JP5170731B2 - Traffic monitoring system - Google Patents

Description

  The present invention relates to a traffic monitoring system, and more particularly to a traffic monitoring system for monitoring unauthorized entry into a restricted access area.

  2. Description of the Related Art Conventionally, in order to monitor unauthorized entry into a restricted access area, there is a system that captures a region near the entrance to the restricted access area with a monitoring camera and records the captured image on a hard disk or the like.

  For example, in Patent Document 1 below, a distance sensor (distance image capturing means) is installed above the area near the entrance to the entrance restricted area, and distance information (distance image information) to the subject detected by the distance sensor is included. Based on this, a monitoring system for extracting a person passing through an entrance is disclosed.

Japanese Patent Laying-Open No. 2006-243972 (see FIG. 2A and paragraphs 0048 to 0050 of the specification)

  However, according to the monitoring system disclosed in Patent Document 1, when the intruder attempts to pass through the entrance with a posture in which he / she leans on the floor, the distance from the distance sensor to the floor and the distance sensor to the intruder Since there is no large difference in distance, there is a possibility that the distance sensor cannot detect the presence of an intruder.

  The present invention has been made in view of such circumstances, and is a traffic monitoring system that can reliably monitor that a non-permitted person passes through an entrance to an access restricted area (so-called companionship) due to traffic by a permitter. The purpose is to obtain.

A traffic monitoring system according to a first aspect of the present invention is a camera for photographing a predetermined entrance / exit provided with a door including a region near the entrance / exit, a first detection unit for detecting an open / closed state of the door, Based on the image taken by the camera, the second detection unit that detects a person moving in the vicinity of the entrance and the detection result by the first and second detection units, the door is opened. And a determination unit that determines whether one or more persons have passed through the doorway until the next closing, and based on the video imaged by the camera, a part of the entire area of the image captured by the camera, Shooting by the camera based on a first setting unit for setting a first monitoring area to be monitored by the first detection unit to detect the open / closed state of the door, and an image shot by the camera In the image And a second setting unit that sets the specified entrance / exit as a second monitoring area to be monitored by the determination unit in order to detect a person passing through the entrance / exit. The first detection unit detects an open / closed state of the door based on an image taken by the camera, and the second setting unit is configured to detect the position of the door in a closed state, The doorway is specified based on the position of the door detected by the first detection unit in the opened state.

The traffic monitoring system according to a second aspect of the invention is the traffic monitoring system according to the first aspect of the invention, particularly in the situation where the first detection unit detects the door in the opened state. When the door in the open state is no longer detected, it is detected that the door has been closed because the door has not been detected in the open state for a predetermined period of time. It is characterized by.

Traffic monitoring system according to the third invention is the traffic monitoring system according to the first or second aspect, before Symbol camera is installed above the region near the doorway, said second detector Detects the person by detecting a circle in an image captured by the camera, and the second detection unit includes a plurality of circles when the plurality of overlapping or adjacent circles are detected. One circle is defined, and the one circle is detected as one person.

Traffic monitoring system according to the fourth invention is the traffic monitoring system according to the first or second aspect, before Symbol camera is installed above the region near the doorway, said second detector Detects the person by detecting a circle in the image captured by the camera, and when the circle moves in the image captured by the camera, the second detection unit detects center position information of the circle And determining whether the circles before and after the movement are circles corresponding to the same person based on the radius information.

  According to the traffic monitoring system according to the first aspect of the invention, the first detection unit detects the open / closed state of the door, the second detection unit detects a person moving in the vicinity area of the doorway, Based on the detection results of the first and second detection units, it is determined whether one or more persons have passed through the doorway between the opening and closing of the door. Therefore, it is possible to reliably monitor the passage of the non-permitted person through the doorway (so-called companionship) due to the passage by the permitted person.

In the traffic monitoring system according to the first aspect of the invention, the first detection unit detects the open / closed state of the door based on the video imaged by the camera. Therefore, it is not necessary to install a dedicated sensor for detecting the open / closed state of the door, so that the cost can be reduced.

