JP6342169B2 - Object detection sensor and program - Google Patents

Description

  The present invention relates to an object detection sensor that detects a distance to an object to be measured in a warning area by light projection and reception.

  Conventionally, in order to monitor a wide range of warning areas such as outdoors, various types of exploration signals such as laser light, visible light, ultrasonic waves, infrared rays, etc. are irradiated into the warning area and reflected return signals from the object are received. An object detection sensor for detecting an object in a warning area is known.

  For example, Patent Document 1 discloses a security system using a laser sensor that projects laser light while rotating and scanning a predetermined angle range and determines the presence of an intruder based on a distance value calculated when receiving reflected light. Has been.

Japanese Patent Laid-Open No. 10-244102

  The laser sensor disclosed in Patent Document 1 targets an intruder, and when there is a change in the distance data acquired by the laser sensor in an arbitrarily set two-dimensional monitoring area, the change occurs a predetermined number of times. If it continues, the presence of an intruder is detected, and it is further determined whether or not it is an intruder based on the amount of movement of the intruder. As described above, the laser sensor disclosed in Patent Document 1 prevents the erroneous determination due to planting or installation by adding the duration of the change of the distance data and the amount of movement to the determination condition when detecting an intruder. Yes.

  Here, the detection target of the object detection sensor is not limited to the intruder, and the vehicle on which the intruder gets (stolen) may be the detection target. On the other hand, there are cases where it is desired to exclude vehicles and make only intruders to be detected.

When a vehicle is included as a detection target, according to the technique of Patent Document 1, the duration of change and the amount of movement are used as determination conditions. Therefore, if this determination condition is satisfied, the vehicle can be detected. .
However, the vehicle and the intruder are greatly different in characteristics such as the maximum movement speed, shape, and size. For this reason, when the intruder and the vehicle are set as detection targets using the determination condition as in Patent Document 1, these detection targets cannot be detected with high accuracy.

For example, a vehicle moving fast has a short time during which the distance data changes. In order to reliably detect such a vehicle, if the determination condition of the duration of change is shortened, the possibility that a short-term change due to an erroneous determination factor such as planting will also satisfy the determination condition increases. The possibility of misjudgment increases.
Conversely, if the determination condition for the duration of change is lengthened in order to eliminate such erroneous determination, there is a high possibility that a vehicle moving quickly will not be detected, and there is a vehicle as a detection target. However, there are cases where this cannot be detected.

  Further, when it is desired to exclude a vehicle as a detection target, the vehicle and the intruder cannot be distinguished from each other under the same determination condition as in Patent Document 1.

  Therefore, an object of the present invention is to propose an object detection sensor that can detect a detection target with high accuracy even when monitoring a warning area where a vehicle may exist.

  In order to achieve the above object, an object detection sensor according to the present invention is an object detection sensor that monitors a warning area and detects a detection target existing in the warning area, from one end to the other end of the warning area. The detection unit that scans every measurement cycle and generates distance measurement data indicating the distance to the measurement point in each direction in the alert area, and the same for the current distance measurement data compared to the past distance measurement data Grouping means for grouping the measurement points whose distances have changed due to the object to be measured as a change area, and a line for detecting line segment data from the measurement points that can be approximated as a line segment among the measurement points grouped as the change area Segment detection means, first vehicle shape evaluation means for calculating the first vehicle shape degree by evaluating the likelihood of the vehicle shape for each line segment data from the line segment data, and two line data from the line segment data The second vehicle shape evaluation means for calculating the second vehicle shape degree by evaluating the vehicle shape likelihood for each combination of the minute data, and the first vehicle shape degree and the second vehicle shape degree are used. And a determination means for determining whether the change region is the detection target based on the third vehicle shape. And

  In such a configuration, the object detection sensor evaluates the likelihood of the vehicle shape for each piece of line segment data included in the change area generated in the warning area, and each combination of the two line segment data. Evaluate the shape of the vehicle. And based on the vehicle shape likeness calculated using both of these two evaluation results, it acts to determine whether or not the change area is a detection target.

  According to such a configuration, since the vehicle shape likelihood is calculated based on the evaluation result of one line segment data and the evaluation result of the combination of two line segment data, the vehicle shape likelihood can be calculated with high accuracy. As a result, it is possible to improve the accuracy of determining whether or not the change area is a detection target.

  In the object detection sensor according to the present invention, the determination unit includes a vehicle degree evaluation unit that calculates a vehicle presence degree by evaluating a degree of presence of the vehicle over the plurality of measurement periods from the third vehicle shape degree. The means for determining whether the change region is the detection target based on the vehicle presence degree, wherein the vehicle degree evaluation means has the third vehicle shape degree higher than a first threshold value. The higher the number of times, the higher the vehicle presence level may be calculated.

  In such a configuration, the object detection sensor acts so as to evaluate the degree of existence of the vehicle higher as the number of times the distance measurement data having a high vehicle shape is obtained.

  According to such a configuration, in addition to the shape of the vehicle in the change area, it is possible to evaluate from the viewpoint of the time axis that the shape of the vehicle is high, so the possibility that the vehicle exists in the warning area can be accurately detected. it can.

  In the object detection sensor according to the present invention, the vehicle degree evaluation means may calculate the vehicle presence level higher according to the number of times than when the number of times is less than when the number is less. The degree of increasing the presence may be increased.

  In such a configuration, the object detection sensor acts to highly evaluate the degree of existence of the vehicle even in a situation where the number of times the distance measurement data having a high vehicle shape is obtained is not so many.

  According to such a configuration, even for a vehicle passing at a high speed, the degree of vehicle presence can be highly evaluated, and the possibility that the vehicle is present in the alert area can be accurately detected.

  Further, in the object detection sensor according to the present invention, the vehicle degree evaluation means sets a predetermined bottom value when the third vehicle shape degree is higher than a second threshold value that is larger than the first threshold value. You may add to the abundance.

  In such a configuration, the object detection sensor acts to add a predetermined bottom value to the vehicle presence when distance measurement data having a very high vehicle shape is obtained.

  According to such a configuration, it is possible to immediately evaluate the degree to which the vehicle exists with respect to a vehicle having a high moving speed, and it is possible to accurately detect the possibility that the vehicle exists in the alert area.

  In order to achieve the above object, the program according to the present invention scans from one end to the other end of the alert area at every measurement cycle, and generates distance measurement data indicating the distance to the measurement point in each direction in the alert area. A program that is executed by an object detection sensor that detects the detection target existing in the warning area by monitoring the warning area based on the distance measurement data. Among the measurement points grouped as the change area, and a process for grouping the measurement points whose distance has changed due to the same object to be measured with respect to the current distance measurement data as compared with the distance measurement data of A process for detecting line segment data from measurement points that can be approximated as a line segment, and a first vehicle shape by evaluating the likelihood of the vehicle shape for each line segment data from the line segment data A process for calculating a degree, a process for calculating a second vehicle shape degree by evaluating the likelihood of a vehicle shape for each combination of two line segment data from the line segment data, the first vehicle shape degree and the first A process of calculating a third vehicle shape degree using the second vehicle shape degree, and a process of determining whether the change region is the detection target based on the third vehicle shape degree. It is characterized by.

