JP5667500B2 - Intrusion detection device - Google Patents

Description

  The present invention relates to an intrusion detection apparatus that measures a monitoring area to detect an intruder and detects the presence of a shield that obstructs the measurement.

  In an image sensor or the like for detecting an intruder, in order to continuously detect the intruder, the presence of a shielding object or the like that blocks part or all of the field of view is detected abnormally.

  In the image sensor described in Patent Document 1, the degree of similarity between the current image and the reference image is determined for each determination block obtained by dividing the image of the monitoring area, and the image sensor is determined if the number of determination blocks having a low degree of similarity is equal to or greater than a predetermined number. It was determined that there was an anomaly such as an obstacle that blocked the field of vision.

JP 2000-194866 A

  However, in the prior art, the monitoring area for detecting the presence of an intruder is the same as the monitoring area for detecting the presence of a shield. For this reason, even if a regular user such as a resident places an article such as belongings in the monitoring area, the presence of the article can be determined to be abnormal, which may reduce the convenience for the user.

  This also means that if an area where goods can be placed is provided with priority on user convenience, the monitoring area will be narrowed, and if security is given priority and the monitoring area is widened, user convenience will be reduced. This caused a trade-off.

  Furthermore, it takes skill and time to reliably set a monitoring area that satisfies the trade-off between user convenience and security.

  The present invention has been made in view of the above problems, and an object of the present invention is to realize an intrusion detection device that can improve user convenience while maintaining security.

  Another object of the present invention is to realize an intrusion detection apparatus capable of easily and reliably setting a monitoring area that satisfies the trade-off between user convenience and security.

To solve such problems, the present invention is important equivalent to an intrusion detection device for detecting an intruder to intrusion detection area including the periphery, the output image by imaging an intrusion detection area at predetermined time intervals Compares images at different points in time with an imaging unit that performs intrusion detection area, an important area where important objects are present, and a storage unit that stores an occlusion detection area for detecting the presence of shielding objects that hinder imaging A change area extraction unit for extracting an area, a movement distance calculation unit for calculating a movement distance from the extraction position of the change area at each time point, and a stationary determination in which the movement distance of the change area extracted in the intrusion detection area is set in advance. An intrusion determination unit that determines the presence of an intruder based on being greater than or equal to the distance, and the movement distance of the change area extracted in the shielding detection area is less than the stationary determination distance A shielding judgment unit judges standing, image input can be set and input unit key areas on the display to display a, based on the significant area that has been inputted from the setting input unit, entering at the periphery of the important areas user to provide an intrusion detection device is characterized in that example Bei a monitoring area setting unit, a calculating set the shielding detection area calculation possible range the moving distance when moving the stationary determination distance.

In such an intrusion detection device, it is preferable that the monitoring area setting unit sets a shielding detection area in a range in which an intruder can move around the important area in a predetermined time. According to this, it is possible to set a suitable occlusion detection area in an intrusion detection apparatus configured using an imaging unit in which the distance that an intruder can move during one time recorded at the imaging interval is larger than the stationary determination distance. .

Moreover, it is preferable that the monitoring area setting unit sets a shielding detection area in a range in which the intruder moves the stillness determination distance around the important area. According to this, it is possible to set a suitable occlusion detection area in an intrusion detection device configured using an imaging unit in which the distance that an intruder can move during one time recorded at the imaging interval is equal to or less than the stationary determination distance. .

  According to the present invention, since the intrusion detection area can be set to be smaller than the intrusion detection area without dividing the intrusion detection area from the intrusion detection area, the convenience of the user is improved while maintaining security. It is possible to realize an intrusion detection device that can

  In addition, according to the present invention, since an occlusion detection area is automatically set by setting and inputting an important area that is easily visible, an intrusion detection apparatus that can easily and reliably set a monitoring area can be realized.

1 is a block diagram illustrating an overall configuration of an outer periphery monitoring system 1. FIG. 2 is a block diagram illustrating a configuration of an image sensor 2. FIG. It is an example of a monitoring image obtained by imaging a building and its outer periphery. It is the figure which illustrated the building area set with respect to the monitoring image of FIG. It is the figure which illustrated the intrusion detection area set with respect to the monitoring image of FIG. It is the figure which illustrated the shielding detection area set with respect to the monitoring image of FIG. It is a schematic diagram explaining width W, height H, and stillness determination distance Td. It is the figure which illustrated another occlusion detection area set with respect to the monitoring image of FIG. It is a flowchart explaining a prior setting process. It is a flowchart explaining an image monitoring process. It is a flowchart explaining an abnormality determination process.

  Hereinafter, as an example of a preferred embodiment of the intrusion detection device of the present invention, an outer periphery monitoring system 1 that monitors the outer periphery of a building and notifies a security center when an intruder and a shield are detected will be described.