In the traffic monitoring system according to the first aspect of the invention, the first setting unit monitors the opening / closing state of the door in a part of the entire area of the image captured by the camera. A first monitoring area to be set is set. Therefore, it is sufficient for the first detection unit to monitor only the first monitoring area, and it is not necessary to monitor the entire area of the photographed image, so that the processing load can be reduced.

Further , according to the traffic monitoring system of the first invention, the second setting unit specifies an entrance / exit in a photographed image by the camera, and detects the person passing through the entrance / exit by using the specified entrance / exit. The determination unit sets the second monitoring area to be monitored. Therefore, the determination unit only needs to monitor the second monitoring area, and does not need to monitor a larger area, so that the processing load can be reduced. In addition, since an intruder attempting to enter the restricted access area always passes through the entrance / exit, it is possible to reliably detect the presence of the intruder by setting the entrance / exit itself as the second monitoring area.

Further , according to the traffic monitoring system of the first invention, the second setting unit includes the position of the door in the closed state and the door detected by the first detection unit in the opened state. The entrance / exit is specified based on the position. Therefore, the second setting unit can appropriately set the second monitoring area according to the degree of opening of the door at the current time.

According to the traffic monitoring system according to the second aspect of the invention, the first detection unit detects that the door has been closed because the state in which the opened door is no longer detected continues for a predetermined time. Therefore, even if the first detection unit cannot instantaneously detect a door that is actually opened due to light adjustment, noise, or the like, the first detection is that the door is closed. It is possible to avoid erroneous detection by the detection unit. As a result, it is possible to accurately determine the passerby's single / multiple determination by the determination unit.

According to the traffic monitoring system according to the third aspect of the invention, when the second detection unit detects a plurality of overlapping or adjacent circles, the second detection unit defines one circle including the plurality of circles, and the one circle Is detected as one person. Accordingly, when a plurality of circles such as the head and shoulders are detected for the same person, one circle including the plurality of circles is detected as one person. As a result, for example, even if a person moving in the vicinity of the entrance / exit makes a sudden movement and the amount of movement of the head of the person increases momentarily, the person is accurately identified as the same person. Can be tracked.

According to the traffic monitoring system according to the fourth aspect of the present invention, when the circle moves within the image captured by the camera, the second detection unit detects the circles before and after the movement based on the center position information and the radius information of the circle. It is determined whether the circles correspond to the same person. The size of the circle corresponding to the person will differ depending on the camera installation position, installation angle, etc., but by using not only the center position information of the circle but also the radius information, the moving person can be tracked more accurately Can be done.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In addition, the element which attached | subjected the same code | symbol in different drawing shall show the same or corresponding element.

  1 and 2 are a side view and a top view, respectively, showing an environment in which a traffic monitoring system according to an embodiment of the present invention is introduced. The traffic monitoring system according to the present embodiment is a system for monitoring a person passing through an entrance / exit to an access restricted area. Here, the entry restricted area means an area where entry of persons other than those who are permitted to enter the room in advance (hereinafter referred to as “permitted person”) is prohibited.

  Referring to FIGS. 1 and 2, a door 3 is provided at the entrance 2 of the restricted access area. 1 and 2, the right side of the door 3 is an access restriction area. A card reader 4 is installed on the wall next to the door 3. A predetermined card key is given to the authorized person in advance, and the lock of the door 3 is temporarily released by reading the contents of the card key by the card reader 4. In addition, a camera 1 capable of moving image shooting is installed on the ceiling above the area 5 near the entrance 2. The camera 1 photographs the entrance / exit 2 provided with the door 3 including the vicinity area 5 of the entrance / exit 2 (see the broken line in FIG. 1).

  FIG. 3 is a block diagram showing a functional configuration of the image processing unit 10 in the traffic monitoring system according to the present embodiment. A processing unit 11 is connected to the rear stage of the camera 1. A difference image generation unit 12 and an image storage unit 13 are connected to the subsequent stage of the processing unit 11. A first setting unit 14, a first detection unit 15, a second detection unit 16, and a second setting unit 17 are connected to the subsequent stage of the difference image generation unit 12. A determination unit 19 is connected to the subsequent stage of the first detection unit 15 and the second detection unit 16. A timer 18 is connected to the first detection unit 15.