  According to the present invention, it is possible to accurately detect a detection target even when monitoring a warning area where a vehicle may exist.

It is the schematic which shows the whole structure of the security system using the object detection sensor which concerns on this invention. It is a block diagram which shows the structure of the object detection sensor which concerns on this invention. It is a block diagram which shows the structure of the security apparatus in a security system. It is a figure which shows the outline | summary of the detection method of the detection target by the object detection sensor which concerns on this invention. It is a figure which shows the concept of the detection process of the change area | region by the object detection sensor which concerns on this invention. It is a figure which shows the concept of the calculation process of the vehicle presence degree by the object detection sensor which concerns on this invention. It is a flowchart which shows the monitoring operation | movement of the object detection sensor which concerns on this invention. It is a flowchart which shows the detection target determination process of the object detection sensor which concerns on this invention. It is a flowchart (1) which shows the vehicle presence degree calculation process of the object detection sensor which concerns on this invention. It is a flowchart (2) which shows the vehicle presence degree calculation process of the object detection sensor which concerns on this invention.

Embodiments of the present invention will be specifically described below with reference to the drawings.
In the present embodiment, a security system that performs outdoor monitoring using an object detection sensor in a monitoring building is illustrated, but the scope of the present invention is not limited to this.

FIG. 1 is a configuration diagram showing a security system 1 using an object detection sensor 2 according to the present invention.
FIG. 1 is a schematic plan view showing the relationship between an object detection sensor 2 installed on an outdoor wall surface of a monitoring building 3, a warning area 4 of the object detection sensor 2, and a security device 5 installed in the monitoring building 3. It is shown on the figure. In the example of FIG. 1, three object detection sensors 2 are installed around the monitoring building 3. The object detection sensors 2 are each connected to the security device 5 via a communication line, and the security device 5 is connected to a remote monitoring center 6 via a communication line network 7. Although not particularly illustrated, security sensors such as a heat ray sensor and an open / close sensor are also installed inside the monitoring building 3 and connected to the security device 5.

  The object detection sensor 2 performs spatial scanning at a predetermined measurement period while irradiating laser light in a preset alert area 4 and receives reflected light reflected by an object on the optical path, thereby The position of an object as an object to be measured existing in is detected. In this way, the object detection sensor 2 monitors an object appearing in the alert area 4 and outputs a detection signal including its own address information to the guard device 5 when it is determined that the object is a detection target.

  The security device 5 monitors the inside and outside of the monitoring building 3 serving as a monitoring area. Then, the security device 5 determines an abnormality in the monitoring area based on the detection signal of the object detection sensor 2 and outputs an abnormality signal to the monitoring center 6.

  The monitoring center 6 is a facility provided with a center device 61 operated by a security company or the like. The center device 61 is composed of one or a plurality of computers, and realizes the function of the monitoring center 6 related to the present invention. In the monitoring center 6, various devices are controlled by the center device 61, the abnormality signal received from the security device 5 is recorded, information on the abnormality is displayed on the display 62, and a plurality of monitoring areas to be monitored by the monitor are displayed. Monitoring.

<Object detection sensor>
Next, the configuration of the object detection sensor 2 will be described with reference to FIG. FIG. 2 is a block diagram showing a configuration of the object detection sensor 2.
The object detection sensor 2 is installed with a horizontal or constant depression angle on the outdoor wall surface of the monitoring building 3 so that the detection target in the alert area 4 is irradiated with laser light. Since the object detection sensor 2 in the present embodiment targets a person and a vehicle as detection targets, the installation height on the wall surface is set to a height at which the person and the vehicle that have entered the alert area 4 are irradiated with laser light. Further, the object detection sensor 2 operates upon receiving power supply from the security device 5.

  The object detection sensor 2 is composed of a communication unit 21 connected to the security device 5 for communication, a detection unit 22 for irradiating and receiving laser light, and a storage unit configured to store various setting information, programs, and the like. 23 and a control unit 24 that is configured by an MPU, a microcomputer, and the like and controls each unit.

  The communication unit 21 is connected to the security device 5, receives a security start signal and a security release signal output from the security device 5, and outputs the signals to the control unit 24. Further, when the control unit 24 determines the presence of the detection target in the alert area 4, the communication unit 21 transmits a detection signal including its own address information to the guard device 5.

  The detection unit 22 scans the alert area 4 with a laser beam and detects the position of an object as a measurement object that reflects the laser beam. The detection unit 22 includes, for example, a laser oscillation unit 221 that emits near-infrared light having a wavelength of about 890 nm, a scanning mirror 222 that reflects laser light and irradiates the object detection sensor 2, and scanning that drives the scanning mirror 222 to rotate at a constant speed. A control unit 223, a reflected light detection unit 224 that includes a light receiving element and is provided in the vicinity of the laser oscillation unit 221, and a distance measurement data generation unit 225 that generates distance measurement data as a laser light irradiation result are provided.

  The irradiation direction of the laser light emitted from the laser oscillation unit 221 is controlled by the scanning mirror 222 and the scanning control unit 223 to scan at least the entire warning area 4. This scanning can be performed on a horizontal plane according to the installation angle of the object detection sensor 2 or on a plane that approaches the ground as the distance increases with a depression angle. The scanning is performed at a predetermined measurement cycle (for example, 30 msec). For example, the scanning may be repeatedly performed in the same direction, or the backward scanning may be performed after scanning in the forward direction.

  The distance measurement data generation unit 225 includes a distance between the object detection sensor 2 calculated from the time required from the irradiation of the laser light to the detection of the reflected light and the object to be measured (measurement point) reflecting the laser light, and a scanning control unit. The relative position of the object that reflects the laser beam, that is, the measurement point that reflects the laser beam, is calculated based on the angle of the scanning mirror 222 that is rotationally driven by 223 (the direction in the alert area 4). The relative position is the position of the measurement point with reference to the object detection sensor 2, and specifically, the position of the surface that reflects the laser beam on the object. In addition, if the reflected light does not return within a predetermined time, the distance measurement data generation unit 225 determines that there is no object within the distance that can be irradiated with the laser light, and records the predetermined pseudo data as a relative position. To do. The pseudo data may be a predetermined value, for example, a distance value that is an outer periphery of the alert area 4 to be monitored by the object detection sensor 2 or an appropriate value that is equal to or greater than an effective measurement distance by laser light.

The measurement data obtained by the distance measurement data generation unit 225 is referred to as distance measurement data in this embodiment. The distance measurement data is specifically the result of measuring the alert area 4 at a predetermined angular interval (for example, 0.25 °) by one scan by the detection unit 22. For example, when distance measurement data is acquired at intervals of 0.25 ° for a range of 180 °, 721 distance values are obtained. A set of these 721 distance values becomes one distance measurement data. The distance measurement data is stored as information (table) of a collection of a plurality of measurement point data in which angles (directions) and distances are associated with each other.
The distance measurement data generation unit 225 generates distance measurement data and outputs it to the control unit 24 every time one scan of the detection unit 22 ends at a predetermined cycle interval (for example, 30 msec).