[Configuration of Perimeter Monitoring System 1]
FIG. 1 is an overall view of the outer periphery monitoring system 1. The outer periphery monitoring system 1 includes an image sensor 2, an authentication device 3, a mobile terminal 4, a controller 5, and a center device 6.

  The image sensor 2 and the authentication device 3 are connected to a controller 5 via a communication line, and the controller 5 is connected to a center device 6 installed at a remote place such as a security center via a wide-area communication network such as a telephone line or an internet line. Is done.

  The image sensor 2 is installed outdoors and monitors the outer periphery of the building, which is an intrusion detection area. When the image sensor 2 detects an intruder in the intrusion detection area, the image sensor 2 outputs an intrusion abnormality signal, and detects a blocking object that hinders the detection of the intruder in the shielding detection area set to overlap with a part of the intrusion detection area. Then, a shielding abnormality signal is output. These abnormal signals output from the image sensor 2 are transmitted to the center device 6 via the controller 5.

  The authentication device 3 authenticates authorized users such as residents and employees in the intrusion detection area by communicating with the portable terminal 4 in the intrusion detection area. The mobile terminal 4 possessed by the authorized user is registered in the authentication device 3 in advance, and when the authentication device 3 detects the registered mobile terminal 4, it outputs an authentication signal to the image sensor 2 via the controller 5. When the image sensor 2 detects a person, the image sensor 2 determines whether the user is an authorized user or an intruder by referring to the presence or absence of an authentication signal.

  FIG. 2 is a block diagram showing the configuration of the image sensor 2. The image sensor 2 includes an imaging unit 21, a setting input unit 22, a storage unit 23, and a communication unit 25 connected to a control unit 24.

  The imaging unit 21 is a measurement unit that measures an intrusion detection area at predetermined time intervals and outputs a measurement signal. The imaging unit 21 is realized by a surveillance camera, and is fixedly installed at a height of about 3 to 4 m so as to face the intrusion detection area with its optical axis obliquely downward. The imaging unit 21 captures an intrusion detection area, for example, at 1/5 second intervals, outputs a monitoring image, and sequentially outputs the monitoring image to the control unit 24. Hereinafter, the unit of time recorded at this time interval is referred to as time.

  FIG. 3 illustrates a monitoring image 100 obtained by imaging the building 101 and its outer periphery. Incidentally, 102 in the figure is a part outside the site in the field of view.

  The setting input unit 22 is an interface device that inputs various settings used in the image sensor 2. The setting input unit 22 is temporarily connected to the communication unit 25 and communicates with the control unit 24 before the operation of the outer periphery monitoring system 1. The setting input by the setting input unit 22 is stored in the storage unit 23 via the control unit 24.

  The setting input unit 22 includes an input display unit such as a liquid crystal touch panel and a communication unit, displays a monitoring image received from the imaging unit 21, superimposes the position indicated on the display, and displays the position on the control unit 24. Can be sent.

  The setting input unit 22 may be connected to the controller 5 so that the setting input unit 22 and the control unit 24 communicate with each other via the controller 5 and the communication unit 25.

  The storage unit 23 is a storage device such as a ROM (Read Only Memory) or a RAM (Random Access Memory). The storage unit 23 stores various programs and various data, and inputs and outputs such information to and from the control unit 24. Various data includes a camera parameter 231, a building area 232, an intrusion detection area 233, a shielding detection area 234, a background image 235, and change area information 236.

The control unit 24 includes a CPU (Central Processing Unit), a DSP (Digital Signal).
Processor, a control unit such as MCU (Micro Control Unit), and the like, by reading a program from the storage unit 23 and executing it, a monitoring area setting unit 241, a change region extraction unit 242, a change region tracking unit 243, an intrusion It functions as the determination unit 244 and the shielding determination unit 245.

  The communication unit 25 includes a communication circuit and driver software between the controller 5, and the control unit 24 is connected to one of the communication units 25, and the controller 5 and the setting input unit 22 are connected to the other. An abnormal signal or the like from the control unit 24 is transmitted to the controller 5 via the communication unit 25, and an authentication signal or the like from the controller 5 is input to the control unit 24 via the communication unit 25. When the setting input unit 22 is connected to the communication unit 25, the monitoring image from the imaging unit 21 is transmitted to the setting input unit 22 via the control unit 24 and the communication unit 25, and the setting information from the setting input unit 22 and the like. Is input to the control unit 24 via the communication unit 25.

  Hereinafter, the camera parameter 231, the building area 232, the intrusion detection area 233, the shielding detection area 234, the background image 235, and the change area information 236 stored in the storage unit 23 will be described.

