JP2018181114A - Intrusion monitoring method, intrusion monitoring program, and intrusion monitoring system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an intrusion monitoring method capable of detecting a person or object, who or which intrudes into an object-of-monitoring range, and reliably giving an intrusion alarm.SOLUTION: An intrusion monitoring method includes the steps of defining an object-of-monitoring range 201, within which an alarm action is performed when an object intrudes into the range, within a range within which an object detection sensor detects an object, and defining an object recognition range 202, within which the fact that a moving object has entered the range is recognized, around the object-of-monitoring range 201 (S10), identifying an object supposing the object enters the object recognition range 202 (S11), and outputting an alarm signal when the identified object enters the object-of-monitoring range 201 (S13).SELECTED DRAWING: Figure 8

Description

  The present invention relates to an intrusion monitoring method, an intrusion monitoring program, and an intrusion monitoring apparatus.

  Conventionally, as an intrusion monitoring technology, there is a technology in which an intrusion detection sensor is provided separately from the monitoring camera. In this technology, a plurality of intrusion detection sensors are provided in a region where intrusion is to be detected. Then, based on the intrusion detection information from the plurality of intrusion detection sensors, the traveling direction of the intruder is predicted, and when intrusion is detected by the intrusion detection sensor in the predicted traveling direction, the camera is Start recording video information. As a result, it is possible to record only the video information according to the traveling direction of the intruder without always recording the video information of the camera. Therefore, the storage capacity is reduced, and it is possible to efficiently search for video information in which the intruder's action is recorded (Patent Document 1).

JP, 2007-142742, A

  In the prior art, a plurality of intrusion detection sensors monitor. For this reason, when there are a large number of people and objects that can be intruders in the monitoring target range, the large number of intrusion detection sensors react one after another, and the process can not catch up with it.

  Therefore, an object of the present invention is to provide an intrusion monitoring method capable of detecting an object intruding into a monitoring target range and reliably issuing an intrusion alert.

  Another object of the present invention is to provide an intrusion monitoring program capable of detecting an object intruding into a monitored area and reliably issuing an intrusion alert.

  Furthermore, another object of the present invention is to provide an intrusion monitoring device capable of detecting an object intruding into a monitored area and reliably emitting an intrusion alert.

  The above object is achieved by the following means.

(1) Within a range where an object detection sensor detects an object, a monitoring target range is set to perform an alarm operation when an object intrudes, and recognition that a moving object has entered around the monitoring target range Setting a target object recognition range (a);
If there is an object within the object recognition range, a step (b) of recognizing the object;
Performing an alarm operation when part or all of the recognized object intrudes into the monitored area (c);
Intrusion monitoring method.

(2) An interval from the outer periphery of the monitoring target range corresponding to the type of the object to the outer periphery of the object recognition range is determined in advance and stored in the computer,
The interval corresponding to the type of the object selected by the user is read from the storage in the computer, and the object recognition range is set around the monitoring target range at the read interval. Intrusion monitoring method described.

  (3) The object recognition range described in the above (1) or (2), wherein, when there are a plurality of monitoring target ranges, one object recognition range is set to surround all the plurality of monitoring target ranges. Intrusion monitoring methods.

  (4) The intrusion monitoring method according to (1) or (2), wherein the object recognition range sets the object recognition range for each of the monitoring target ranges when there are a plurality of monitoring target ranges.

(5) storing, in the computer, sorting criteria for sorting whether or not an object that has entered the object recognition range is an object that should be alerted;
Said step (b)
Recognizing an object that has entered the object recognition range (b1);
Determining whether or not the recognized object is an object to be alarmed by comparing it with the sorting criterion, and if it is an object to be alarmed, selecting (b2) as an object to be alarmed; Have
The step (c) is
The intrusion monitoring method according to any one of the above (1) to (4), wherein an alarm operation is performed when an object selected to perform an alarm operation intrudes into the monitoring target range.

  (6) An intrusion monitoring program that causes a computer to execute the intrusion monitoring method according to any one of (1) to (5) above.

(7) an object detection sensor that detects an object;
The object detection sensor is set within a range for detecting an object, and it is set in a monitoring target range in which an alarm operation is performed when an object enters, and is recognized around the monitoring target range to recognize that a moving object enters. A storage unit that stores an object recognition range;
A control unit that recognizes an object that has entered the object recognition range, and outputs an alarm signal when the recognized object intrudes into the monitoring target range;
Intrusion monitoring device.

  According to the present invention, an object recognition range for recognizing an intruding object (including a person) is provided around a monitoring target range in which an alarm operation is performed. As a result, it is possible to recognize a person or a thing in the object recognition range and to immediately warn when the person or the thing intrudes into the monitoring target range. On the other hand, since recognition is not performed outside the object recognition range, the load on the recognition processing can be reduced to prevent alarm omission, and an intrusion alarm to the monitoring target range can be reliably issued.