  The storage unit 23 is composed of ROM, RAM, or HDD, stores address information and various programs for specifying the object detection sensor 2 itself, and further stores various information for operating the object detection sensor 2. Remember. Specifically, the storage unit 23 includes warning area information indicating the set warning area 4, reference data generated by the control unit 24, object tracking information detected by the detection unit 22, Current state information indicating the state of the alert area 4 is stored. The storage unit 23 stores distance measurement data for the past predetermined period output from the detection unit 22.

The warning area information is information indicating the warning area 4 set according to the security planning of the monitoring area by a security company or the like as a range to be monitored by the object detection sensor 2, for example.
The warning area information is input by designating a range on the scanning plane by the detection unit 22 from a setting terminal, an operation unit (not shown), or the like when the object detection sensor 2 is installed or when security planning of the monitoring area is changed. The range of the warning area 4 that is input is associated with an angle (direction) from the detection unit 22 and a distance value for each predetermined angular interval (for example, 0.25 °) scanned by the detection unit 22. It is stored in the storage unit 23 as a table of angles (directions) and distances. In the present embodiment, as shown in FIG. 1, an example in which the alert area 4 is set in a semicircular shape centering on the object detection sensor 2 will be described.
Note that the alert area information is not limited to this, and it is only necessary to store the positional relationship between the information indicating the range of the alert area 4 and the object detection sensor 2 so that the object detection sensor 2 can be identified. It may be set and stored in two-dimensional coordinates indicating a simple positional relationship.

  The reference data is comparison reference information used for extracting an object that has newly appeared in the alert area 4 in comparison with the current distance measurement data in the detection target determination process described later. Is generated from distance measurement data acquired at any past time from the current to the present. The reference data may be stored as a table of angles (directions) and distances. Further, the reference data may be generated at any past time point, and may be updated using distance measurement data acquired at any time. In the present embodiment, an example will be described in which reference data is generated and stored from distance measurement data acquired by the first scan after the detection unit 22 starts scanning.

The tracking information is information used to track an object that has newly appeared in the alert area 4 in a detection target determination process described later over a plurality of periods. The tracking information includes the position and size of the object in the current cycle, whether the object is determined as a detection target, the position and size of the object that first appeared in the alert area 4, and the position and size in each cycle up to the present. It is stored in correspondence.
The tracking information is stored in association with the vehicle presence level and the counter. Details of such information will be described later.

  The current state information stores whether or not a detection target exists in the current alert area 4 as a determination result by the control unit 24. When the presence of the detection target is determined by the control unit 24, the state with detection is stored, and when the detection target is not determined, the state without detection is stored.

  The control unit 24 includes a microcomputer including a CPU, a ROM, a RAM, and the like and peripheral circuits thereof, and controls the above-described units. For this purpose, the control unit 24 is acquired from the detection unit 22 and the drive control unit 241 that controls the driving of the detection unit 22 as functional modules realized by the microcomputer and a computer program executed on the microcomputer. A reference data generation unit 242 that generates reference data from the distance measurement data, and a detection target determination unit 243 that determines the presence or absence of a detection target are provided.

The drive control unit 241 outputs a drive signal to the detection unit 22 when a security start signal is input from the security device 5 via the communication unit 21, starts driving the detection unit 22, and is scanned by the scan control unit 223. Driving of the mirror 222 and irradiation of laser light by the laser oscillation unit 221 are started. In addition, when a security release signal is input from the security device 5, the drive control unit 241 outputs a drive stop signal to the detection unit 22, stops driving the detection unit 22 at the end of scanning at that time, and scans a mirror 222. Is stopped and laser beam irradiation is stopped.
Thus, it becomes possible to prevent breakage of the drive parts due to continuous operation by driving the detection unit 22 in accordance with the start of security of the security device 5.

The reference data generation unit 242 generates reference data using the distance measurement data acquired from the detection unit 22. As described above, in the present embodiment, an example will be described in which the reference data is generated and stored from the distance measurement data acquired in the first scan after the scanning by the detection unit 22 is started. That is, when the drive signal is output from the drive control unit 241 and the detection unit 22 starts scanning, the reference data generation unit 242 stores the distance measurement data output in the first scan in the storage unit 23 as reference data. To do. When the distance value corresponding to a certain angle is not within the alert area 4 as the position of the measurement point in the distance measurement data, the distance to the outer periphery of the alert area 4 corresponding to the angle is stored as reference data. Since the distance to the measurement point by the scanning is stored in the reference data, existing objects such as planting and outer walls existing in the alert area 4 at the time of the scanning are taken in as the reference data.
However, the present invention is not limited to this, and the reference data generation unit 242 sets the distance value for each scanning angle in the distance measurement data a predetermined number of times (for example, scanning performed for 5 minutes) after the detection unit 22 starts scanning. The frequency may be obtained, and the distance value with the highest frequency may be adopted as the reference value of the scanning angle to generate reference data.

The detection target determination unit 243 compares the current distance measurement data with the reference data, detects an object appearing in the alert area 4 as a change area, and calculates a feature amount and a movement amount of the object. Then, based on the feature amount and the movement amount, it is determined whether or not the object is a detection target.
In the outdoor environment, there are many moving objects such as small animals compared to indoors, and there may be shaking such as planting and flying objects due to the wind. Therefore, a new object appears in the alert area 4 and is immediately detected. It can lead to misjudgment. For this reason, in this embodiment, if the detection target determination part 243 detects the object which appeared in the alert area 4, this object will be evaluated over multiple periods, and it will be determined whether it is a detection target.

Specifically, the detection target determination unit 243 calculates, for each corresponding angle, a difference between the distance value for each scanning angle obtained from the distance measurement data and the distance value for each angle stored in the reference data, A measurement point that is closer than the reference data, that is, a measurement point whose distance value has changed is detected as a change point. Then, the detection target determination unit 243 groups measurement points whose distance values have changed due to the same object to be measured as a change area. Further, the detection target determination unit 243 performs a tracking process for determining whether or not there is a change area caused by the same object in the detection result of the previous cycle with reference to the tracking information for the change area. The association with the detection result of the previous cycle is performed based on the distance and size between the objects detected in both cycles. That is, when the distance between the objects detected in both cycles is within the threshold value and the variation in size is within the threshold value, the change areas are associated.
When association is performed, the tracking information associates the position and size of the change area detected in the current period and whether or not it is determined as the detection target with the position and size in each period until now. Is remembered.

