d are the second reference images, they are set in the first reference image 331b. The motion detection area 330b is set as the motion detection area 330a in the second reference image 331a, is set as the motion detection area 330c in the second reference image 331c, and is the motion detection area in the second reference image 331d. 330d is set.

  After the setting of the second motion detection region in the second reference image, the operations after the spatial region conversion (S106) by the detection region calculation unit 108 are repeated.

  After setting the motion detection area in such a reference image, it is possible to monitor whether or not an intruder exists in the motion detection area. That is, the image comparison unit 110 inputs a monitoring image obtained by performing shooting at any time by the imaging unit 301 in the mobile robot 300, and based on the sequence information in the sequencer 101, the same shooting range as the monitoring image. A reference image obtained by shooting is acquired from the reference image recording unit 105. Further, the image comparison unit 110 compares the input monitoring image with the acquired reference image, and outputs the image change to the detection region extraction unit 111 as a comparison result. The detection area extraction unit 111 inputs the comparison result, acquires the data of the motion detection area recorded in the detection area recording unit 109 via the detection area calculation unit 108, and changes the image in the motion detection area. Extract. The warning generation unit 112 generates a warning when a change in the image is extracted by the detection area extraction unit 111. In addition, when the change of the image is extracted by the detection region extraction unit 111, the warning generation unit 112 may generate a trigger for performing some processing other than the generation of the warning.

  As described above, in the present embodiment, the image area setting device is obtained by shooting when the imaging unit 301 in the mobile robot 300 performs shooting while changing the shooting range as the mobile robot 300 moves. A first motion detection area corresponding to the operator's operation is set in the first reference image which is one of the reference images, and the first reference image which is another reference image is set in the first reference image. The position and orientation of the mobile robot 300 when the reference image and the second reference image are taken, and the relevance of the first reference image based on the correspondence between the first reference image and the first motion detection area. Accordingly, the second motion detection area is set. For this reason, even when the shooting range changes, if the first motion detection region is set in the first reference image, the first motion is not detected in the second reference image without manual intervention. A second motion detection area on which the detection area is projected can be set.

  Further, the image area setting device determines whether or not an area corresponding to the first motion detection area is included in the second reference image, and if included, sets the second motion detection area. Therefore, it is possible to set an appropriate motion detection area as required. Further, the image area setting device sets a first motion detection area as a three-dimensional space area formed by the planar area and the depths of the front and rear in the first reference image. And the 2nd motion detection area | region as a three-dimensional space area | region can be set appropriately using the information regarding depth of back and back.

  In the above-described embodiment, the image comparison unit 110 compares the reference image with the monitoring image, and then the detection region extraction unit 111 extracts the change in the image in the motion detection region. The motion detection area and the area corresponding to the motion detection area in the monitoring image may be extracted, and the images of these extracted areas may be compared with each other.

  Further, the sequencer 101 and the operation control unit 102 may be configured in the mobile robot 300. In this case, the image information management unit 104 acquires sequence information from the sequencer 101 by wireless communication or the like.

  Further, in the above-described embodiment, as shown in FIG. 6, the detection area calculation unit 108 has a cone whose top is the position corresponding to the plane area 323 in FIG. The frustum-shaped region 310 formed by cutting the region 312 by the front depth 315 and the rear depth 314 is set as the first motion detection region. As shown in FIG. A flat surface corresponding to the region 323 may be set as a bottom surface at the position of the front depth 315, and the rectangular parallelepiped first motion detection region 340 having a depth up to the position of the rear depth 314 may be set. Further, as shown in FIG. 12, the detection region calculation unit 108 may set a spherical first motion detection region 355 having a diameter 357 with the position of the depth 358 from the mobile robot as the center 356. Furthermore, as shown in FIG. 13, the detection area calculation unit 108 may set a planar first motion detection area 368 at the position of the depth 369.

  In addition, the detection area calculation unit 108 measures the depth of a feature point by a stereoscopic method using a plurality of reference images, displays the feature point as a reference point on the reference image display unit 106 together with a measurement result, The depth corresponding to the selection of the reference point by the person may be calculated. For example, the detection area calculation unit 108 measures the depth from the reference points A to E in FIG. 14, and enables the specification of the reference point in the depth input field displayed together with the reference image 279 as shown in FIG.