  The camera parameters 231 include internal parameters such as external parameters such as the installation position and imaging direction of the imaging unit 21, focal length, angle of view, lens distortion, and other lens characteristics of the imaging unit 21, and the number of pixels of the imaging device. These parameters obtained by actual measurement or the like are set in advance and stored in the storage unit 23. By applying the camera parameter 231 to a camera model such as a known pinhole camera model, the pixel position in the monitoring image is represented by coordinates on the imaging surface of the imaging unit 21 (imaging surface coordinates) and coordinates in the real space (real coordinates). It is possible to convert coordinates between the two. Hereinafter, the conversion from the imaging surface coordinates to the actual coordinates using the camera parameter 231 is referred to as reverse perspective conversion, and the conversion from the actual coordinates to the imaging surface coordinates using the camera parameter 231 is referred to as perspective transformation.

  The building is an intruder's purpose of intrusion in the outer periphery monitoring system 1 and is an important monitoring object. An important area where important objects exist in the outer periphery monitoring system 1 is a building area 232. The building area 232 is set and input as an image area in the monitoring image from the setting input unit 22 by the administrator, and is stored in the storage unit 23 in the imaging plane coordinates. A shielding detection area 234 is calculated based on the building area 232.

  FIG. 4 illustrates a building area 111 set for the monitoring image 100 of FIG.

  The intrusion detection area 233 is an area where the outer periphery monitoring system 1 detects the presence of an intruder as an abnormality. The intrusion detection area 233 is the entire field of view of the imaging unit 21 unless otherwise specified. However, when the administrator inputs a non-monitoring area from the setting input unit 22, the non-monitoring area is excluded from the field of view of the imaging unit 21. It becomes an area. The intrusion detection area 233 is stored in the imaging plane coordinates as an image area in the monitoring image.

  FIG. 5 illustrates an intrusion detection area 121 set for the monitoring image 100 of FIG. The intrusion detection area 121 is set by removing the portion corresponding to the off-site portion 102.

  The shielding detection area 234 is an area for detecting the presence of a shield and the presence of an intruder as an abnormality in the intrusion detection area 233. The shielding detection area 234 surrounds the building area 232 in the intrusion detection area 233 and is set to be narrower than the intrusion detection area 233, and is stored as an image area in the monitoring image in the imaging plane coordinates. The shielding detection area 234 is automatically calculated based on the building area 232 and stored in the storage unit 23.

  In FIG. 6, an area surrounded by a thick line is an example of the occlusion detection area 131 set for the monitoring image 100 of FIG. 3. The dotted line in the figure is the outline of the building area 111. A range of width W and height H surrounding the building area 111 is calculated in real coordinates, and this range is converted into imaging surface coordinates to set the shielding detection area 131.

  FIG. 7 is a schematic diagram illustrating the relationship between the width W of the shielding detection area, the stillness determination distance Td, the measurement interval (imaging interval), and the height H of the shielding detection area. V in the figure is a distance that an intruder can move at one time that is engraved at a measurement interval. The movable distance V is set assuming an intruder that moves at high speed, that is, an intruder that runs through. The distance V is a value that depends on the measurement interval. S in the figure is the maximum width of the intruder. The width S is set assuming an intruder taking a forward leaning posture, that is, an intruder who runs through. The stillness determination distance Td is a threshold value that is compared with the moving distance of the change region in order to discriminate between an intruder that is a moving object and a shield that is a stationary object. The stillness determination distance Td is set to 0.5 m, for example.

  The width W of the occlusion detection area is the minimum width at which the intruder can be detected based on the measurement signal (monitoring image), in other words, the minimum distance at which the movement distance when the intruder moves the stationary determination distance Td can be calculated. Set to the width of. The reason why the width is set to the minimum is that the width W is also a value that restricts the installation of the article by the authorized user, so that the narrowest one is more convenient for the authorized user. There are the following two methods for setting the width W depending on the length of the measurement interval.

  One of the setting methods of the width W is a setting method when the measurement interval is long, that is, when the movable distance V is larger than the stillness determination distance Td. FIG. 7A is a diagram for explaining the setting when V> Td. In this case, the width W is set to a distance at which an intruder can be measured at least at two consecutive times. That is, the width W is set to a width corresponding to the distance that the intruder can move during the measurement interval. Specifically, the width W is set to a width (for example, 2 m) obtained by adding the maximum width S of the intruder to the movable distance V. The reason why the maximum width S of the intruder is added is to ensure that the feature amount of the changed area can be correctly extracted in the tracking process.

  Another setting method of the width W is a setting method when the measurement interval is short, that is, when the movable distance V is equal to or less than the stillness determination distance Td. FIG. 7B is a diagram for explaining the setting when V ≦ Td. In this case, the width W is set to a width corresponding to the stillness determination distance Td. Specifically, the width W is set to a width (for example, 1.5 m) obtained by adding the maximum width S of the intruder to the stillness determination distance Td. The reason why the maximum width S of the intruder is added is to ensure that the feature amount of the changed area can be correctly extracted in the tracking process as described above.

  The height H is set to the highest possible intruder height (for example, 2 m). By providing a height component in the shielding detection area, a shielding object such as a flag at the bottom can be detected.