It is a block diagram explaining composition of a surveillance system. It is a functional block diagram for explaining functional composition of a control device. It is explanatory drawing which shows a monitoring object range, an object recognition range, and an intruder while showing a distance image example. It is explanatory drawing explaining how to determine the space | interval from the outer periphery of the monitoring object range to the outer periphery of an object recognition range. It is explanatory drawing explaining selection of an object recognition range, and the example of a setting screen. It is explanatory drawing explaining the 1st example in the case of setting multiple monitoring object ranges in one distance image. It is explanatory drawing explaining the 2nd example in the case of setting multiple monitoring object ranges in one distance image. 5 is a flowchart showing the procedure of the intrusion monitoring process of the first embodiment. 7 is a flowchart showing the procedure of the intrusion monitoring process of the second embodiment. It is a sectional view explaining the composition of a rider. It is explanatory drawing explaining the state which scans the inside of the monitoring space of the penetration monitoring apparatus MD with the laser spot beam SB (it shows with hatching) radiate | emitted according to rotation of the mirror unit MU.

  Hereinafter, embodiments of the present invention will be described with reference to the attached drawings. The present invention is not limited to the following embodiments. In the description of the drawings, the same elements will be denoted by the same reference symbols, without redundant description. Also, the dimensional proportions of the drawings are exaggerated for the convenience of the description, and may differ from the actual proportions. Further, in the following description, in the frames arranged in time series, the current frame is referred to as a current frame, a frame before the current frame is referred to as a past frame, and a frame one before is referred to as a previous frame.

(Embodiment 1)
FIG. 1 is a block diagram for explaining the configuration of the intrusion monitoring apparatus.

  In the intrusion monitoring device 100 according to the first embodiment, a rider (LiDAR (Light Detection And Ranging)) 20, which is an object detection sensor, and a control device 10 for recognizing and alerting an object from a distance image acquired by the rider 20 are integrated. Device (system).

  The details of the rider 20 will be described later, but generally speaking, the laser (light) is scanned to measure the distance from the reflected light to the object in real time. For this reason, the distance from the installation position of the rider 20 to an object (an object such as a person or an object) and the size and shape of the object can be obtained. Thereby, an image called a distance image can be obtained. The distance images are arranged in chronological order as one scan and one frame (displayed continuously) to become a moving image.

  The control device 10 is a computer. The configuration of the control device 10 includes a central processing unit (CPU) 11, a random access memory (RAM) 12, a read only memory (ROM) 13, a hard disk drive (HDD) 14, and the like.

  The CPU 11 serves as a control unit, reads a program corresponding to the processing content from the HDD 14, develops the program in the RAM 12, and executes the developed program. Here, the operation of the rider 20 is mainly controlled, and recognition of an object, output of an alarm signal (instruction of alarm operation) and the like are performed by a processing procedure described later.

  The ROM 13 stores programs necessary for the basic operation of the control device 10 as a computer.

  The HDD 14 is a storage unit together with the RAM 12 and stores programs and data necessary for each process. The HDD 14 may be replaced by a non-volatile semiconductor memory (for example, a flash memory).

  In addition, the control device 10 includes a touch panel (also used as a display) and buttons, an input device 15 such as a mouse, a display 16 for displaying an image (moving image), etc., and further, for example, a server 30 and other external devices (not shown) A network interface 17 for connecting to (not shown) or the like is provided.

  The system also includes an alarm device 21. The alarm device 21 issues an alarm in a manner that can be recognized by a person, such as sound, light such as flash light or a rotating light, for example. In place of (or together with) such an alarm, other actions may be triggered. Other actions include, for example, start of recording of the distance image 200 and stop of an automatic machine (a robot, a carriage, various machines, etc.) operating in relation to the monitoring target range 201. For example, when recording is performed, the server 30 connected via the network interface 17 is recorded. By starting recording at this stage, it is possible to reliably record an intruding object, but when there is no intruding object, it is not necessary to perform recording, so it is possible to suppress the usage of recording media. A dedicated recording device may be connected to the control device 10 in addition to the server 30 for recording. In addition, by stopping the automatic machine, it is also possible to use it for the safe operation of the automatic machine. These sound and light alarms and other actions are collectively called alarm operation.

  In the first embodiment, an example of the device in which the rider 20, the control device 10, and the alarm device 21 are integrated is shown. However, the present invention is not limited to this, and the rider 20, the control device 10, and the alarm device 21 may be devices or systems in a form in which each or a part is separated and connected via a network or a dedicated line. In this case, the rider 20 only has a processing circuit required to operate the rider 20, and the control device 10 performs commands for the recognition of the object and the alarm operation separately. In the case of such a separation mode, the control device 10 is not limited to a dedicated computer, and may be a general-purpose personal computer or the like.

  Moreover, although the example using the rider 20 as an object detection sensor was shown in this Embodiment 1, it replaces with this and if it can recognize an object and its position within the field of view of a sensor, it will not be limited in particular. For example, a movie camera can be used.

  FIG. 2 is a functional block diagram for explaining a functional configuration of the control device. FIG. 3 is an explanatory view showing an example of a distance image and explaining a monitoring target range, an object recognition range, and an intruder.

  The functional configuration of the control device 10 includes a monitoring target range management unit 111, an object recognition range management unit 112, an object recognition processing unit 113, and an intrusion detection processing unit 114.