  Further, the plane area specified in the operation for setting the motion detection area by the operator is not limited to a quadrangle, and may be another shape. Furthermore, the plane area specified in the operation for setting the motion detection area by the worker may be a set of a plurality of small areas specified by the worker. In addition, when the operator designates two points, a square or a rectangle having a line segment between the two points as a diagonal line may be designated as a plane region. Alternatively, an area surrounded by a contour line derived from a difference in luminance or color in the reference image or an object extracted based on a difference between two reference images having different shooting times may be a designated unit of the plane area. May be.

  Further, the detection area calculation unit 108 may set a motion non-detection area (mask area) as an area where motion should not be detected in the reference image by the same procedure as the setting of the motion detection area. In this case, the detection area extraction unit 111 does not extract image changes in the mask area.

  Further, when a plurality of first motion detection areas are set in the first reference image, and a plurality of second motion detection areas are set in the second reference image accordingly, a warning generation unit In 112, a warning for enabling determination of which motion detection region has undergone the image change or a trigger for performing some processing may be generated.

  In addition, when a plurality of first motion detection areas are set in the first reference image, or when a plurality of second motion detection areas are set in the second reference image, the detection area calculation is performed. The unit 108 may set a priority for each motion detection area. In this case, the priority may be set according to the operator's operation instruction. Alternatively, the priority may be set such that the priority is higher in the motion detection area closer to the mobile robot 300, and the priority is higher in the motion detection area having a smaller depth calculated by the stereoscopic method. May be set. Information about the priority is recorded in the detection area recording unit 109 as motion detection area data. At the time of motion detection, the detection area extraction unit 111 notifies the warning generation unit 112 of the priority of the motion detection area together with the image change in each motion detection area. A warning or a trigger for performing some processing may be generated in accordance with the above.

  Further, when an image change occurs in a portion where a plurality of motion detection areas overlap, the warning generation unit 112 may generate a warning corresponding to each motion detection area or a trigger for performing some processing. .

  As described above, the image area setting device and the image area setting method according to the present invention have the effect that the motion detection area can be easily set even when the shooting range changes as the shooting means moves. It is useful for an image area setting apparatus and an image area setting method for setting a motion detection area in an image obtained by photographing, which is used in a monitoring system that detects an intruder or the like by an image.

The figure which shows an example of operation of a mobile robot The figure which shows an example of the image obtained by imaging | photography of a mobile robot Block diagram of surveillance system with built-in image area setting device Flow chart of operation of image area setting device The figure which shows an example of reference | standard image information The figure which shows the 1st example of the 1st motion detection area | region in real space. The figure which shows an example of the 1st motion detection area | region in the reference | standard image corresponding to FIG. The figure which shows an example of the data of the motion detection area | region in a reference | standard image The figure which shows the other example of the data of the motion detection area | region in a reference | standard image The figure which shows an example of the motion detection area | region in each reference | standard image The figure which shows the 2nd example of the 1st motion detection area | region in real space. The figure which shows the 3rd example of the 1st motion detection area | region in real space. The figure which shows the 4th example of the 1st motion detection area | region in real space. Diagram showing an example of a reference point in real space The figure which shows an example of the reference | standard point in the reference | standard image corresponding to FIG. The figure which shows an example of the motion detection area | region in the real space when the conventional monitoring system is used The figure which shows an example of the motion detection area | region in the image when the conventional monitoring system is used

Explanation of symbols

DESCRIPTION OF SYMBOLS 101 Sequencer 102 Operation control part 103 Reference | standard image selection part 104 Reference | standard image information management part 105 Reference | standard image recording part 106 Reference | standard image display part 107 Detection area input part 108 Detection area calculation part 109 Detection area recording part 110 Image comparison part 111 Detection area extraction Unit 112 warning generation unit 300 mobile robot 301 imaging unit

Claims (22)