  The shielding detection area surrounding the building area covers the intrusion route to the building, and the intruder detection performance can be ensured at a minimum by setting the width W of the shielding detection area as described above. Therefore, by setting such a shielding detection area within the intrusion detection area narrower than the intrusion detection area, the intruder detection performance can be ensured to the minimum without narrowing the intrusion detection area, and the article installation by authorized users can be maximized. acceptable.

  8 is an example of another shielding detection area 141 set for the monitoring image 100 in FIG. The shielding detection area 141 in FIG. 8 is set more simply than the shielding detection area 131 in FIG.

  The background image 235 is an image in which only the background of the intrusion detection area is captured. The background image 235 is generated and stored in the storage unit 23 prior to the abnormality detection process. The background image 235 does not include an intruder, a regular user, or a newly brought-in shield, and by comparing the background image 235 with the monitoring image, a portion where the monitoring image has changed due to the appearance of these objects can be obtained. It can be detected as a change area.

  The change area information 236 is data in which the feature amount of the change area extracted from the monitoring image at each time is stored for each corresponding area. The feature amount includes a change area extraction position, a change area size, a change area shape, a luminance histogram of a monitoring image in the change area, and the like. The change areas similar to each other among the change areas extracted at each time are associated as change areas generated from the same object by the tracking process, and the feature amount of each change area is an identification code (area) for each associated area. ID). Each feature amount is used for tracking, and in particular, the extraction position is also used for calculating the moving distance of the change area and determining abnormality.

  Hereinafter, the monitoring area setting unit 241, the changed region extracting unit 242, the changed region tracking unit 243, the intrusion determining unit 244, and the shielding determining unit 245 that the control unit 24 has as functions thereof will be described.

  Based on the input from the setting input unit 22, the monitoring area setting unit 241 sets an important area (building area 232), an intrusion detection area 233, and a shielding detection area 234 that are monitoring areas. The operation of the monitoring area setting unit 241 will be described later.

  The change area extraction unit 242 compares the monitoring image input from the imaging unit 21 with the background image 235, extracts the change area in the monitoring image, and outputs the extracted change area to the change area tracking unit 243.

  Specifically, the change area extraction unit 242 extracts a change area by background difference processing. That is, the change area extraction unit 242 calculates the absolute value of the difference in pixel value between the corresponding pixels of the monitoring image and the background image 235, and selects pixels that are close to each other among the pixels whose absolute value of the difference is greater than or equal to the difference threshold value. Change areas are extracted together. The difference threshold is set in advance based on a preliminary experiment.

  In another embodiment, the change area extraction unit 242 extracts a change area by background correlation processing. That is, the change area extraction unit 242 calculates a correlation value of pixel values instead of the above difference, and extracts a pixel whose correlation value is equal to or less than a preset correlation threshold as a change area.

  In addition, the change area extraction unit 242 adapts the background image 235 to environmental changes in the monitoring space by appropriately updating the background image 235 using the monitoring image in order to continue to extract the changing area accurately.

  First, the change area extraction unit 242 captures the accepted occlusion object (acceptance occlusion object) in the background image 235 so that the change area caused by the occlusion object is not mixed with the change area that appears in front of the occlusion object. Specifically, the background image 235 is replaced with the monitoring image in the change area of the acceptable shielding object.

  Secondly, the change area extraction unit 242 reflects the illumination variation of the monitoring space in the background image 235 by replacing the background image 235 with the average image with the monitoring image in the area excluding the changing area.

  The object area tracking unit 243 is a movement distance calculation unit that calculates the movement distance of the change area by the tracking process. The change area tracking unit 243 associates similar change areas among the change areas extracted from the monitoring images at the preceding and following times, calculates the movement distance of the change area from the extracted position of the change area, The calculated movement distance is output to the intrusion determination unit 244 and the shielding determination unit 245. Also, the change area tracking unit 243 assigns a common area ID to each associated change area, and causes the storage unit 23 to store change area information 236 in which the feature amount of each change area is associated with the area ID.

  When a plurality of change areas exist simultaneously, the above process is performed for each change area. Incidentally, when an article is installed by a regular user, the regular user and the article are tracked as one change area during transportation, and when separated after the installation, the regular user and the article are tracked as separate change areas.

  Here, when the movement distance is calculated using the imaging surface coordinates, a difference in the movement distance occurs depending on the extraction position of the change area. That is, in the imaging plane coordinates, the moving distance is easily calculated larger as the changed region is extracted near the imaging unit 21. Therefore, the change area tracking unit 243 converts the extraction position of the change area into real coordinates by inverse perspective transformation and calculates the movement distance in the real coordinates.