  The function of each part will be described with reference to the example of the distance image shown in FIG. In the example of the distance image in FIG. 3, the dotted line indicates a standard of the distance in the depth direction (z direction) with reference to the position (0 m) of the sensor. Further, arrows of x, y and z indicate the directions of the three-dimensional coordinate system in this image.

  The distance image 200 is the entire range in which the rider 20 can detect an object, as shown in FIG. A monitoring target range 201 that issues an alarm when an object (including a part of an object, a person and a part of a person, etc.) enters in the distance image 200 is set, and the monitoring target range 201 is set. An object recognition range 202 that recognizes a moving object (including an object and a person, and so on) is set around the outside (except for the bottom surface such as the ground or floor).

  The monitoring target range management unit 111 performs setting, storage, and the like of the monitoring target range 201. Further, the object recognition range management unit 112 performs setting, storage, and the like of the object recognition range 202. Details of the method of setting the monitoring target range 201 and the object recognition range 202 will be described later.

  The object recognition processing unit 113 determines whether an object moving into the object recognition range 202 has entered, and if it is determined that an object has entered, it recognizes the object.

  The determination as to whether or not an object has entered compares the current frame with the past frame from the frames arranged in chronological order. Then, if there is an object which is not in the past frame but partially in the object recognition range 202 of the current frame, it is determined that the object is in the object recognition range 202. Therefore, the range to be compared may be only within the object recognition range 202, and it is not necessary to compare within the entire frame.

  The past frame to be compared with the current frame may be a frame immediately before the current frame (referred to as a previous frame), or may be one of any preset past frames. In the case of any past frame, it is preferable to designate a frame that has no moving object in the frame. Such previous frame or any past frame is stored in the RAM 12 as a reference frame to be compared with the current frame. When the previous frame is used as the reference frame, the stored previous frame is always updated.

  As described above, by calculating whether or not an object appears only in the predetermined object recognition range 202, the processing speed is faster than when the object is recognized on the entire frame.

  Referring to FIG. 3, for example, it is assumed that a person 301 is in a past frame. The person 301 is not recognized at this point. When the person 301 moves and enters the object recognition range 202 in the current frame, the person 302 is recognized. At this time, it is not necessary to recognize whether the object in the object recognition range 202 is a person or not, and it may be recognized that some object has moved and entered the object recognition range 202. As recognition of an object, for example, clustering can be performed. In clustering, pixels constituting an image that is an object partially included in the object recognition range 202 are set as a clustering target. Therefore, the range to be subjected to the clustering process can be limited within the object recognition range 202. When clustering is performed, for example, when the object is a human, the pixels constituting the shape may be finely extracted, but in the case of three dimensions, clustering may be performed as a rectangular solid surrounding the whole human as an image. In FIG. 3, it is a line surrounding the persons 302-303. Among them, it is the person 302 who partially enters the object recognition range 202 to cluster. On the other hand, it is not necessary to perform clustering on the persons 301 not in the object recognition range 202. The same applies to a two-dimensional image obtained by a movie camera or the like, and an object such as a person may be surrounded by a rectangle or the like for clustering.

  In this way, by clustering with a rectangular parallelepiped (a square in the case of two dimensions. The same applies to the following), when tracking the movement of this object between frames, it is necessary to pay attention to fine movements such as human hands and feet. Processing is simplified because it is sufficient to track the movement of the rectangular parallelepiped.

  In addition, while the object recognition processing unit 113 recognizes an object in the object recognition range 202, the range or size (the size of a rectangular solid) to be clustered may be changed according to the movement of the object. For example, in the case of a person, the range or size of clustering may be changed according to a change such as walking or swinging an arm.

  The intrusion detection processing unit 114 determines whether or not the object recognized by the object recognition processing unit 113 has entered the monitoring target range 201, and outputs an alarm signal to the alarm device 21 when there is an intrusion. Thus, the alarm device 21 issues an alarm.

  At this time, the intrusion detection processing unit 114 does not have to recognize the size and shape of the object, and determines only whether the object recognized by the object recognition processing unit 113 has moved and entered the monitoring target range 201. do it. That is, when an object recognized by the object recognition processing unit 113 in the past frame is detected in the monitoring target range 201 in the current frame, an alarm operation is instructed. For example, in the case where the clustering is a rectangular solid surrounding an object, an alarm operation is performed when a part of the rectangular solid enters the monitoring target range 201. At this time, the intrusion detection processing unit 114 does not have to change the clustering range or size in accordance with the movement of the object. Thereby, the intrusion detection processing unit 114 can perform an alarm operation at the moment when it enters a part of a human body, for example, a hand, a foot, or the monitoring target range 201. Moreover, since only the movement of a cuboid (clustered object) is tracked, the processing is simplified and the calculation time is shortened compared with the tracking of fine movements such as human hands and feet, and the alarm operation is performed faster. It can be performed.

  Of course, instead of tracking the area surrounding such an object as a rectangular solid, tracking may be performed focusing on fine movements such as human hands and feet. Even in that case, since the object recognition processing unit 113 tracks a part or all of the object recognized once, the alarm operation can be performed promptly.