Next, the detection target determination unit 243 calculates the feature amount of the change area. In the present embodiment, as feature quantities, a person presence indicating a person-like height, a car presence indicating a vehicle-like height, a bird presence indicating a bird-like height, a fog-like degree The fog presence indicating the height is used. These feature amounts are calculated in the range of the minimum value 0 to the maximum value 100.
Hereinafter, the calculation processing of each feature amount will be described in order. However, since the vehicle presence will be described later in detail, the human presence, the bird presence, and the fog presence will be described here.

The person presence level is a feature amount focusing on the size of the person in the width direction and the characteristics of the moving speed.
Specifically, first, of the measurement points included in the change region, the measurement point having the maximum value and the measurement point having the minimum value are selected as both end points of the change region. Then, the distance between both end points is calculated from the angle and distance information of the both end points, and the human presence is calculated higher as the distance is closer to a predetermined range in consideration of the length in the width direction of the person.
Also, the moving speed of the change area is calculated from the tracking information, and if the moving speed is so fast that it cannot be considered as a person, the human presence is set to a very low value.

The bird presence level is a feature amount focusing on the flight speed of the bird and the characteristics of the movement in the flight state.
In the specific calculation process, the moving speed of the change area is calculated from the tracking information, and the bird presence is calculated higher as the moving speed is higher.
In addition, since the birds are flapping in the flight state, the size and shape of the change area frequently change. Considering this, referring to the tracking information, the bird presence degree is calculated to be higher as the size and shape of the change area change more frequently.

The degree of fog presence is a feature amount focusing on the characteristic of fog density.
A specific calculation process calculates a variance value as a difference between adjacent measurement points for measurement points included in the change region. Since the fog concentration is uneven, the greater the dispersion value, the higher the fog presence.

Each feature amount calculated in this way is used as follows when determining whether or not it is a detection target.
The detection target determination unit 243 calculates the movement amount from the position where the change area first appears in the alert area 4 to the current position. If the amount of movement is equal to or greater than a predetermined distance, the change area is determined as a detection target.
Here, the detection target determination unit 243 sets the predetermined distance as the determination condition to a dynamically different value according to each feature amount of the change area described above. Specifically, the person presence level and the vehicle presence level, which are feature amounts of the person and vehicle that are the detection targets, are totaled as the detection target feature amounts, and the bird presence level and fog that are the feature amounts of the birds and fog that are not the detection targets. When the non-detection target feature quantity is greater than the detection target feature quantity (for example, when it is greater than a predetermined amount), the possibility of not being a detection target is high. The predetermined distance is set longer (for example, 3 m) than the predetermined distance (for example, 1 m).
As described above, when the feature amount of the non-detection target is large, the detection target determination unit 243 eliminates erroneous determination by setting a loose determination criterion.
If it is determined as a detection target, the presence of detection is stored in the current state information of the storage unit 23, and if it is not determined as a detection target, the detection presence information is deleted from the current state information, and information indicating no detection is stored. .

This will be described in more detail with reference to FIGS. FIG. 4 illustrates a detection target detection method by the detection target determination unit 243. As shown in the figure, when the distance d _n to the current measurement point is shorter than the distance d _b to the measurement point (reflection point by the measurement object) stored in the reference data at a certain scanning angle, the measurement object Q is detected as a change area, the change area by the measurement object Q detected in the current period is associated with the change area by the measurement object Q ′ detected in the previous period, and this change area is the warning area 4. It is determined whether or not it is a detection target based on the moving distance from the position where it first appears to the current position.

FIG. 5 shows a concept of a change area detection process performed by the detection target determination unit 243 by comparison of distance measurement data. In FIG. 5, the horizontal axis is an angle, and the vertical axis is a distance value corresponding to an angle component. The detection target determination unit 243 calculates a difference in distance value for each angle component in the current distance measurement data and the reference data, and a change point where the current distance measurement data is closer than the reference data by a predetermined distance or more. Check for the presence or absence. The change point is a point where the distance difference is a negative value and is −Δ or less (a point where the distance difference is Δ or more on the − side) in the lower diagram of FIG. 5. If there is a change point, the detection target determination unit 243 examines the continuous section, and groups the continuous sections as a change area.
Specifically, the distance between two change points is calculated, and if this distance is within a predetermined distance (for example, within 70 cm), it is determined that it can be regarded as a continuous section, and change points constituting one change area To do.
The detection target determination unit 243 determines the position (angle and distance value) of the measurement point included in the change area if the size of the grouped change area is equal to or larger than a predetermined size (for example, 15 cm) that can be determined as a part of a person or a vehicle. ) Is stored in the tracking information as current cycle information. Then, the detection target determination unit 243 associates the change area with the change area of the previous cycle.
Note that change points (isolated points) that could not be determined as continuous sections may be removed as noise, and integration or noise removal by expansion / contraction processing may be performed.

FIG. 6 is a diagram illustrating the concept of the vehicle presence degree calculation process performed by the detection target determination unit 243. This figure shows an example of a situation where a vehicle is present in the alert area 4 of the object detection sensor 2 in an environment where dust is generated by the movement of the vehicle. In the drawing, black circles indicate measurement points that have changed, and are measurement points that are grouped as one change region by the above-described process of detecting the change region.
In this example, since the laser light emitted from the sensor is reflected by dust, some measurement points are in front of the side surface of the vehicle, and the external shape of the vehicle is not accurately detected. . For this reason, although one change area is detected, the number of measurement points reflected by the side surface of the vehicle is smaller than when there is no dust.

First, the detection target determination unit 243 performs a process of detecting line segment data from measurement points that can be approximated as a line segment among measurement points grouped as a change area.
In the process of approximating the line segment, for example, three adjacent measurement points are first selected from the grouped measurement points. Then, of the two measurement points adjacent to each other, the difference between the distance values in the distance measurement data is within a predetermined distance range that can be regarded as a short distance, and the measurement points of these three measurement points When the angle between two line segments formed by connecting two adjacent points is within a predetermined angle range that can be regarded as a straight line, all the measurement points grouped in the process of determining one line segment It can be realized by doing. FIG. 6 shows an example in which three line segments are approximated.
When line segment approximation is performed, the positions (angle and distance value) of two measurement points at both ends of the line segment among the measurement points that can be approximated to one line segment are specified as line segment data.
Note that the process for approximating the line segment may use a known method, for example, a method such as a Hough transform or a least square method.

  Next, the detection target determination unit 243 evaluates each piece of line segment data for each piece of line segment data (a white arrow in the figure), and calculates a vehicle shape degree f1 that indicates the degree of vehicle shape-likeness. Perform the process. For the evaluation of this one line segment data, the length of the line segment is calculated from the line segment data, and the larger this is, the higher the evaluation is made and the higher the vehicle shape degree f1 is calculated.

As an example of the evaluation method of the one line segment data, for example, the following method can be considered.
When the length of the line segment is less than the first predetermined length, the vehicle shape degree f1 = 0.
-When the length of the line segment is not less than the first predetermined length and not more than the second predetermined length, the vehicle shape factor f1 = 0 when the first predetermined length and the second predetermined length The vehicle shape f1 is increased approximately proportionally so that the vehicle shape f1 = 50.
When the length of the line segment is larger than the second predetermined length, the vehicle shape degree f1 = 50.