  The intrusion determination unit 244 outputs an intrusion abnormality signal based on the fact that the moving distance of the change area extracted in the intrusion detection area is greater than or equal to a preset stillness determination distance. Specifically, the intrusion determination unit 244 compares the extraction position of each change area at each time with the intrusion detection area 233, compares the movement distance of the change area to the time with the stationary determination distance, and at the time The presence / absence of the authentication signal is confirmed, and if there is a change area where the extraction position is within the intrusion detection area 233 and the moving distance is equal to or greater than the stationary determination distance, and the authentication signal is not confirmed, an intruder abnormal signal is assumed to be present in the monitoring space. Is output. On the other hand, when there is no change area extracted in the intrusion detection area 233, when there is no change area whose movement distance exceeds the stationary determination distance, and when an authentication signal is confirmed, the intrusion determination unit 244 does not output an intrusion abnormality signal. .

  The intrusion determination unit 244 further compares the size, shape, and movement of the change area with a preset reference value for humanity and a reference value for disturbance (such as small animal), and the monitoring image in the change area. The disturbance attribute value and human attribute value of the change region are calculated by weighting and adding the texture similarity of the background image 235 (similarity is low for humans and high for light and shadows). It can be added to the output condition of the intrusion abnormality signal that the human attribute value is higher.

  The shielding determination unit 245 determines the presence of the shielding object based on the movement distance of the change area extracted in the shielding detection area being less than the stationary determination distance. Specifically, the occlusion determination unit 245 compares the extraction position of each change area at each time with the occlusion detection area 234, compares the movement distance of the change area up to the time with the stationary determination distance, and the change area. Compared with the stationary determination time set in advance, the intrusion detection is prevented when the extraction position is within the shielding detection area 234, the moving distance is less than the stationary determination distance, and there is a change area of the tracking time. A shielding abnormality signal is output because there is a shielding object. The stillness determination time is a determination standby time for preventing erroneous recognition of a moving object that is temporarily stopped, and is set to 10 seconds, for example. The tracking time can be obtained by converting the number of extraction positions of each change area stored in the change area information 236.

  On the other hand, even if the movement distance is a change area that is less than the stillness determination distance, if the change area is tracked outside the shielding detection area 234, the shielding determination unit 245 determines that the change area does not interfere with intrusion detection. This is recognized as being due to a shielding object, and a shielding abnormality signal is not output. Further, the occlusion determination unit 245 outputs a change area due to the acceptable occlusion object to the change area extraction unit 242 in order to update the background image 235.

  The shielding determination unit 245 may accept a change area that is extremely small compared to the size of the intruder regardless of the tracking position. In this case, the shielding determination unit 245 compares the size (number of pixels) of the change region with a preset allowable size and does not output a shielding abnormality signal if the size is less than the allowable size. The acceptable size can be, for example, the size of a flower pot (20 cm square).

[Operation of outer periphery monitoring system 1]
Hereinafter, the operation of the outer periphery monitoring system 1 will be described focusing on the operation of the image sensor 2.

<Pre-setting process>
Prior to monitoring, the image sensor 2 performs pre-setting processing for a monitoring area and the like. When the setting input unit 22 is connected to the communication unit 25, communication via the communication unit 25 becomes possible between the control unit 24 and the setting input unit 22, and the pre-setting process illustrated in the flowchart of FIG. 9 is started.

  First, the setting input unit 22 receives input of camera parameters 231. The setting input unit 22 outputs the camera parameters 231 input by the administrator to the control unit 24, and the control unit 24 stores the camera parameters 231 input from the setting input unit 22 in the storage unit 23 (S1). The setting of the camera parameter 231 is referred to from each unit, and the perspective transformation and the reverse perspective transformation using the camera parameter 231 can be performed in each unit.

  Next, the monitoring area setting unit 241 of the control unit 24 outputs the monitoring image from the imaging unit 21 to the setting input unit 22, and the setting input unit 22 displays the monitoring image on the display unit and receives the input of the building area 232. . The setting input unit 22 outputs the building area 232 input by the administrator while visually confirming the building image in the monitoring image to the monitoring area setting unit 241, and the monitoring area setting unit 241 is input from the setting input unit 22 The building area 232 is stored in the storage unit 23 (S2). Since the building is easily visible on the monitoring image, the administrator can easily set the building area 232.

  The building area is graphically input using a pointing device on the displayed monitoring image. In the example of FIG. 3, four points P <b> 1, P <b> 2, P <b> 3, and P <b> 4 are input by the setting input unit 22 corresponding to the display of the monitoring image 100 including the building 101 of FIG. The monitoring area setting unit 241 calculates the inside of a quadrangle 111 connecting these two adjacent points with a straight line as a building area, and sets the inside and outside of the building area 111 to different pixel values. Is stored in the storage unit 23.

  Subsequently, the monitoring area setting unit 241 sets the intrusion detection area 233 (S3). The setting input unit 22 receives an input of a non-monitoring area that is outside the premises of the building owner. The setting input unit 22 outputs the non-monitoring area input by the administrator to the monitoring area setting unit 241, and the monitoring area setting unit 241 removes the non-monitoring area input from the setting input unit 22 from the entire area of the monitoring image. The intrusion detection area 233 is stored in the storage unit 23. If no non-monitoring area is input, the intrusion detection area 233 is the entire area of the monitoring image.