  Referring to FIG. 3, for example, when the recognized person 302 moves and the person 303 enters the monitored range 201, it is detected and an alarm is issued. In FIG. 3, persons 301, 302, and 303 indicate the same intruder at different times (different frames).

  A method of setting the monitoring target range 201 and the object recognition range 202 will be described. The setting method of the monitoring target range 201 and the object recognition range 202 by the monitoring target range management unit 111 and the object recognition range management unit 112 is basically the same, and is performed in the distance image 200. For example, a three-dimensional coordinate system is set in the distance image 200, and each range is set by coordinates in the horizontal (x), vertical (y), and depth (z) directions of the screen.

  For example, each set range is always stored in the RAM 12 during operation and immediately made available to the CPU 11. Further, it is preferable that each set range be stored in the HDD 14 and read out from the HDD 14 to the RAM 12 after restart after the power is turned off.

  The setting of the monitoring target range 201 is performed by the user. The monitoring target range management unit 111 sets the monitoring target range 201 at the coordinates designated by the user in the distance image 200. For setting the coordinates, for example, the user designates eight points that are corners of the monitoring range 201 in the horizontal (X), vertical (Y), and depth (Z) directions (see FIG. 3) of the three-dimensional coordinate system. . The monitoring target range management unit 111 connects each designated coordinate to a rectangular parallelepiped (two-dimensional quadrilateral) to set the monitoring target range 201. The set monitoring target range 201 is stored by the monitoring target range management unit 111 (stored in the HDD 14 and read out to the RAM 12). The specification of such a range is merely an example, and in addition, for example, a line serving as a side of a rectangular parallelepiped (a square in two dimensions) serving as the monitoring target range 201 is drawn, and the coordinates of the line are stored, etc. May be.

  Next, the size of the object recognition range 202 may be arbitrarily set by the user. Further, a plurality of intervals from the outer periphery of the monitoring target range 201 to the outer periphery of the object recognition range 202 may be determined in advance, and the user may select one of them.

  When the user sets the object recognition range 202 arbitrarily, it is the same as the setting of the monitoring target range 201 described above, and is performed by specifying the three-dimensional coordinates set in the distance image 200, and the object recognition range management unit 112 (Stored in the HDD 14 and read out to the RAM 12).

  On the other hand, when preparing a plurality of intervals from the outer periphery of the monitoring target range 201 to the outer periphery of the object recognition range 202, for example, in the case of only people (for example, pedestrians), when people, bicycles, cars, etc. are mixed, For example, in the case of a car only, it is determined according to the type of object, and stored in the computer (stored in the HDD 14 and read out in the RAM 12).

  FIG. 4 is an explanatory diagram for explaining how to determine the distance from the outer periphery of the monitoring target range to the outer periphery of the object recognition range. This figure shows the distance w from the outer periphery of the monitoring target range 201 to the outer periphery of the object recognition range 202. In addition, in FIG. 4, a part of each range was shown. The interval w is the size of the object recognition range 202, and is the same around the monitoring target range 201, that is, in the x, y, and z directions.

  The distance w corresponding to the type of object may be determined, for example, by the moving speed of the object and the speed required for object recognition. The moving speed of the object differs depending on the object to be recognized. On the other hand, the speed required to recognize an object is the time taken to recognize an object in the distance image 200. Although the speed required for object recognition varies depending on the control device (computer), it is, for example, 1 second, 0.5 seconds, and the like. Incidentally, the time taken to simply display the distance image 200 acquired by the rider 20 is an inter-frame time, which is 1/10 second if display of 10 frames per second is possible.

  For example, when monitoring a person, the moving speed of the person (pedestrian) is 4 km / h = about 1 m / s. The size of the object recognition range 202 is set based on this velocity. Therefore, looking at the margin (assuming a person who runs fast etc.) and assuming 1.5 m / s, if the speed for object recognition is 1 second, the interval w is 1.5 m from the end of the monitoring range 201 Do. If the speed for object recognition is 0.5 seconds, the interval w is set to 0.75 m from the end of the monitoring target range 201.

  Further, for example, in the case of a bicycle, if the speed of the bicycle is 20 km / h = 5.6 m / s, the interval w is set to 5.6 m from the end of the monitoring target area 201 if the speed for object recognition is 1 second. If the speed required for object recognition is 0.5 seconds, the interval w is 2.8 m from the end of the monitoring range 201.

  Also, for example, in a car, if the speed of the car is 40 km / h = 11 m / s, the interval w is 11 m from the end of the monitoring range 201 if the speed for object recognition is 1 second. If the speed required for object recognition is 0.5 seconds, the interval w is set to 5.5 m from the end of the monitoring target range 201.

  Such a predetermined interval w is stored in the object recognition range management unit 112 (stored in the HDD 14 and read out to the RAM 12). Then, the user is allowed to select one of a person, a bicycle, and a car as a monitoring target.

  FIG. 5 is an explanatory view for explaining an example of selection of an object recognition range and a setting screen. For example, the object recognition range management unit 112 displays on the display 16 a selection candidate such as a person, a bicycle, and a car as shown in FIG. 5, radio buttons for selecting them, setting contents, etc. Is displayed. In addition, in FIG. 5, the case where the speed concerning object recognition is 1 second as a setting content is shown.