  Furthermore, the detection target determination unit 243 evaluates all combinations of the two line segment data (broken arrows in the figure), and calculates a vehicle shape degree f2 that indicates the height of the degree of vehicle shape. Do. For the evaluation of the combination of the two line segment data, the length of each line segment is calculated from the two line segment data, and the higher the total value, the higher the evaluation. Also, the angle formed by the two line segments is calculated from the two line segment data, and the higher the evaluation is, the higher the evaluation is. The higher the evaluation is, the higher the vehicle shape degree f2 is calculated.

As an example of an evaluation method of the combination of the two line segment data, first, an evaluation value is calculated as follows for the total value of the lengths of the two line segment data.
In the case where the total value is 0 or more and the third predetermined length or less, the evaluation value is approximately proportionally so that the evaluation value is 0 when the total value is 0 and the evaluation value is 25 when the total value is 3 increase.
When the total value is larger than the third predetermined length, the evaluation value is 25.
Then, according to the angle formed by the two line segment data, the evaluation value is multiplied by a predetermined value to obtain the vehicle shape degree f2. For example, the evaluation value is multiplied by a predetermined value as follows.
When the absolute value of the difference between the formed angle (recess angle) and 90 degrees is not less than 0 and not more than the first predetermined difference, the evaluation value is multiplied by 2.
When the absolute value of the difference between the formed angle (recess angle) and 90 degrees is greater than the first predetermined difference and less than or equal to the second predetermined difference, the evaluation value is multiplied by 1.
When the absolute value of the difference between the formed angle (inferior angle) and 90 degrees is larger than the second predetermined difference, the evaluation value is multiplied by 0.5.

The detection target determination unit 243 performs a process of summing all the calculated vehicle shape degrees f1 and the vehicle shape degrees f2 and calculating this as a vehicle shape degree F that indicates the height of the degree of vehicle shape likelihood of the change region. .
In the evaluation method according to the above-described example, for example, even if only two line segment data are detected for one side of the vehicle, each line segment data is sufficiently long. If the length is equal to or longer than the second predetermined length, two vehicle shape factors f1 = 50 can be obtained, so that the vehicle shape factor F can be set to 100, which is the maximum value.
Further, for example, one line segment data is detected for each of the two side surfaces of the vehicle, and even if the length of one of the line segment data is short, the length of the other line segment is sufficiently long. If the length of the other line segment is equal to or longer than the second predetermined length in the line segment data evaluation method, one vehicle shape factor f1 = 50 can be obtained. In the evaluation method of the two line segment data, the total value of the lengths of the two line segment data is sufficiently long and exceeds the third predetermined length, and the angle formed by the two line segment data is extremely 90 degrees. If the absolute value of the difference from near 90 degrees is not less than 0 and not more than the first predetermined difference, one vehicle shape degree f2 = 50 can be obtained, so that the vehicle shape degree F can be set to the maximum value of 100.
Furthermore, even if the above-mentioned case conditions are not satisfied, if another line segment data is obtained, the vehicle shape degree F can be calculated to be high by evaluation including this.
The vehicle shape degree F is used for the calculation of the vehicle presence degree described above.

  As described above, in the present embodiment, the vehicle shape F is not calculated in consideration of all the changed measurement points in the change region at a time, but for each approximate line segment data and two approximate lines. First, an evaluation is performed for each set of minute data, and then the vehicle shape likelihood F is calculated by combining these evaluations. For this reason, even if a measurement point that has changed only a part of the vehicle shape due to the influence of the environment such as dust or mist as shown in the figure, the curved surface part of the vehicle shape or the paint of the vehicle, etc. can be detected, The vehicle shape degree F can be calculated with high accuracy.

Next, a vehicle presence degree calculation process using the vehicle shape degree F will be described. The vehicle presence level is calculated by detecting whether or not the vehicle shape degree F calculated in each measurement cycle is in a specific state, and calculating according to the number of times this state has occurred.
Specifically, the detection target determination unit 243 calculates the vehicle presence level by detecting the number of measurement periods in which the vehicle shape factor F is higher than a predetermined threshold value M (first threshold value). The detection of the number of times and the calculation of the vehicle presence degree are performed by the counter and the vehicle presence degree (initial value is the initial value when the tracking information is stored for the newly detected change area in the tracking process described above. Both are set by setting 0). If the vehicle shape degree F is higher than the threshold value M, 1 is added to the counter.

Then, the greater the number of times the vehicle shape degree F becomes higher than the threshold value M (counter value), the higher the vehicle presence degree is calculated. At this time, in the initial state where the value of the counter is small, that is, the number of times that the vehicle shape degree F is higher than the threshold value M is not so many, the degree of increasing the vehicle presence degree is the vehicle presence degree. The highest ratio with respect to the maximum value 100 is set.
For example, when the vehicle presence level is changed from the minimum value 0 to the maximum value 100 at 900 msec (30 cycles if the measurement cycle is 30 msec), the vehicle presence level included in the tracking information stored in the previous cycle is 5 is added when the counter value is 1 to 10, 4 is added when the counter value is 11 to 20, and 1 is added when the counter value is 21 to 30.

Thus, the calculation accuracy of the vehicle presence can be improved by calculating the vehicle presence high considering the number of times that the vehicle shape F is higher than the threshold value M.
Also, since the vehicle presence is calculated greatly even in a situation where the number of times is not so large, the vehicle presence can be calculated even for a vehicle passing at a high speed.

Further, the vehicle presence degree is obtained by subtracting a predetermined value from the vehicle presence degree when the vehicle shape degree F is lower than the threshold value L (threshold value L <threshold value M).
Specifically, the detection target determination unit 243 calculates the vehicle presence level as the number of measurement cycles in which the vehicle shape degree F is lower than the threshold value L increases. The process of calculating the low vehicle presence level is also performed using the above-described counter. In a state where the value of the counter is large, that is, the number of times that the vehicle shape degree F is higher than the threshold value M is still many, the degree of decreasing the vehicle presence degree is the lowest ratio with respect to the maximum value 100 of the vehicle presence degree. To be.
For example, when the vehicle presence level is changed from the maximum value 100 to the minimum value 0 at 900 msec (30 cycles if the measurement cycle is 30 msec), the counter is calculated from the vehicle presence level included in the tracking information stored in the previous cycle. 1 is subtracted when the value is 20 to 29, 4 is subtracted when the counter value is 10 to 19, and 5 is subtracted when the counter value is 0 to 9.

As described above, when the vehicle shape degree F is lower than the threshold value L, the calculation accuracy of the vehicle presence degree can be improved by calculating the vehicle presence degree low.
In addition, when the vehicle shape degree F is higher than the threshold value M, the vehicle presence degree is not greatly subtracted, so that once the vehicle shape degree F is increased, the vehicle presence degree becomes extremely high due to noise caused by short-term environmental factors. You can prevent it from being lowered.