  An image D0 captured by the camera 1 is input to the processing unit 11. The processing unit 11 performs edge enhancement processing and binarization processing on the image D0.

  FIG. 4 is a diagram illustrating a background image D1 that is captured by the camera 1 with the door 3 closed and subjected to edge enhancement processing and binarization processing by the processing unit 11. In the background image D1, the floor of the door 3 and the nearby area 5 is shown. With reference to FIG. 3, the background image D <b> 1 is stored in the image storage unit 13.

  FIG. 5 is a diagram illustrating an image D <b> 2 captured by the camera 1 with the door 3 opened by the person 20 who is the permitter and subjected to edge enhancement processing and binarization processing by the processing unit 11. In the image D2, in addition to the door 3 and the floor in the vicinity area 5, a part of the entrance 2 and a person 20 are shown. With reference to FIG. 3, the image D <b> 2 is input to the difference image generation unit 12.

  The difference image generation unit 12 generates a difference image D3 based on the image D2 input from the processing unit 11 and the background image D1 read from the image storage unit 13. At that time, the difference image generation unit 12 compares the pixel values of the corresponding pixels in the image D2 and the background image D1, respectively, and determines the pixel having a larger pixel value in the image D2 than in the background image D1 in the difference image D3. Is set as a black pixel.

  FIG. 6 is a diagram illustrating the difference image D <b> 3 generated by the difference image generation unit 12. By taking the difference between the background image D1 and the image D2, the difference image D3 includes a part of the edge of the person 30 corresponding to the person 20 in the image D2, and the door 31 corresponding to the door 3 in the image D2. A part of the edge and a part of the edge of the entrance 32 corresponding to the entrance 2 in the image D2 appear. With reference to FIG. 3, the difference image D <b> 3 is input to the first setting unit 14, the first detection unit 15, the second detection unit 16, and the second setting unit 17.

<Processing by the first setting unit 14>
FIG. 7 is a diagram for explaining the contents of processing by the first setting unit 14 shown in FIG. When the traffic monitoring system according to the present embodiment is introduced and the camera 1 is installed at a predetermined location, initial setting is first performed. In such an initial setting, the background image D1 shown in FIG. 4 is acquired, and then the situation in which the door 3 is opened is photographed by the camera 1. Thereby, the difference image generation part 12 shown in FIG. 3 produces the difference image D3 shown in FIGS. The difference image D3 is input to the first setting unit 14.

  When the door 3 is opened, the edge 41 of the door 31 appears in the difference image D3. Therefore, the first setting unit 14 detects the edge 41 in the difference image D3 by a straight line detection method using Hough transform. And the 1st setting part 14 detects the position of the edge 42 corresponding to the door 3 of the closed state as a position of the edge 41 in the moment detected for the first time.

  Next, the first setting unit 14 sets a rectangular or fan-shaped area having a side of the edge 42 (a square area in the example shown in FIG. 7) in the entire image area 25 of the difference image D3 as the first area. Set as the monitoring area 26. Whether the first monitoring area 26 is set to the left or right of the edge 42 can be determined based on the position of the edge 41 detected after several seconds. In the example shown in FIG. 7, since the edge 41 is detected on the left side of the edge 42, the first monitoring area 26 is set to the left of the edge 42.

  The first monitoring area 26 is an area to be monitored by the first detection unit 15 in order to detect the open / closed state of the door 3. The position information D4 of the first monitoring area 26 is the first setting unit 14. Held by. With reference to FIG. 3, the position information D <b> 4 is input from the first setting unit 14 to the first detection unit 15.

  As described above, the first setting unit 14 performs the first detection in order to detect the open / closed state of the door 3 in a part of the entire image area 25 of the image captured by the camera 1 (specifically, the difference image D3). The first monitoring area 26 to be monitored by the unit 15 is set. Therefore, the first detection unit 15 only needs to monitor the first monitoring region 26 and does not need to monitor the entire image region 25. Therefore, the processing load of the first detection unit 15 can be reduced. It becomes possible.

<Processing by the first detection unit 15>
When the traffic monitoring system according to the present embodiment is introduced and the initial setting is completed, monitoring of the vicinity region 5 (see FIGS. 1 and 2) using the camera 1 is started.