  The object recognition range management unit 112 reads (checks) a radio button corresponding to a selection candidate such as a person, a bicycle, or an automobile, reads the corresponding interval w from the HDD 14, and matches it with the monitoring target range 201. Set. The object recognition range management unit 112 temporarily stores the set object recognition range 202 in the RAM 12 and further stores it in the HDD 14.

  The above-described speed and recognition speed are merely examples, and the present invention is not limited to these values.

  By the way, there are also cases where the monitoring target range 201 is not limited to one. Therefore, in the first embodiment, a plurality of monitoring target ranges 201 can be set in one distance image 200.

  FIG. 6 is an explanatory diagram for explaining a first example in the case where a plurality of monitoring target ranges are set in one distance image.

  In the example shown in FIG. 6, two monitoring target ranges 201 are provided in one distance image 200. Each monitoring target range 201 is set in the distance image 200 by the user.

  On the other hand, the object recognition range 202 is one range so as to include two monitoring target ranges 201. In this case, such a form may be inevitably obtained by selecting the object recognition range 202 at the predetermined interval w described above. In addition, after the predetermined interval w is selected, if two object recognition areas 202 are temporarily provided, the user may join them. Of course, the user may arbitrarily set the object recognition range 202 from the beginning to have such a form. Further, the monitoring target range 201 is not limited to two, and may be more. Also in this case, in the first example, one object recognition range 202 is set to include a plurality of monitoring target ranges 201.

  FIG. 7 is an explanatory view for explaining a second example in the case where a plurality of monitoring target ranges are set in one distance image.

  In the example shown in FIG. 7, two monitoring target ranges 201 are provided in one distance image 200. Each monitoring target range 201 is set in the distance image 200 by the user.

  On the other hand, an object recognition range 202 is provided around each of two monitoring target ranges 201. In this case, such a form may be inevitably obtained by selecting the object recognition range 202 at the predetermined interval w described above. Of course, the user may set it arbitrarily. Further, the monitoring target range 201 is not limited to two, and may be more. In particular, when there are three or more monitoring target ranges 201, the near ones may be set as the first example, and the far ones may be set as the second example.

  Next, an intrusion monitoring method will be described. FIG. 8 is a flow chart showing the processing procedure of the intrusion monitoring method of the first embodiment.

  The intrusion monitoring process is performed by the control device 10 (specifically, the CPU 11 of the computer constituting the control device 10) executing an intrusion monitoring program created based on the processing procedure shown in FIG.

  First, the control device 10 sets the monitoring target range 201 and the object recognition range 202 according to an input from the user (S10). The processing of S10 is the function of the monitoring target range management unit 111 and the object recognition range management unit 112. After each range is set, the monitoring operation is started.

  The control device 10 determines whether an object has entered the object recognition range 202 (S11). This state loops until an object enters (S11: NO).

  In S11, when the control device 10 determines that an object is included in the object recognition range 202 (S11: YES), the control device 10 recognizes the object (S12). The processes of S11 and S12 are functions of the object recognition processing unit 113.

  Subsequently, it is determined whether the object recognized in S12 intrudes into the monitoring target range 201 (S13). If an object does not intrude into the monitoring target range 201 (S13: NO), the process returns to S11. When S13 becomes NO, there is no intrusion of an object in the monitoring target range 201. On the other hand, it must always be determined whether the object is in the object recognition range 202 or not. Therefore, if S13 is NO, the process returns to S11 and the process is continued. If the object recognition range 202 continues to have an object after returning to S11, S11 becomes YES, so that the object is continuously recognized in S12, and the process directly enters S13. Therefore, as long as there is an object in the object recognition range 202, the state of determining the presence or absence of the object intrusion into the monitoring target range 201 is continued. After returning to S11, if the object is no longer recognized in the object recognition range 202 (that is, if the previously recognized object has left (or is removed)), S11 becomes NO, so the process of S11 is performed as it is. It becomes a loop state.

  In S13, when the control device 10 determines that the object recognized in the object recognition range 202 has intruded into the monitoring target range 201 (S13: YES), the control device 10 outputs an alarm signal to the alarm device 21. Are output (alarm issuance) (S14). Thus, an alarm operation is performed. Thereafter, the process ends. After the end of the process, for example, the alarm operation is continued until the user stops the alarm operation. The processing of S13 to S14 is the function of the intrusion detection processing unit 114.

  It should be noted that automatic return may be performed after the alarm operation. For example, when an object leaves the monitoring target range 201, the process returns to S11. For example, when it is sufficient to simply record an object (intruder) that has invaded into the monitoring target range 201, it is automatically returned (return to S11). Also, the machine is stopped only while the intruder is in the monitoring target area 201, and if it is possible to resume the operation of the stopped machine if it leaves the area, the machine is automatically returned (return to S11). In addition, there are cases where it is possible to return to the original state when the alarm operation is not necessary.

  According to the first embodiment, the following effects can be obtained.