  If the vehicle shape F is between the threshold value L and the threshold value M, the current value is maintained without adding or subtracting the value.

Furthermore, when the vehicle shape degree F is higher than a predetermined threshold value H (threshold value H> threshold value M, second threshold value), that is, when the vehicle shape degree F is very high, the detection target determination unit 243 responds to the counter. A predetermined bottom value (for example, 50) is added to the above-described vehicle presence degree calculated in the above.
This addition of the bottom value is performed only when the vehicle shape degree F is higher than the threshold value H, and when the vehicle shape degree F becomes equal to or lower than the threshold value H in the next measurement period. Subtracts the bottom price.
Thereby, only when the vehicle shape F is very high, the vehicle presence can be increased more quickly, and the vehicle presence can be immediately increased with respect to a vehicle having a high moving speed.

<Security device>
Next, the configuration of the security device 5 will be described with reference to FIG. FIG. 3 is a block diagram showing the configuration of the security device 5. The security device 5 is installed in the monitoring building 3 and vigilancely monitors the monitoring area, and notifies the remote monitoring center 6 of the location of the abnormality.

  The security device 5 includes a sensor I / F (interface) 51 connected to the object detection sensor 2 and other security sensors (not shown), and a communication unit 52 connected to the remote monitoring center 6 via the communication network 7. And an operation unit 53 operated by a user in the monitoring area, a storage unit 54 composed of an HDD, a memory, etc., and a control unit 55 composed of an MPU, a microcomputer, etc. and controlling each unit. Composed. The control unit 55 includes, as function modules, a mode setting unit 551 that sets / changes the security mode of the monitoring area and an abnormality processing unit 552 that determines that an abnormality has occurred in the monitoring area. The storage unit 54 stores management information such as security mode information and current state information, various processing programs and parameters, identification information of the security device 5, and the like.

  The mode setting unit 551 collates information input from the operation unit 53 when the user sets the security mode, and if the verification is OK, the security mode is set to the security set mode or the security release based on the input of the operation unit 53. Set to mode. The security mode set by the mode setting unit 551 is stored in the security mode information of the storage unit 54.

Here, the security set mode is set when the monitoring area including the monitoring building 3 becomes unattended, such as at night or on holidays, and when the various sensors detect a change in the event, the remote monitoring center via the communication unit 52 is set. 6 is a mode for notifying abnormality. Further, the security release mode is a mode that is set when the monitoring area is manned and does not perform abnormality notification by detection of various sensors.
The mode setting unit 551 outputs a security start signal to the object detection sensor 2 via the sensor I / F 51 when set to the security set mode, and via the sensor I / F 51 when set to the security release mode. A security release signal is output to the object detection sensor 2.

  When the abnormality processing unit 552 receives an input of a detection signal from various sensors when the current security mode stored in the storage unit 54 is the security set mode, the abnormality processing unit 552 determines that an abnormality has occurred in the monitoring area, and the current state information In addition, the abnormality type corresponding to the detection signals input from the various sensors and the information of the detected sensors are stored. In addition, when the occurrence of the abnormality is confirmed, the abnormality processing unit 552 transmits an abnormality signal including the sensor detected as the abnormality type and the identification information of the security device 5 to the remote monitoring center 6 via the communication unit 52.

<Description of operation>
About the security system 1 comprised as mentioned above, the operation | movement is demonstrated with reference to drawings. Here, an operation related to the object detection sensor 2 will be mainly described. FIG. 7 is a flowchart showing the operation of the monitoring program executed by the object detection sensor 2.

  The drive control part 241 will output a drive signal to the detection part 22, if the security start signal is received from the security apparatus 5 (step ST1-Yes), and will start the drive of the detection part 22 (step ST2). When the detection unit 22 receives the drive signal and starts scanning the alert area 4 and the distance measurement data is output (step ST3-Yes), the reference data generation unit 242 converts the distance measurement data obtained by the first scan into the distance measurement data. Based on this, reference data is generated and stored in the storage unit 23 (step ST4).

  When the distance measurement data is acquired after generating the reference data (step ST5-Yes), the detection object determination unit 243 compares the current distance measurement data with the reference data and performs detection object determination processing ( Step ST6). The detection target determination process will be described later. And if the detection information which is not output to the security device 5 is memorize | stored in the present condition information of the memory | storage part 23 based on the result of a detection target determination process (step ST7-Yes), this address information from the communication part 21 will be mentioned. A detection signal including is transmitted to the security device 5 (step ST8).

  If the object detection sensor 2 has not received a security release signal from the security device 5 (step ST9-No), the process of steps ST5 to ST8 is repeated to monitor the security area 4. On the other hand, when a security release signal is received from the security device 5 (step ST9-Yes), the drive control unit 241 outputs a drive stop signal to the detection unit 22 to stop driving the detection unit 22 (step ST10), and the storage unit The information with detection stored in the current state information 23 is changed to information with no detection (step ST11), and the series of processes is terminated. In step ST3 and ST5, if distance measurement data is not acquired even after exceeding a predetermined scanning period (for example, 30 msec), the process may be terminated as a device abnormality.

The basic operation of the object detection sensor 2 has been described above.
Next, the detection target determination process in step ST6 of FIG. 7 will be described with reference to FIG. FIG. 8 is a flowchart of the detection target determination process.

In FIG. 8, the detection target determination unit 243 reads the distance measurement data and the reference data acquired in the current cycle (step ST31), and the distance value detected from the current distance measurement data for each angle component (direction). performs difference calculation between the stored in d _n and the reference data distance value _{d b} (step ST32).

  Then, the detection target determination unit 243 checks whether there is a change point where the current distance measurement data is closer than the reference data by a predetermined distance or more from the difference result between the current distance measurement data and the reference data (step) ST33). The change point is a point where the distance difference is a negative value and is −Δ or less (a point where the distance difference is Δ or more on the − side) in the lower diagram of FIG. If there is a change point (Yes in step ST33), the detection target determination unit 243 examines the continuous section, groups the consecutive change points that are close in distance, and detects them as a change region (step ST34). Here, since the angular interval when the detection unit 22 scans is sufficiently dense as compared with the detection target (person or vehicle), the discontinuous change points (isolation points) and the size of the change region are small. An object that is smaller than a predetermined size (for example, 15 cm) that can be determined as a part of a detection target (a person or a vehicle) may be removed as noise.

The detection target determination unit 243 stores the position (angle and distance value) of the change point as the change region in the tracking information of the storage unit 23 as information on the current cycle. Then, referring to the tracking information, the processing results of the previous period and the current period are compared, and a tracking process for determining whether or not a tracking target exists for each change area is performed (ST35). In the tracking process, the presence / absence of a tracking target is determined based on whether or not there is a change area having substantially the same size within a predetermined angle and distance range between the previous cycle and the current cycle. If there is a corresponding change area, the change area becomes a tracking target. In this determination, if there is no tracking target, the change area of the current cycle is a newly appearing change area.
It should be noted that the tracking information of the change area in the previous cycle and not determined as the tracking target in the current cycle is erased by determining that the object that caused the change area has moved out of the alert area.