  The first detection unit 15 detects the open / closed state of the door 3 based on the video imaged by the camera 1. This will be specifically described below.

  As described above, the position information D4 regarding the first monitoring area 26 set by the first setting unit 14 is input from the first setting unit 14 to the first detection unit 15. In addition, referring to FIG. 3, the difference image D <b> 3 generated by the difference image generation unit 12 is also input to the first detection unit 15.

  Referring to FIG. 7, the first detection unit 15 monitors the first monitoring area 26 in the entire image area 25 of the difference image D3, and performs the first monitoring area by a straight line detection method using Hough transform. 26 to search for an edge 41. When the edge 41 is not detected, the first detection unit 15 determines that the door 3 is in a closed state.

  When the door 3 is opened, an edge 41 of the door 31 appears in the first monitoring area 26. The first detection unit 15 detects that the door 3 is open by detecting the edge 41. The first detection unit 15 can also detect the degree of opening of the door 3 based on the position of the edge 41 in the first monitoring area 26.

  Thus, the 1st detection part 15 detects the door 3 of the open state by the edge 41 being detected. Further, the first detection unit 15 detects that the door 3 is closed when the detected edge 41 is not detected.

  Here, referring to FIG. 3, a timer 18 is connected to the first detection unit 15. The first detection unit 15 measures the elapsed time using the timer 18 from the time when the detected edge 41 is no longer detected. Then, when the state in which the edge 41 is not detected continues for a predetermined time (for example, several seconds) or more, it is detected that the door 3 is closed. On the other hand, after the detected edge 41 is no longer detected, when the edge 41 is detected again before the predetermined time elapses, the first detection unit 15 closes the door 3. Judge that it is not.

  As a result, the door 3 is closed even when the first detection unit 15 cannot instantaneously detect the door 3 that is actually opened due to light adjustment or noise. It can be avoided that the first detection unit 15 detects it erroneously. As a result, it is possible to accurately determine the passer-by-passer determination by the determination unit 19 described later.

<Processing by Second Detection Unit 16>
FIG. 8 is a diagram for explaining the contents of the processing by the second detection unit 16 shown in FIG. The second detection unit 16 detects a person who moves in the vicinity area 5 (see FIGS. 1 and 2) of the entrance 2 based on the video imaged by the camera 1. This will be specifically described below.

  As shown in FIG. 3, the difference image D <b> 3 generated by the difference image generation unit 12 is input to the second detection unit 16. In the example shown in FIG. 8, the edge of the person 30 corresponding to the person 20 shown in FIG. 5 appears in the difference image D3. Thus, if the person 20 exists in the vicinity area | region 5, the edge of the person 30 will appear in the difference image D3.

  Referring to FIG. 8, the second detection unit 16 monitors the entire image area 25 of the difference image D3, and uses a circle (true circle and ellipse) detection method using the Hough transform to detect a circle in the difference image D3. Search for edges. When the circular edge is not detected, the second detection unit 16 determines that no person is present in the vicinity region 5.

  If the person 20 exists in the vicinity area | region 5, the edge of the person 30 will appear in the difference image D3. In the example illustrated in FIG. 8, the second detection unit 16 detects a perfect circle 50 corresponding to the head of the person 30 and an ellipse 51 corresponding to the torso of the person 30.

  Next, the second detection unit 16 defines one circle 52 including the perfect circle 50 and the ellipse 51. The circle 52 defined in this way is handled as a circle corresponding to the person 20 (person 30). As shown in FIG. 8, the inner circumference of the circle 52 is in contact with the outer circumferences of the perfect circle 50 and the ellipse 51. In the example shown in FIG. 8, one circle 52 is defined including the perfect circle 50 and the ellipse 51 that overlap each other, but when there are a plurality of circles that are close to each other without overlapping, A plurality of these circles are also included to define one circle 52. For example, the second detection unit 16 obtains the maximum radius among a plurality of radii of the plurality of circles, includes a plurality of circles having a separation distance shorter than the maximum radius as one group, and includes one circle 52. Is specified.