  In the first embodiment, an object recognition range 202 for recognizing an object is provided along with a monitoring target range 201 for performing an alarm operation when an object (including an intruder) intrudes. As described above, the object recognition range is narrowed down to the object recognition range 202 provided around the monitoring target range 201 instead of the wide range where the rider 20 can detect the object. For this reason, the processing load concerning object recognition can be minimized, and an object which can invade can be recognized certainly. Then, when the recognized object intrudes into the monitoring target range 201, the alarm operation can be reliably performed without any delay in processing.

  In addition, by limiting the monitoring target range 201, since the alarm operation is not performed even if an object comes in except the monitoring target range 201, false alarm can be suppressed.

  Further, since the alarm operation is performed only when the object intrudes into the monitoring target area 201, the user can be actively noticed.

  In addition, since the size of the object recognition range 202 can be determined according to the object to be recognized, even when many objects and people enter the object recognition range 202, processing can be performed without delay.

  In addition, the size of the object recognition range 202 is previously stored in correspondence with the type of the object, by determining the distance from the outer periphery of the monitoring target range 201 to the outer periphery of the object recognition range 202. Thus, by causing the user to select an object to be recognized in the object recognition range 202, the object recognition range 202 can be automatically set. Moreover, since the size of the object recognition range 202 is determined in accordance with the type of recognition, the user can set the object recognition range 202 by a simple operation of selecting it.

  Further, even when there are a plurality of monitoring target ranges 201, by providing the object recognition range 202 corresponding to them, even if there are a plurality of monitoring target ranges 201, the alarm operation can be performed reliably without a delay in processing. it can.

  In the case where there are a plurality of monitoring target ranges 201, it is possible to provide one object recognition range 202 surrounding the plurality of monitoring target ranges 201 at one time. By doing this, when there are a plurality of monitoring target ranges 201, it is possible to reliably recognize an object that is not known to be within the monitoring target range 201.

  Conversely, when there are a plurality of monitoring target ranges 201, the object recognition ranges 202 can be provided independently corresponding to them. By doing this, the range of the object recognition range 202 does not increase, and even if the monitoring target range 201 increases, the increase of the object recognition process can be minimized.

Second Embodiment
In the second embodiment, the alarm operation is performed only when a predetermined object intrudes. For this reason, in the second embodiment, as a function of the object recognition processing unit 113, a function of selecting whether an object that has entered the object recognition range 202 is an object to be warned or not is provided. The configuration of the apparatus and the other functions are the same as in the first embodiment, and thus the description thereof is omitted.

  In the object recognition processing unit 113, the object recognition processing unit 113 stores, for example, the size, the shape, etc. in advance as a selection reference of what the object to be warned is like (the selection criterion is the HDD 14). , And read out and stored in the RAM 12 during operation). Then, the size and the shape of the object in the object recognition range 202 are compared with the size and the shape of the stored sorting reference object, and if they are the same, it is determined that the object is an alarm. Of course, it is preferable to set an acceptable range in size and shape when recognizing an object. The allowable range may be, for example, the range of the measurement accuracy of the rider 20, or may be the allowable range with a margin. However, if the margin is too large, it may lead to a false alarm, so it is preferable that the tolerance range be close to the measurement accuracy of the rider 20.

  Although it can be recognized from the distance image 200 by the rider 20 as a three-dimensional size or shape, it may be sorted according to only one of the size and the shape.

  Further, when using an object detection sensor other than the rider 20, sorting is performed according to the characteristics of the sensor. For example, when a color movie camera is used as a sensor, the size and shape of an object in a screen (frame) can be recognized, for example. Although the size changes depending on the location of the image, the change is not so large within the object recognition range 202. Therefore, it is preferable to determine how much the size of the object changes in the object recognition range 202 and store it as the allowable range with respect to the size of the sorting standard determined in advance. The same is true for the form. Also, color can be recognized by using a color movie camera. In the case of using a color, the accuracy of the alarm operation is improved by selecting whether or not the object is an alarm according to the size and the shape instead of the color alone.

  FIG. 9 is a flowchart of the intrusion monitoring process according to the second embodiment. Also in the description of this processing procedure, the same step number (S number) is attached to the step performing the same processing as that of the first embodiment, and the description will be omitted or simplified.

  When each range is set in S10, the control device 10 recognizes that the object is included in the object recognition range 202 (S11: YES), and recognizes the object (S12). Thereafter, in the second embodiment, it is determined (sorted) whether or not the recognized object is an object to be alerted (S21). If the object recognized in S12 is not an alarming object (S21: NO), the process returns to S11.

  On the other hand, if the object recognized in S12 is an object to be alerted in S21 (S21: YES), the process proceeds next, and it is determined whether the object has entered the monitoring target range 201 (S13). The processes after S13 are the same as in the first embodiment. That is, in the second embodiment, only when it is determined that the object that has entered the object recognition range 202 is an object that needs to be alerted in S21, the alert operation is performed if there is an object that has entered the monitoring target range 201. It is done. Therefore, if S21 is not YES, the alarm operation is not performed even if an object intrudes into the monitoring target range 201.

  According to the second embodiment, in addition to the effects of the first embodiment, the following effects can be obtained.