  In the example of FIG. 4, three points Q <b> 1, Q <b> 2, and Q <b> 3 are input at the setting input unit 22 corresponding to the display of the monitoring image 100 including the off-site portion 102 of FIG. 3. The monitoring area setting unit 241 calculates the intrusion detection area 121 by removing the triangular non-monitoring area connecting these three points from the entire area, and the binary image 120 in which the inside and outside of the intrusion detection area 121 are set to different pixel values. Is stored in the storage unit 23.

  Subsequently, the monitoring area setting unit 241 sets the shielding detection area 234. The width W and height H used here are set in advance.

  First, the monitoring area setting unit 241 calculates a ground contact area having a width W on the outer periphery of the building area 232 (S4). In other words, the monitoring area setting unit 241 selects the contour pixels of the building area 232 and inversely transforms the coordinates of each contour pixel into actual coordinates (contour actual coordinates). At this time, the contour pixels on the outer frame of the monitoring image may be excluded. In the case of the building area 111 illustrated in FIG. 3, the pixels on the line segments P2-P3 and P3-P4 are selected and converted as contour pixels. Then, the monitoring area setting unit 241 calculates a circle having a radius W with each of the contour outline coordinates as the center, and perspective-transforms the sum area of each circle into the imaging plane coordinates. The monitoring area setting unit 241 calculates a ground contact area by excluding a portion overlapping with the building area 232 from the sum area thus obtained.

  Next, the monitoring area setting unit 241 calculates the shielding detection area 234 by adding the height component of the height H to the ground contact region, and stores it in the storage unit 23 (S5). In other words, the monitoring area setting unit 241 performs reverse perspective transformation of the coordinates of each pixel in the ground area into real coordinates. Then, the monitoring area setting unit 241 calculates a line segment having a length H in the vertical direction (the height direction of the image capturing unit 21) from each of the actual coordinates of the ground contact area, and perspective-converts each line segment to the image capturing surface coordinates. A region filled with each line segment thus converted is a shielding detection area 234 having a width W and a height H. From the building area 111 illustrated in FIG. 3, a shielding detection area 131 indicated by a thick line in FIG. 4 is calculated. The monitoring area setting unit 241 causes the storage unit 23 to store a binary image 130 in which the inside and outside of the shielding detection area 131 are set to different pixel values.

  The monitoring area setting unit 241 generates a monitoring image in which the set monitoring area is superimposed and transmits the generated monitoring image to the setting input unit 22, and the setting input unit 22 displays the received monitoring image on the display unit. By this display, the administrator and the user can easily confirm the range in which the article installation is permitted and the range in which the article installation is not permitted.

  When the monitoring area is set in this way, the pre-setting process ends.

  In step S4, the ground contact area is calculated based on the circle. However, the ground contact area can be simply calculated based on the line segment. In this case, the monitoring area setting unit 241 calculates line segments having a length W perpendicular to the contour line from the contour actual coordinates on the ground contact surface in the real space, and perspective-converts these line segments to change the ground region. Can be calculated. The shielding detection area 141 in FIG. 6 is an example calculated from a ground contact area based on a line segment.

<Image monitoring processing>
When the administrator who has confirmed that the monitoring space is unmanned restarts the image sensor 2 that has completed the pre-setting process, the image monitoring process shown in the flowchart of FIG. 10 is started.

  After activation, the change area extraction unit 242 of the control unit 24 creates an initial background image 235 by averaging the monitoring images from the imaging unit 21 for a preset initialization time, and stores the initial background image 235 in the storage unit 23. (S10) After the initialization time has elapsed, each time a monitoring image is input from the imaging unit 21 to the control unit 24, the processes in steps S11 to S19 are repeated. Hereinafter, the time when a new monitoring image is input is referred to as the current time.

  First, the control unit 24 acquires a monitoring image at the current time (S11), and the change region extraction unit 242 of the control unit 24 extracts the change region by performing a differential process on the monitoring image at the current time and the background image 235 (S12). After the extraction process, the changed region extracting unit 242 outputs the changed region extraction result to the changed region tracking unit 243 of the control unit 24. If there is no change region, the change region extraction unit 242 outputs to the change region tracking unit 243 that there is no change region.

  The change area tracking unit 243 associates a change area similar between the change area extracted at the current time and the change area extracted one time ago (S13).

  Therefore, the change area tracking unit 243 stores the extraction position and the feature amount of each change area extracted up to one time ago in the storage unit 23 as the change area information 236. The change area tracking unit 243 obtains the size (number of pixels) of each change area extracted at the current time, calculates the difference between each change area stored and calculates each change extracted at the current time. The difference between the shape of each change area that is stored by calculating the area shape (aspect ratio) is calculated, and the brightness of each change area that is stored by calculating the luminance histogram of each change area extracted at the current time Calculate the difference from the histogram, extrapolate the extracted position at the current time to the stored extracted position history of each changed area, and the extracted position of each changed area detected at the current time The difference is calculated, and each difference is applied to a preset evaluation function to calculate the similarity for each combination of change areas, and a combination exceeding a preset similarity threshold is associated.