  Next, the second detection unit 16 obtains the center position C and the radius R of the circle 52 defined as described above, holds these pieces of information as position information regarding the person 20, and moves the person 20. Use for tracking.

  FIG. 9 is a diagram for explaining a first example of tracking processing of person movement by the second detection unit 16. Referring to FIGS. 9A and 9B, a circle 52a is a circle 52 detected in the first frame, and a circle 52b is detected in a second frame after the first frame. A circle 52. That is, in the first frame, the circle 52a having the center position Ca and the radius Ra is detected, and in the second frame, the circle 52b having the center position Cb and the radius Rb is detected. The center position Cb is separated from the center position Ca by a distance L. The second frame may be a frame that is continuous with the first frame, or may be a frame that is several frames after the first frame.

  With reference to FIG. 9A, the distance L is equal to or less than the radius Ra. In this case, the second detection unit 16 determines that the circle 52b is a circle corresponding to the same person as the person corresponding to the circle 52a. That is, the second detection unit 16 determines that the person corresponding to the circle 52a and the person corresponding to the circle 52b are the same person.

  With reference to FIG. 9B, the distance L is larger than the radius Ra. In this case, the second detection unit 16 determines that the circle 52b is a circle corresponding to a person different from the person corresponding to the circle 52a. That is, the second detection unit 16 determines that the person corresponding to the circle 52a and the person corresponding to the circle 52b are different persons.

  As described above, when the circle 52 moves in the image captured by the camera 1, the second detection unit 16 determines whether the movement before and after the movement based on the center positions Ca and Cb of the circles 52a and 52b and the radius Ra of the circle 52a. It is determined whether or not the circles 52a and 52b are circles corresponding to the same person. The size of the circle 52 corresponding to the person 20 varies depending on the installation position, the installation angle, and the like of the camera 1, but not only the center position C of the circle 52 but also the radius R can be used. Tracking can be performed more accurately.

  FIG. 10 is a diagram for explaining a second example of the person movement tracking process by the second detection unit 16. Here, for example, a situation is assumed in which the person 20 moving in the vicinity area 5 of the doorway 2 makes a sudden movement, and the amount of movement of the head of the person 20 increases momentarily. FIG. 10A shows an example in which the movement of the person is tracked using the circle 52, while FIG. 10B shows the movement of the person using only the circle 50 corresponding to the head of the person. An example of tracking is shown.

  Referring to FIG. 10B, circle 50a is circle 50 detected in the first frame, and circle 50b is circle 50 detected in the second frame. That is, the circle 50a is detected in the first frame, and the circle 50b is detected in the second frame. Similarly to the above, the second frame may be a frame continuous with the first frame, or may be a frame several frames after the first frame.

  In the example shown in FIG. 10B, the distance L is larger than the radius Ra. Therefore, in this example, the second detection unit 16 erroneously determines that the circle 50b is a circle corresponding to a person different from the person corresponding to the circle 50a.

  On the other hand, referring to FIG. 10A, a circle 52a is a circle 52 detected in the first frame, and a circle 52b is a circle 52 detected in the second frame. That is, the circle 52a is detected in the first frame, and the circle 52b is detected in the second frame.

  In the example shown in FIG. 10A, the distance L is not more than the radius Ra. Therefore, in this example, the second detection unit 16 correctly determines that the circle 52b is a circle corresponding to the same person as the person corresponding to the circle 52a.

  As described above, when the second detection unit 16 according to the present embodiment detects a plurality of circles that overlap or are close to each other in the difference image D3, the second detection unit 16 defines one circle that includes the plurality of circles, and One circle is detected as one person. Therefore, as shown in FIG. 8, when a plurality of circles 50 and 51 such as the head and shoulders are detected for the same person, one circle 52 including the plurality of circles 50 and 51 is obtained. Detected as one person. As a result, as shown in FIG. 10 (A), for example, the person 20 moving in the vicinity area 5 of the doorway 2 makes a sudden movement, and the movement amount of the head of the person 20 increases momentarily. Even in this case, the person 20 can be accurately tracked as being the same person.