  Next, the detection target determination unit 243 performs a feature amount calculation process for calculating the above-described feature amount for each change region (ST36). In this feature amount calculation process, a vehicle presence degree calculation process, which will be described later, which is a process for calculating the vehicle presence degree as one of the feature quantities, is also performed.

  When the calculation of the feature amount is completed, the detection target determination unit 243 reads the position at the time of newly appearing in the alert area 4 for each change area from the tracking information, and from the position that newly appeared in the alert area 4 to the current position. The moving distance is calculated (step ST37). The movement distance is calculated from the straight line distance from the newly appearing position to the current position.

  If all the movement distances calculated for each change area do not satisfy the predetermined distance (step ST38-No), the detection target determination process ends. On the other hand, if any of the movement distances calculated for each change area is equal to or greater than the predetermined distance (step ST38-Yes), the corresponding change area is determined as a detection target, and it is determined that the detection target exists in the alert area 4. The presence of detection is stored in the current state information of the storage unit 23 (step ST39). As a result, a detection signal is output to the security device 5 in step ST8 of FIG. 7, and when the abnormality is confirmed by the security device 5, an abnormality is notified to the remote monitoring center 6.

  If there is no negative change point in ST33 (ST33-No), no object that causes a change point has appeared in the alert area 4, and therefore the tracking information is deleted. .

Next, the vehicle presence degree calculation process in step ST36 of FIG. 8 will be described with reference to FIGS. 9 and 10 are flowcharts of the vehicle presence degree calculation process.
First, the detection target determination unit 243 refers to the tracking information and selects an arbitrary change region (ST51). Then, line segment data is detected from measurement points that can be approximated as line segments among the measurement points in the selected change region (ST52).

  The detection target determining unit 243 selects one arbitrary line segment data from the line segment data detected in ST52 (ST53), and calculates the vehicle shape degree f1 (ST54). Then, it is determined whether or not all evaluations have been completed for the line segment data detected in ST52 (ST55). If not completed (ST55-No), the process returns to ST53 and one line segment data that has not yet been evaluated. The vehicle shape degree f1 is calculated (ST54). When all the evaluations are completed (ST55-Yes), the process proceeds to ST56. A series of processes of ST53 to ST55 is an evaluation process (first vehicle shape evaluation process) for evaluating the likelihood of the vehicle shape for each piece of line segment data.

  Next, the detection target determination unit 243 selects any two line segment data sets from the line segment data detected in ST52 (ST56), and calculates the vehicle shape factor f2 (ST57). Then, it is determined whether or not the evaluation is completed for all the two line segment data sets detected in ST52 (ST58). If not completed (ST58-No), the process returns to ST56 and evaluation is still performed. A set of two line segment data that are not present is selected, and the vehicle shape degree f2 is calculated (ST57). When all the evaluations are completed (ST58-Yes), the process proceeds to ST59. The series of processes of ST56 to ST58 is an evaluation process (second vehicle shape evaluation process) for evaluating the likelihood of the vehicle shape for each combination of two line segment data.

  In S59, the detection target determination unit 243 sums all the vehicle shape degrees f1 calculated in the first vehicle shape evaluation process and all the vehicle shape degrees f2 calculated in the second vehicle shape evaluation process. The vehicle shape F is calculated. Then, the process proceeds to the flow of FIG.

When the flow proceeds to the flow of FIG. 10, the detection target determination unit 243 determines whether the vehicle shape degree F calculated in ST59 of FIG. 9 is higher than the threshold M (ST71) or lower than the threshold L (S74). .
If it is determined that the vehicle shape F is higher than the threshold M (ST71-Yes), 1 is added to the counter (ST72), and a predetermined value is added to the vehicle presence according to the counter value (ST73).
If it is determined that the vehicle shape F is lower than the threshold L (ST74-Yes), 1 is subtracted from the counter (ST75), and a predetermined value is subtracted from the vehicle presence according to the counter value (ST76).
The specific processing for adding / subtracting to the vehicle presence level has been described above, and thus the description thereof is omitted.
When the vehicle shape F is between the threshold value L and the threshold value M (ST71-No and ST74-No), the process proceeds to ST77 without adding or subtracting the counter and the vehicle presence degree.

In ST77, the detection target determination unit 243 determines whether the vehicle presence is higher than the threshold value H (ST77).
When the vehicle presence level is higher than the threshold value H (ST77-Yes), the bottom value is added to the vehicle presence level, and the process proceeds to ST81.
If the vehicle presence level is less than or equal to the threshold value H (ST77-No), it is determined whether or not the bottom value has been added to the previous time (ST79). If the previous value has been added (ST79-Yes), the vehicle is present. The bottom value is subtracted from the degree (ST80), and if it was not added last time (ST79-No), the process of ST80 is skipped and the process proceeds to ST81.
A series of these processes of ST71 to ST80 is an evaluation process (vehicle ratio evaluation process) for calculating the vehicle presence by evaluating the degree of presence of the vehicle in the change region.
Note that the vehicle presence calculated in the vehicle degree evaluation process is stored in the tracking information.

When the vehicle degree evaluation process is completed, the detection target determination unit 243 determines whether or not the process of calculating the vehicle presence degree has been performed for all the change areas stored in the tracking information (ST81).
If the process has not been performed for all the change areas (ST81-No), the process returns to ST51 in FIG. 9 to select a change area that has not yet been processed, and similarly perform the process of calculating the vehicle presence level.
If processing has been performed for all the change regions (ST81-Yes), the processing ends.

If the change area is not stored even if the tracking information is referred to in ST51, the process is terminated.
If no line segment data is detected in ST52, the process proceeds to ST59. In this case, since the vehicle shape degrees f1 and f2 are not calculated, the vehicle shape degree F is zero.
Further, when only one line segment data is detected in ST52 (for example, when one side surface of the vehicle is in a positional relationship perpendicular to the laser beam irradiation direction of the object detection sensor 2), in ST53 to ST55, After performing the first vehicle shape evaluation process for one line segment data, the second vehicle shape evaluation processing of ST56 to ST58 is skipped, and the process proceeds to ST59. In this case, the vehicle shape degree F in ST59 is the vehicle shape degree f1 for the one line segment data.

  The embodiments of the present invention have been described above by way of example, but the scope of the present invention is not limited to these embodiments, and can be changed or modified according to the purpose within the scope of the claims. is there.

  For example, in the present embodiment, the approximate length of a line segment is used as a method for evaluating the likelihood of a vehicle shape for one line segment data. However, instead of or in addition to this, another evaluation method is used. It may be adopted. As another evaluation method, for example, even if it can be approximated to a line segment, there are variations in the measurement points constituting this line segment based on the approximate line segment. The degree of variation may be calculated and used to evaluate one line segment data.