  The changed area tracking unit 243 that has completed the association updates the changed area information 236 with the association result (S14).

  That is, the change area tracking unit 243 assigns the same area ID as the change area of the storage unit 23 in which the correspondence is obtained to the extraction position and the feature amount of the change area at the current time, and additionally writes the change area information 236. In addition, the change area tracking unit 243 assumes that the change area at the current time for which the association has not been obtained is a new appearance area, and assigns a new area ID to the extraction position and the feature amount of the change area to change area information 236. Add to. Further, the change area tracking unit 243 determines that the change area of the storage unit 23 for which the association has not been obtained is the disappearance area, and deletes the extraction position and the feature amount of the change area from the change area information 236.

  Subsequently, the change area tracking unit 243 refers to the change area information 236 and calculates the movement distance from the extraction position of each change area (S15). That is, the change area tracking unit 243 calculates the movement distance by accumulating the distances between the actual positions at the preceding and succeeding times by converting the extraction position of each change area into the actual position by reverse perspective transformation.

  When the tracking process and the movement distance calculation process progress in this way, an abnormality determination is performed by the intrusion determination unit 244 and the shielding determination unit 245 of the control unit 24 (S16). The abnormality determination process will be described with reference to FIG.

  In the abnormality determination, the control unit 24 sequentially sets each change region stored in the change region information 236 as a target change region, and executes the loop process of S30 to S42 for the change region being tracked.

  In the loop processing, first, the intrusion determination unit 244 compares the moving distance of the attention change region with the stationary determination distance (S31), and if the moving distance is less than the stationary determination distance, the subsequent steps S32 to S36 are skipped (in S31). NO → S37). Further, the intrusion determination unit 244 confirms whether or not the extraction position of the attention change area at the current time is within the intrusion detection area (S32), and if the extraction position is not within the intrusion detection area, the subsequent steps S33 to S36 are skipped. (NO in S32 → S37).

  On the other hand, if the movement distance is equal to or greater than the stationary determination distance and the extraction position is within the intrusion detection area (YES in S31), the intrusion determination unit 244 determines that the intent change area may be a change area by an intruder, and the intrusion determination unit 244 performs step The process proceeds to S33. The intrusion determination unit 244 calculates the human attribute value and disturbance attribute value of the attention change area and compares these values (S33, S34). If the human attribute value is greater than or equal to the disturbance attribute value, the attention change area is determined by the person. Whether or not the authentication signal has been received is confirmed as a change area (YES in S34 → S35). If the authentication signal has not been received, intrusion determination unit 244 generates intrusion abnormality information on the assumption that the region of interest change is a region changed by an intruder, and temporarily stores it in storage unit 23 (NO in S35 → S36).

  On the other hand, if the human attribute value is less than the disturbance attribute value, the intrusion determination unit 244 skips Steps S35 and S36 (NO in S34 → S37), assuming that the region of interest change is due to a disturbance such as a small animal. If the reception of the authentication signal has been received, the intrusion determining unit 244 skips step S36 (YES in S35 → S37), assuming that the target change area is that of the authorized user. Disturbances and regular user change areas will be tracked until they disappear.

  Subsequently, the shielding determination unit 245 determines the attention change area.

  First, in steps S37 and S38, the shielding determination unit 245 confirms that the attention change area is a change area due to a stationary object. That is, the shielding determination unit 245 compares the moving distance of the attention change region with the still determination distance (S37), and skips the subsequent steps S38 to S42 if the moving distance is equal to or greater than the still determination distance (NO in S37 → S43). ). If the moving distance is less than the stationary determination distance (YES in S37), shielding determination unit 245 further compares the tracking time with the stationary determination time (S38). If tracking time is less than stationary determination time, shielding determination unit 245 skips subsequent steps S39 to S42 (NO in S38 → S43).

  The occlusion determination unit 245 that has confirmed that the attention change area is due to a stationary object checks whether or not the extraction position of the attention change area at the current time is within the occlusion detection area (S39), and further the number of pixels of the attention change area ( (Size) is counted and compared with the acceptable size (S40). If the extracted position is within the shielding detection area and the size is greater than or equal to the permissible size, shielding abnormality information is generated and temporarily stored in the storage unit 23, assuming that the region of interest change is due to a shielding object that prevents intrusion detection (in S39). YES → YES in S40 → S41).

  On the other hand, if the extraction position of the attention change area is not within the shielding detection area, or if the size is less than the allowable size, the attention change area is assumed to be due to the acceptable shielding object, and the information of the attention changing area is determined as the acceptable shielding information. (NO in S39 → S40, YES in S40 → S40).