<Processing by Second Setting Unit 17>
FIG. 11 is a diagram for explaining the contents of processing by the second setting unit 17 shown in FIG. 3. The second setting unit 17 specifies the entrance / exit 2 in the image captured by the camera 1 based on the video imaged by the camera 1. And the area | region containing the specified entrance / exit 2 is set as a 2nd monitoring area | region which the determination part 19 should monitor in order to detect the person 20 who passes the entrance / exit 2. FIG. This will be specifically described below.

  As illustrated in FIG. 3, the difference image D <b> 3 generated by the difference image generation unit 12 is input to the second setting unit 17. Referring to FIG. 11, when the door 3 is opened, the edge 41 of the door 31 appears in the difference image D3. An edge 41 is detected in the image D3. Moreover, the 1st detection part 15 detects the position of the edge 42 corresponding to the door 3 of the closed state as a position of the edge 41 in the moment detected for the first time.

  The second setting unit 17 receives the position information of the edges 41 and 42 from the first detection unit 15. A line segment 43 connecting one end of the edge 41 and one end of the edge 42 is defined, and a triangular area surrounded by the edges 41 and 42 and the line segment 43 is set as the second monitoring area 44. As a result, the second monitoring area 44 is set as an area including the entrance 2 itself in the difference image D3.

  The second monitoring area 44 is an area to be monitored by the determination unit 19 in order to detect the person 20 passing through the doorway 2, and the position information D5 of the second monitoring area 44 is obtained by the second setting unit 17. Retained. Referring to FIG. 3, position information D <b> 5 is input from second setting unit 17 to determination unit 19.

  As described above, the second setting unit 17 specifies the entrance / exit 2 in the image captured by the camera 1 and determines a region including the specified entrance / exit 2 itself to detect the person 20 passing through the entrance / exit 2. 19 is set as the second monitoring area 44 to be monitored. Therefore, the determination unit 19 only needs to monitor the second monitoring area 44, and does not need to monitor a wider area, so that the processing load can be reduced. In addition, since an intruder who tries to enter the restricted access area always passes through the entrance 2, the presence of the intruder can be reliably detected by setting the entrance 2 as the second monitoring area 44. Become.

  Further, the second setting unit 17 includes the position of the door 3 in the closed state (edge 42) and the position of the door 3 detected by the first detection unit 15 in the opened state (edge 41). ) And the gateway 2 is specified. Therefore, the second setting unit 17 can appropriately set the second monitoring area 44 according to the opening degree of the door 3 at the current time.

<Processing by the determination unit 19>
12-14 is a figure for demonstrating the content of the process by the determination part 19 shown in FIG. 3, respectively. In these figures, time advances in the order of (A) → (B) →... → (F). With reference to FIG. 3, the determination unit 19 determines whether the door 3 is open or closed based on the detection result D6 of the open / closed state of the door 3 by the first detection unit 15 and the detection result D7 of the moving person by the second detection unit 16. It is determined whether one or more persons have passed through the doorway 2 between the opening and closing of the. This will be specifically described below.

  Referring to FIG. 3, detection result D6 of the open / closed state of door 3 is input from first detection unit 15 to determination unit 19. Specifically, information notifying that the door 3 has been opened and information notifying that the door 3 has been closed next are input from the first detection unit 15 to the determination unit 19 as the detection result D6. The

  In addition, the detection result D <b> 7 of the moving person is input from the second detection unit 16 to the determination unit 19. Specifically, position information of one or more circles 52 (see FIG. 8) detected by the second detection unit 16 at each time is input from the second detection unit 16 to the determination unit 19 as a detection result D7. Is done.

  Further, the position information D5 of the second monitoring area 44 (see FIG. 11) at each time is input from the second setting unit 17 to the determination unit 19.

  Based on the input information, the determination unit 19 has one or a plurality of circles 52 that cross the second monitoring area 44 between the opening and the next closing of the door 3. It is determined whether there is one person or a plurality of persons who have passed through the doorway 2.

  With reference to FIGS. 5 and 12, at the time of FIG. 12A, the second monitoring region 44 is set in the difference image D <b> 3 as the door 3 starts to be opened by the person 20. In addition, a circle 52 corresponding to the person 20 is detected. At this time, the person 20 stands in front of the door 3, and the circle 52 and the second monitoring area 44 are also in the difference image D3. Do not overlap each other.