  In the second embodiment, since the object to be alerted is selected, the alarm operation can be reliably performed when the object that needs to be alerted intrudes, but the alarm is triggered even if the object that does not require the alert operation intrudes Do not operate. Therefore, the accuracy of the alarm operation can be increased, and the possibility of false alarm can be reduced.

(rider)
The rider 20 will be described below. The rider 20 is common to the first and second embodiments. The rider 20 is also called a laser radar.

  The rider 20 emits a laser, measures the time until it is reflected back to the object (TOF (Time of Flight)), and determines the distance. By performing this while scanning within a certain range, a three-dimensional distance image in which the distance from the installation position of the rider 20 is known can be obtained (see FIG. 3).

  FIG. 10 is a cross-sectional view for explaining the configuration of the rider.

  The rider 20 shown in FIG. 10 is an example used as the intrusion monitoring device MD. The intrusion monitoring device MD is shown attached to the inclined wall surface WL, but may differ from the actual shape such as the shape and length of the component. Further, in FIG. 10, it is assumed that the intrusion monitoring device MD is installed in the upside-down state.

  The intrusion monitoring device MD includes, for example, a pulsed semiconductor laser LD that emits a laser beam, a collimating lens CL that converts divergent light from the semiconductor laser LD into parallel light, and laser light collimated by the collimating lens CL. A mirror unit MU that scans and emits light toward a monitoring space by a rotating mirror surface and reflects light reflected from an object, and a lens LS that collects reflected light from the object reflected by the mirror unit MU And a photo diode PD for receiving the light collected by the lens LS, a motor MT for rotating the mirror unit MU, and a case CS for housing these. The photodiode PD has a plurality of pixels arranged in the Y direction. Further, the processing circuit is provided to obtain distance information according to the time difference between the emission timing of the semiconductor laser LD and the light reception timing of the photodiode PD. The control circuit 10 may double as this processing circuit, or a dedicated processing circuit for obtaining distance information may be provided.

  A semiconductor laser LD and a collimating lens CL constitute an emitting unit LPS, a lens LS and a photodiode PD constitute a light receiving unit RPS, a mirror unit MU constitutes a scanning unit, and these constitute a light emitting / receiving unit. .

  It is preferable that the light axes of the light emitting unit LPS and the light receiving unit RPS be orthogonal to the rotation axis RO of the mirror unit MU.

  The box-like housing CS fixed to the wall surface WL or the like has an upper wall CSa, a lower wall CSb opposite to the upper wall CSa, and a side wall CSc connecting the upper wall CSa and the lower wall CSb. An opening CSd is formed in part of the side wall CSc, and a transparent plate TR is attached to the opening CSd.

  The mirror unit MU has a shape in which two square pyramids are joined in an opposite direction and integrated, that is, four pairs (but not limited to four pairs) of mirror surfaces M1 and M2 inclined in opposite directions. ) Have. The mirror surfaces M1 and M2 are preferably formed by vapor deposition of a reflective film on the surface of a resin material (for example, PC) in the shape of a mirror unit.

  The mirror unit MU is coupled to the shaft MTa of the motor MT fixed to the housing CS, and is rotationally driven. Here, the axis line (rotational axis line) of the axis MTa extends in the Y direction tilted with respect to the vertical direction, and the ZX plane formed by the Z direction and the X direction orthogonal to the Y direction However, the axis of the axis MTa may be aligned in the vertical direction.

  An object detection principle of the intrusion monitoring device MD will be described. In FIG. 10, divergent light intermittently emitted in a pulse form from the semiconductor laser LD is converted into a parallel light beam by the collimator lens CL, is incident on the first mirror surface M1 of the rotating mirror unit MU, and is reflected there. Further, after being reflected by the second mirror surface M2, the light is transmitted as a laser spot light having, for example, a vertically long rectangular cross section toward the outside monitoring space through the transparent plate TR. The direction in which the emitted laser spot light is reflected by the object and returns as the reflected light is referred to as a light transmission / reception direction. The laser spot beams traveling in the same light emitting and receiving direction are detected by the same pixel.

  FIG. 11 is an explanatory view for explaining a state in which the inside of the monitoring space of the intrusion monitoring device MD is scanned with the laser spot light SB (indicated by hatching) emitted according to the rotation of the mirror unit MU.

  Here, in the combination of the first mirror surface M1 and the second mirror surface M2 of the mirror unit MU, the crossing angles are different.

  The laser light is sequentially reflected by the rotating first mirror surface M1 and the rotating second mirror surface M2. First, the laser light reflected by the first mirror surface M1 and the second mirror surface M2 of the first pair is horizontally shifted from the left to the right in the region Ln1 of the monitoring space according to the rotation of the mirror unit MU. Is scanned.

  Next, the laser light reflected by the first mirror surface M1 and the second mirror surface M2 of the second pair is horizontally oriented from the left in the second region Ln2 from the top of the monitoring space according to the rotation of the mirror unit MU. It is scanned to the right.

  Next, the laser light reflected by the third mirror surface M1 and the second mirror surface M2 of the third pair horizontally from the left in the third region Ln3 from the top of the monitoring space according to the rotation of the mirror unit MU It is scanned to the right.