  Further, in the present embodiment, as a method for evaluating the likelihood of vehicle shape for a combination of two line segment data, the total value of the lengths of the respective line segment data and the angle formed by the two line segment data are used. However, only either of them may be used, or another evaluation method may be adopted instead of or in addition to these. As another evaluation method, for example, the degree of variation of two line segment data may be calculated as described above, and the sum of the degree of variation of the two line segment data may be used. Further, in this embodiment, the higher the evaluation is made as the angle formed by the two line segment data is closer to 90 degrees, the connection is close to 90 degrees and the positions of the two line segment data are not separated from each other. In such a case, since it is highly possible that any corner of the rectangle as the outer shape of the vehicle can be detected, higher evaluation may be performed.

Furthermore, in the present embodiment, the vehicle shape F is calculated by adding all the vehicle shape f1 and the vehicle shape f2, but using all the vehicle shape f1 and all the vehicle shape f2. What is necessary is just to calculate a vehicle shape degree, and is not restricted to this.
For example, the vehicle shape degree f1 is an evaluation of the vehicle shape likeness using one line segment data, and the vehicle shape degree f2 is an evaluation of the vehicle shape likeness using two line segment data. However, the amount of data that is the basis of evaluation is larger than the vehicle shape f1, and the reliability is high. Therefore, instead of simply summing, the vehicle shape degree F2 may be weighted higher than the vehicle shape degree f1 and summed to calculate the vehicle shape degree F. .

In the present embodiment, the vehicle presence degree is calculated from the vehicle shape degree F, and it is determined whether or not it is a detection target based on this, but the present invention is not limited to this. Whether F is a detection target or not may be determined.
For example, in the present embodiment, the vehicle shape F is calculated over a plurality of predetermined cycles to calculate the vehicle shape F, but the evaluation over the plurality of predetermined cycles is not performed, and the vehicle shape F is used as a feature amount. It may be used directly to determine whether it is a detection target together with other feature quantities.

In the present embodiment, the movement amount is used as the determination condition for determining the detection target. However, instead of or in addition to this, another method may be used.
For example, the time during which the fluctuating area stays in the alert area 4 may be calculated from the tracking information, and the determination condition may be that this staying time is a predetermined time or longer.
Further, the moving path of the change region may be detected from the tracking information, and the determination condition may be that this path is a path unique to the detection target.

  Furthermore, in the present embodiment, the detection target is a person and a vehicle, but according to the security planning of the warning area 4 monitored by the object detection sensor 2, only the vehicle may be the detection target or the vehicle may not be the detection target. Also good. In the present embodiment, when the vehicle is not set as the detection target, the vehicle presence degree calculated as the feature amount is used as the feature amount of the non-detection target, based on the comparison with the feature amount of the detection target as described above. Thus, it may be determined whether or not it is a detection target.

  In the present embodiment, four feature quantities are used. However, the present invention is not limited to this, and any one of the four feature quantities may be used, or another feature quantity may be used.

  In the present embodiment, an example in which tracking processing is performed between distance measurement data one cycle before (previous cycle with respect to the current cycle) and one cycle after (current cycle with respect to the previous cycle) has been described. Instead, a tracking process may be performed between distance measurement data within a predetermined time (for example, data acquired during one second). In this case, as described above, it is determined whether or not the distance measurement data acquired within a predetermined time in the tracking process can be associated. If there is a corresponding change area, it is determined that there is a tracking target. Thereby, even if it is a case where detection is missing by instantaneous noise, it is possible to detect the tracking target without erroneous determination.

DESCRIPTION OF SYMBOLS 1 Security system 2 Object detection sensor 21 Communication part 22 Detection part 221 Laser oscillation part 222 Scanning mirror 223 Scan control part 224 Reflected light detection part 225 Distance measurement data generation part 23 Storage part 24 Control part 241 Drive control part 242 Reference data generation part 243 detection target determination unit 3 monitoring building 4 security area 5 security device 6 monitoring center 7 communication network

Claims (5)

An object detection sensor for monitoring a warning area and detecting a detection target existing in the warning area, scanning from one end to the other end of the warning area at every measurement cycle, and measuring in each direction in the warning area A detector that generates ranging data indicating the distance to the point;
Grouping means for grouping the measurement points whose distances have changed due to the same object to be measured with respect to the current distance measurement data as compared to the past distance measurement data as a change area;
Line segment detection means for detecting line segment data from measurement points that can be approximated as line segments among the measurement points grouped as the change region;
First vehicle shape evaluation means for evaluating the likelihood of the vehicle shape for each piece of line segment data from the line segment data, and calculating a first vehicle shape degree that indicates the height of the degree of the likelihood of the vehicle shape stepwise; ,
A second vehicle shape evaluation that evaluates the likelihood of the vehicle shape for each combination of two line segment data from the line segment data, and calculates a second vehicle shape degree that indicates the height of the degree of the likelihood of the vehicle shape stepwise. Means,
A third vehicle type evaluating means for calculating a third vehicle shape of comprehensively the first car shape degree and the second car shape degree,
An object detection sensor comprising: determination means for determining whether or not the change area is the detection target based on the third vehicle shape. The determination means includes vehicle degree evaluation means for calculating a vehicle presence degree by evaluating a degree of presence of the vehicle from the third vehicle shape degree over a plurality of the measurement periods, and the change based on the vehicle presence degree. Means for determining whether a region is the detection target,
2. The object detection sensor according to claim 1, wherein the vehicle degree evaluation unit calculates the vehicle presence level higher as the number of times the third vehicle shape degree becomes higher than the first threshold increases. The vehicle degree evaluation means increases the degree of increasing the vehicle presence level when calculating the vehicle presence level higher according to the number of times than when the number of times is low. Item 3. The object detection sensor according to Item 2. The vehicle degree evaluation means adds a predetermined bottom value to the vehicle presence degree when the third vehicle shape degree is higher than a second threshold value that is larger than the first threshold value. 3. The object detection sensor according to 3. A detection unit that scans from one end to the other end of the alert area at every measurement cycle and generates distance measurement data indicating a distance to a measurement point in each direction in the alert area, and the alert area is based on the distance measurement data. A program that is executed by an object detection sensor that monitors a region and detects a detection target existing in the alert region,
The program is stored in a computer.
A process of grouping the measurement points whose distances are changed by the same object to be measured with respect to the current distance measurement data as compared with the past distance measurement data as a change area;
A process of detecting line segment data from measurement points that can be approximated as a line segment among the measurement points grouped as the change region;
A process of evaluating the vehicle shape likelihood for each line segment data from the line segment data, and calculating a first vehicle shape degree that indicates the height of the degree of the vehicle shape likelihood step by step ;
A process of evaluating the vehicle shape likelihood for each combination of two line segment data from the line segment data, and calculating a second vehicle shape degree stepwise indicating the height of the degree of the vehicle shape likelihood ;
A process of calculating a third vehicle shape of comprehensively the first car shape degree and the second car shape degree,
A process of determining whether the change area is the detection target based on the third vehicle shape degree;
A program characterized by running

Images