  At the time of FIG. 12B, the area of the second monitoring region 44 is increased as the door 3 is greatly opened by the person 20. However, even at this time, the circle 52 and the second monitoring area 44 do not overlap each other.

  At the time of FIG. 12C, as the person 20 moves and approaches the entrance 2, the circle 52 moves to the right side, and a part of the circle 52 overlaps the second monitoring area 44.

  At the time of FIG. 12D, as the person 20 further moves and a part of the body enters the entrance 2, the circle 52 further moves to the right side, and a part of the circle 52 is the second The monitoring area 44 is crossed.

  At the time of FIG. 12 (E), the area of the second monitoring region 44 is reduced as the door 3 starts to be closed by the person 20.

  At the time of FIG. 12 (F), as the person 20 passes through the entrance 2 and closes the door 3, the circle 52 is no longer detected across the second monitoring area 44, and the difference image D3. The second monitoring area 44 has also disappeared.

  In the example shown in FIG. 12, a circle crossing the second monitoring area 44 from the moment when the door 3 is opened (FIG. 12A) to the next moment when the door 3 is closed (FIG. 12F). Is one. Accordingly, the determination unit 19 determines that one person has passed through the entrance 2. In this case, emergency call processing such as sounding an alarm is not performed.

  On the other hand, in the example shown in FIG. 13, after the circle 52 corresponding to the permitter crosses the second monitoring area 44, the circle 60 corresponding to the intruder crosses the second monitoring area 44. Therefore, in the example shown in FIG. 13, the second monitoring area 44 is crossed between the moment when the door 3 is opened (FIG. 13A) and the next moment when it is closed (FIG. 13F). The number of circles is two. Therefore, the determination unit 19 determines that there are a plurality of persons who have passed through the entrance 2. In this case, emergency call processing such as sounding an alarm is performed.

  In the example illustrated in FIG. 14, in addition to the circle 52, a circle 70 corresponding to a passerby who passes through the neighborhood region 5 is detected. However, the circle 70 does not cross the second monitoring area 44. Therefore, in the example shown in FIG. 14, the second monitoring area 44 is crossed between the moment when the door 3 is opened (FIG. 14A) and the moment when the door 3 is next closed (FIG. 14F). The number of circles is one. Accordingly, the determination unit 19 determines that one person has passed through the entrance 2. In this case, emergency call processing such as sounding an alarm is not performed.

  Thus, according to the traffic monitoring system according to the present embodiment, the first detector 15 detects the open / closed state of the door 3 and the second detector 16 moves in the vicinity area 5 of the entrance 2. 20, the determination unit 19 determines whether a person who has passed through the doorway 2 between the opening and closing of the door 3 based on the detection results of the first detection unit 15 and the second detection unit 16. Determine whether one or more. Therefore, it is possible to reliably monitor the passage of the non-permitted person through the doorway (so-called companionship) due to the passage by the permitted person.

  In the above explanation, an example was given of a situation in which a non-permitted person enters the room after the permitted person entering the restricted access area (so-called follow-up), but it is not permitted to pass by a permitted person leaving the restricted access area. Even when the person enters the room (so-called passing), the traffic monitoring system according to the present embodiment can detect the intrusion of the unauthorized person.

<Modification 1>
FIG. 15 is a side view showing an environment in which a traffic monitoring system according to a modification of the present embodiment is introduced, corresponding to FIG. In FIG. 1, the camera 1 is installed on the ceiling above the nearby region 5, but the camera 1 may be installed on the wall above the door 3 as shown in FIG. 15.

  The camera 1 takes a picture of the vicinity area 5 from obliquely above the front of the door 3. Therefore, the face of the person walking toward the door 3 can be photographed from the front by the camera 1, and the security can be improved.

  A mirror 80 is installed on the ceiling above the neighborhood area 5. The mirror 80 is disposed within the shooting angle of view of the camera 1 (see a thick broken line). A mirror 80 is reflected in the video photographed by the camera 1, and the state of the neighborhood region 5 viewed from above is reflected on the mirror 80. Therefore, the same effect as described above can be obtained by executing the processing by the image processing unit 10 (see FIG. 3) on the video displayed on the mirror 80.