  Next, the laser light reflected by the fourth mirror surface M1 and the second mirror surface of the fourth pair horizontally from left to right in the lowermost region Ln4 of the monitoring space according to the rotation of the mirror unit MU. It is scanned.

  This completes one scan of the entire monitoring space that can be monitored by the intrusion monitoring device MD. Thus, in the case where the laser spot beam is scanned two-dimensionally without a gap (when the trace of the scanned laser spot beam is adjacent (for example, region Ln1 and region Ln2), it means that the adjacent trajectory contacts without gap. This is preferable because it is possible to obtain the distance image 200 that makes it easy for the user to intuitively grasp the space at the time of setting of the intrusion monitoring device MD, including the case of partially overlapping).

  One frame FL is obtained by combining the images obtained by scanning the regions Ln1 to Ln4. Then, after the mirror unit MU makes one rotation, if the first mirror surface M1 and the second mirror surface M2 of the first pair return, again from the top area Ln1 to the bottom area Ln4 of the monitoring space The scan is repeated to obtain the next frame FL.

  In FIG. 10, of the scanned and projected light flux, a part of the laser light that strikes the object and is reflected again passes through the transparent plate TR and is incident on the second mirror surface M2 of the mirror unit MU in the housing CS. Here, the light is reflected by the first mirror surface M1, condensed by the lens LS, and detected for each pixel by the light receiving surface of the photodiode PD.

  Furthermore, the processing circuit obtains distance information according to the time difference between the emission timing of the semiconductor laser LD and the light reception timing of the photodiode PD. As a result, the object is detected in the whole area in the monitoring space, and as a distance image 200 (also referred to as a measurement point marker group) having distance information (three-dimensional measurement value) for each pixel (measurement point) A frame FL (see FIG. 11) can be obtained. The shape of the distance image 200 is the same as the shape of the spot beam SB actually scanned. The distance image 200 is sent from the processing circuit to the control device 10 (computer). Also, they are sent to and stored in the server 30 as needed.

  Although the embodiments to which the present invention is applied have been described above, the present invention is not limited to these embodiments. For example, when an object is recognized in the object recognition range 202, a part of the alarm operation or another operation may be performed as a preliminary operation or the like. For example, when an object is recognized in the object recognition range 202, recording is started as a preliminary operation, a yellow light is turned on, or the like. After that, when the recognized object intrudes into the monitoring target range 201, as the alarm operation, the sound and the red light are turned on (flashing) with the continuation of the recording, and the operation is performed in two steps.

  Besides, the present invention can be variously modified based on the configuration described in the claims, and they are also within the scope of the present invention.

10 control units,
11 CPU,
12 RAM,
13 ROM,
14 HDD,
20 riders,
21 alarm device,
30 servers,
100 intrusion monitoring devices,
111 Monitored Range Management Department,
112 Object recognition range management unit,
113 object recognition processing unit,
114 intrusion detection processing unit,
200 range images,
201 Monitored range,
202 Object recognition range.

Claims (7)

An object recognition that recognizes that a moving object has entered around the monitoring target range, while setting a monitoring target range in which an alarm operation is performed when the object enters, within a range where the object detection sensor detects an object Setting the range (a),
If there is an object within the object recognition range, a step (b) of recognizing the object;
Performing an alarm operation when part or all of the recognized object intrudes into the monitored area (c);
Intrusion monitoring method. The interval from the outer periphery of the monitoring target range corresponding to the type of the object to the outer periphery of the object recognition range is determined in advance and stored in the computer,
2. The apparatus according to claim 1, wherein the interval corresponding to the type of the object selected by the user is read from storage in the computer, and the object recognition range is set around the monitoring target range at the read interval. Intrusion monitoring methods.   The intrusion monitoring method according to claim 1, wherein the object recognition range is set to one object recognition range so as to surround all of the plurality of monitoring target ranges when there are a plurality of the monitoring target ranges.   The intrusion monitoring method according to claim 1, wherein the object recognition range sets the object recognition range for each of the monitoring target ranges when there are a plurality of the monitoring target ranges. The sorting criteria for sorting whether or not the object that has entered the object recognition range is an object to be alerted is stored in the computer,
Said step (b)
Recognizing an object that has entered the object recognition range (b1);
Determining whether or not the recognized object is an object to be alarmed by comparing it with the sorting criterion, and if it is an object to be alarmed, selecting (b2) as an object to be alarmed; Have
The step (c) is
The intrusion monitoring method according to any one of claims 1 to 4, wherein the alarm operation is performed when an object selected to be alarmed intrudes into the monitoring target range.   An intrusion monitoring program that causes a computer to execute the intrusion monitoring method according to any one of claims 1 to 5. An object detection sensor that detects an object;
The object detection sensor is set within a range for detecting an object, and it is set in a monitoring target range in which an alarm operation is performed when an object enters, and is recognized around the monitoring target range to recognize that a moving object enters. A storage unit that stores an object recognition range;
A control unit that recognizes an object that has entered the object recognition range, and outputs an alarm signal when the recognized object intrudes into the monitoring target range;
Intrusion monitoring device.