An image area setting device for setting a motion detection area in an image obtained by shooting while changing the shooting range of the shooting means,
First motion detection area setting means for setting a first motion detection area in a first image obtained by photographing by the photographing means;
A second motion detection corresponding to the first motion detection region in the second image based on the relevance between the first image and the second image obtained by photographing by the photographing means. An image area setting device comprising: second motion detection area setting means for setting an area. The second motion detection region setting means determines whether or not a region corresponding to the first motion detection region is included in the second image, and if included, the first motion detection The image area setting apparatus according to claim 1, wherein an area corresponding to an area is set as the second motion detection area. 2. The second motion detection region setting means sets the second motion detection region based on a correspondence relationship between the first image and the first motion detection region. Or the image area setting apparatus of 2 description. The second motion detection area setting unit includes position information and posture information of the photographing unit when the first image is obtained, and position information of the photographing unit when the second image is obtained. 4. The image region setting device according to claim 1, wherein the second motion detection region is set based on the posture information and the posture information. 5. The first motion detection region setting means sets a three-dimensional space region as the first motion detection region,
5. The image region setting device according to claim 1, wherein the second motion detection region setting unit sets a three-dimensional space region as the second motion detection region. 6. The image area setting device according to claim 5, wherein the first motion detection area setting means sets the three-dimensional space area formed by a plane area and a depth in the first image. The first motion detection region setting means is any one of a cone shape, a frustum shape, and a rectangular parallelepiped shape formed by the planar region and at least one of a front depth and a rear depth in the first image. The image area setting apparatus according to claim 6, wherein the three-dimensional space area is set. The image according to claim 7, wherein the first motion detection area setting unit grasps the front depth and the rear depth by using a reference point whose position in the first image is known. Area setting device. 7. The image area setting according to claim 6, wherein the first motion detection area setting means sets the spherical three-dimensional space area having a predetermined position as a center and a distance from the predetermined position as a diameter. apparatus. The first motion detection region setting means sets a plurality of the first motion detection regions,
The said 2nd motion detection area setting means sets the said some 2nd motion detection area corresponding to each of the said some 1st motion detection area, The any one of Claim 1 thru | or 9 characterized by the above-mentioned. The image area setting device described. The first motion detection region setting means sets a priority for each of the first motion detection regions,
11. The image region setting device according to claim 10, wherein the second motion detection region setting means sets a priority for each of the second motion detection regions. First motion non-detection region setting means for setting a first motion non-detection region in which motion should not be detected in the first image;
A second motion non-detection region where motion corresponding to the first motion non-detection region should not be detected in the second image based on the relevance between the second image and the first image The image area setting device according to claim 1, further comprising second motion detection area setting means for setting An image area setting method for setting a motion detection area in an image obtained by shooting while changing the shooting range by the shooting means,
A first motion detection region setting step for setting a first motion detection region in a first image obtained by photographing by the photographing means;
A second motion detection corresponding to the first motion detection region in the second image based on the relevance between the first image and the second image obtained by photographing by the photographing means. And a second motion detection area setting step for setting the area. The second motion detection region setting step determines whether or not a region corresponding to the first motion detection region is included in the second image, and if included, the first motion detection 14. The image area setting method according to claim 13, wherein an area corresponding to an area is set as the second motion detection area. 14. The second motion detection region setting step sets the second motion detection region based on a correspondence relationship between the first image and the first motion detection region. Or the image area setting apparatus of 14. The second motion detection area setting step includes position information and posture information of the photographing unit when the first image is obtained, and position information of the photographing unit when the second image is obtained. 16. The image region setting method according to claim 13, wherein the second motion detection region is set based on the posture information and the posture information. The first motion detection region setting step sets a three-dimensional space region as the first motion detection region,
The image region setting method according to claim 13, wherein the second motion detection region setting step sets a three-dimensional space region as the second motion detection region. The image region setting method according to claim 17, wherein the first motion detection region setting step sets the three-dimensional space region formed by a planar region and a depth in the first image. 19. The image region setting according to claim 18, wherein the first motion detection region setting step sets the spherical three-dimensional space region having a predetermined position as a center and a distance from the predetermined position as a diameter. Method. The first motion detection region setting step sets a plurality of the first motion detection regions,
20. The second motion detection area setting step sets a plurality of the second motion detection areas corresponding to each of the plurality of first motion detection areas. The image area setting method described. The first motion detection region setting step sets a priority for each of the first motion detection regions,
21. The image region setting method according to claim 20, wherein the second motion detection region setting step sets a priority for each of the second motion detection regions. A first motion non-detection region setting step for setting a first motion non-detection region in which motion should not be detected in the first image;
A second motion non-detection region where motion corresponding to the first motion non-detection region should not be detected in the second image based on the relevance between the second image and the first image 23. The image region setting method according to claim 13, further comprising: a second motion detection region setting step for setting.

Images