<Modification 2>
In the above description, the first detection unit 15 detects the open / closed state of the door 3 based on the video imaged by the camera 1. Not only this but the sensor for exclusive use for detecting the opening-and-closing state of the door 3 is installed in the door 3 or the entrance / exit 2, and the 1st detection part 15 opens and closes the door 3 based on the detection signal from the sensor. A state may be detected. In this case, the second monitoring area 44 is set as the entire image area 25 of the difference image D3, for example.

It is a side view showing the environment where the traffic monitoring system concerning an embodiment of the invention was introduced. It is a top view which shows the environment where the traffic monitoring system which concerns on embodiment of this invention was introduced. It is a block diagram which shows the function structure of the image process part in the traffic monitoring system which concerns on embodiment of this invention. It is a figure which shows a background image. It is a figure which shows a picked-up image. It is a figure which shows a difference image. It is a figure for demonstrating the content of the process by the 1st setting part shown in FIG. It is a figure for demonstrating the content of the process by the 2nd detection part shown in FIG. It is a figure for demonstrating the 1st example of the tracking process of a person's movement by a 2nd detection part. It is a figure for demonstrating the 2nd example of the tracking process of a person's movement by a 2nd detection part. It is a figure for demonstrating the content of the process by the 2nd setting part shown in FIG. It is a figure for demonstrating the content of the process by the determination part shown in FIG. It is a figure for demonstrating the content of the process by the determination part shown in FIG. It is a figure for demonstrating the content of the process by the determination part shown in FIG. It is a side view which shows the environment where the traffic monitoring system which concerns on the modification of embodiment of this invention was introduced.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Camera 2, 32 Entrance / exit 3, 31 Door 5 Neighborhood area 10 Image processing part 12 Difference image generation part 14 1st setting part 15 1st detection part 16 2nd detection part 17 2nd setting part 18 Timer 19 Determination Part 20, 30 Person 25 Total image area 26 First monitoring area 41, 42 Edge 44 Second monitoring area 52 yen

Claims (4)

A camera for photographing a predetermined entrance / exit provided with a door including a region near the entrance / exit,
A first detector for detecting an open / closed state of the door;
A second detection unit that detects a person moving in a vicinity region of the doorway based on an image captured by the camera;
A determination unit that determines whether one or more persons have passed through the doorway between the opening and closing of the door based on the detection results of the first and second detection units;
Based on the video imaged by the camera, a first monitoring area to be monitored by the first detection unit in order to detect the open / closed state of the door is included in a part of the entire area of the image captured by the camera. A first setting unit to be set;
Based on the video imaged by the camera, the entrance should be identified in an image captured by the camera, and the determination unit should monitor the identified entrance / exit itself to detect a person passing through the entrance / exit A second setting unit for setting as a second monitoring area,
The first detection unit detects an open / closed state of the door based on an image taken by the camera,
The second setting unit identifies the doorway based on the position of the door in the closed state and the position of the door detected by the first detection unit in the opened state. A traffic monitoring system.   In the situation where the first detection unit detects the door in the opened state, the first detection unit detects the door in the opened state when the door in the opened state is not detected. 2. The traffic monitoring system according to claim 1, wherein the door is closed by detecting that the state has continued for a predetermined time from the point of time when the door is not operated. Before Symbol camera is installed above the region near the doorway,
The second detection unit detects the person by detecting a circle in an image captured by the camera,
The second detection unit, when detecting a plurality of circles that overlap or close, to define a single circle encompassing the plurality of circles, to detect the one circle as the person alone, claim 1 Or the traffic monitoring system according to 2; Before Symbol camera is installed above the region near the doorway,
The second detection unit detects the person by detecting a circle in an image captured by the camera,
When the circle moves in the image captured by the camera, the second detection unit determines whether the circles before and after the movement correspond to the same person based on the center position information and the radius information of the circle. The traffic monitoring system according to claim 1 or 2 , which determines whether or not.

Images