  When the shielding determination unit 245 finishes the process of step S41 or S42, the control unit 24 confirms whether the determination of all the change regions is completed (S43). If the determination of all the change areas has not been completed (NO in S43), the control unit 24 returns the process to step S30 to start the determination of the next change area, and if the determination of all the change areas has been completed (S43). The control unit 24 advances the processing to step S17 in FIG.

  Returning to FIG. 10, the intrusion determination unit 244 and the shielding determination unit 245 confirm the presence or absence of temporarily stored abnormality information (S17). If there is intrusion abnormality information, the intrusion determination unit 244 outputs the information to the communication unit 25, If there is shielding abnormality information, shielding judgment unit 245 outputs the information to communication unit 25 (YES in S17 → S18). If there is no abnormality information (NO in S17), step S18 is skipped.

  Subsequently, the change area extraction unit 242 updates the background image 235 using the monitoring image (S19). That is, the change area extraction unit 242 weights and adds the image obtained by cutting out the area where the change area is not detected from the monitoring image at the current time and the background image 235 at a ratio of α: (1-α). The image 235 is updated, and if there is an area determined to be an acceptable shielding object change area at the current time, the area is cut out from the monitoring image, and the corresponding portion in the background image 235 is replaced with the cut out image. 235 is updated. However, α is a preset value in the range of 0 <α <1.

  When the above process is completed, the process returns to step S11, and the process for the monitoring image at the next time is performed.

  In the above embodiment, the imaging unit 21 using a monitoring camera as an example of the measurement unit has been illustrated, but a scanning laser sensor may be used as the measurement unit. In this case, the measurement unit measures the monitoring area by laser scanning and outputs a distance image, and the change region extraction unit performs generation / update of the background image and extraction of the change region using the distance image. The process using the change area by the movement distance determination unit (change area tracking unit), the intrusion determination unit, and the shielding determination unit can be performed in the same manner as in the above embodiment.

  The intrusion determination unit 244 can also be realized by an infrared sensor, a laser sensor, a microwave sensor, a thermal image sensor, or the like. In this case, adjustment may be made so that the visual field of the intrusion determination unit 244 and the visual field of the imaging unit 21 coincide.

  Moreover, although the example which monitors the outer periphery of a building was shown in the said embodiment, it is applicable to the use which monitors the circumference | surroundings of various important things regardless of the inside and outside, such as a safe and its periphery.

In the above embodiment, an example in which the shielding detection area is automatically created has been described. However, the administrator may operate the setting input unit 22 to set and input the shielding detection area. In this case, one of the managers may actually measure the movement determination distances to indicate a plurality of representative positions, and the other of the managers may input the positions pointed to by the actually measured person on the display of the setting input unit 22.

DESCRIPTION OF SYMBOLS 1 ... Perimeter monitoring system 2 ... Image sensor 3 ... Authentication apparatus 4 ... Portable terminal 5 ... Controller 6 ... Center apparatus 21 ... Imaging part 22 ... Setting input part 23 ... storage unit 231 ... camera parameter 232 ... building area 233 ... intrusion detection area 234 ... shielding detection area 235 ... background image 236 ... change area information 24 ... control part 241 ... Monitoring area setting unit 242 ... Change region extraction unit 243 ... Change region tracking unit 244 ... Intrusion determination unit 245 ... Occlusion determination unit 25 ... Communication unit

Claims (3)

An intrusion detection device for detecting an intruder into an intrusion detection area including an important object and its outer periphery,
Wherein the intrusion detection area and the image pickup unit for outputting an image captured at predetermined time intervals intrusion detection area, the important area important object is present, and shielding detection area for detecting the presence of a shield that prevents the imaging A storage unit for storing
A change area extraction unit that extracts the change area by comparing the images at different time points;
A movement distance calculation unit for calculating a movement distance from the extraction position of the change area at each time point;
An intrusion determination unit that determines the presence of the intruder based on the movement distance of the change area extracted in the intrusion detection area being equal to or greater than a predetermined stillness determination distance;
A shielding determination unit that determines the presence of the shielding object based on the movement distance of the change region extracted in the shielding detection area being less than the stationary determination distance;
A setting input unit capable of displaying the image and inputting the important area on the display;
Based on the important area input from the setting input unit, the shielding detection area within a range in which the moving distance can be calculated when the intruder moves the static determination distance around the important area is calculated. A monitoring area setting section to be set,
Intrusion detecting apparatus is characterized in that example Bei a. The intrusion detection device according to claim 1,
The monitoring area setting unit is an intrusion detection device that sets the shielding detection area within a range in which the intruder can move around the important area in the predetermined time. The intrusion detection device according to claim 1,
The monitoring area setting unit is an intrusion detection device that sets the shielding detection area in a range in which the intruder moves the static determination distance around the important area.