JP2020149158A - Crime prevention device, crime prevention method, and program - Google Patents

Abstract

To perform processing for crime prevention without allowing a person permitted to enter a crime prevention object area to be determined as a suspicious person or an intruder.SOLUTION: An object included in an acquired distance image is detected as a person based on a distance between the object and an own device and the size of the object. A motion state of a person is determined based on a change of a detection position in a distance image of an object detected as a person. Alarm is issued based on an operation mode corresponding to a motion state of an object detected as a person.SELECTED DRAWING: Figure 1

Description

The present invention relates to a security device, a security method, and a program.

When a suspicious person is detected, a security device is known that notifies the surroundings of the intrusion of the suspicious person by lighting or an alarm sound. As a related technique, Patent Document 1 describes a technique in which an illumination light is generated when the presence of a visitor is detected, a warning light is generated from a security light when the presence of a suspicious person is detected, and an alarm sound is emitted from a speaker. Is disclosed.

Further, in Patent Document 2, the first lighting device (for example, the garden light L1) corresponding to the first sensor is turned on in the first mode and the second lighting is triggered by the detection by the first sensor. A technique of a security system in which a device (for example, a garden light L2) is turned on in a second mode different from the first mode is disclosed.

Further, in Patent Document 3, the amount of illumination light from the illuminator is maximized in the farthest region of the caution area, and the amount of illumination is gradually increased according to the distance to the human body in the region closer than the farthest region. The technology of the security lighting device to reduce the number of lights is disclosed.

Further, Patent Document 4 discloses a technique for detecting motion information of a person based on distance information from a TOF (Time Of Flight) sensor.

Further, in Patent Document 5, the intrusion of an intruder into the detection area is detected, the lighting state of the light source is changed according to the distance to the intruder, and the shorter the distance to the presence position of the intruder, the lighter the light source. A technique for brightening the brightness at the time of lighting is disclosed.

JP-A-2009-80524 Japanese Unexamined Patent Publication No. 2017-16415 Japanese Unexamined Patent Publication No. 2016-6634 Japanese Unexamined Patent Publication No. 2011-27707 Japanese Unexamined Patent Publication No. 2008-310397

When determining a suspicious person or an intruder using the above-mentioned crime prevention technology or person movement detection technology, the person who is allowed to enter the crime prevention target area such as a house is not judged as a suspicious person or an intruder There is a need for a technology that can perform the processing for.

Therefore, an object of the present invention is to provide a crime prevention device, a crime prevention method, and a program that solve the above-mentioned problems.

According to the first aspect of the present invention, the security device includes a detection unit that detects that the object is a person based on the distance and size of the object included in the acquired distance image from the own device, and the above-mentioned An operation mode determination unit that determines an operation mode of the person based on a change in the detection position of the object detected as a person in the distance image, and an operation according to the operation mode of the object detected as a person. It is characterized by including an alarm unit that outputs an alarm based on a mode.

According to the second aspect of the present invention, the crime prevention method detects that the object is a person based on the distance and size of the object included in the acquired distance image from the own device, and is the person. The motion mode of the person is detected based on the change in the detection position of the object detected in the distance image, and the alarm output is performed based on the motion mode according to the motion mode of the object detected as the person. It is characterized by that.

According to a third aspect of the present invention, the program detects a person who detects that the object is a person based on the distance and size of the object included in the acquired distance image from the own device. Means, motion mode detecting means for detecting the motion mode of the person based on a change in the detection position of the object detected as the person in the distance image, depending on the motion mode of the object detected as the person. It is characterized in that it functions as an alarm processing means that outputs an alarm based on the operation mode.

According to the present invention, it is possible to perform processing for crime prevention without determining a person who is allowed to enter the crime prevention target area as a suspicious person or an intruder.

It is a figure which shows the image of the security apparatus by one Embodiment of this invention. It is a figure which shows the appearance of the security device by one Embodiment of this invention. It is a figure which shows the hardware structure of the security apparatus by one Embodiment of this invention. It is a functional block diagram of the security apparatus by one Embodiment of this invention. It is the first figure which shows the processing flow of the security apparatus by one Embodiment of this invention. It is a second figure which shows the processing flow of the security apparatus by one Embodiment of this invention. It is a figure which shows the processing flow of the security apparatus of the minimum composition by one Embodiment of this invention.

Hereinafter, a security device according to an embodiment of the present invention will be described with reference to the drawings.
FIG. 1 is a diagram showing an image of a security device.
FIG. 2 is a diagram showing the appearance of the security device.
As shown in this figure, the security device 1 is installed on an outdoor entrance door or the like. The security device 1 includes a TOF sensor 11 and a light 12 as shown in FIG. The TOF sensor 11 is a device that detects reflected light from an object of emitted light with an image sensor, calculates distance information to the object for each pixel of the image sensor, and generates a distance image. In the present embodiment, the TOF sensor 11 is installed so as to face the gate direction that separates the site of the house from the outside, and generates a distance image in the gate direction. It is assumed that there is a road at the end of the gate from the site of the house, and people and vehicles are coming and going. When the security device 1 detects a suspicious person, the light 12 is used to emit a light indicating an alarm. In the present embodiment, the house side from the gate is the crime prevention target area. The security device 1 may be configured integrally with a lighting device such as an outdoor light.

FIG. 3 is a diagram showing a hardware configuration of the security device.
As shown in FIG. 3, the security device 1 includes a CPU (Central Processing Unit) 101, a ROM (Read Only Memory) 102, a RAM (Random Access Memory) 103, and an SSD (Solid State Drive) in addition to the TOF sensor 11 and the write 12. ) 104, the communication module 105, and the like. The security device 1 may further include a speaker that emits an alarm sound.

FIG. 4 is a functional block diagram of the security device.
The CPU 101 of the security device 1 executes a security processing program. As a result, the security device 1 exerts the functions of the detection unit 111, the operation mode determination unit 112, and the alarm unit 113.

The detection unit 111 detects that the object is a person based on the distance and size of the object included in the acquired distance image from its own device. When there is a difference between the acquired distance image and the distance image acquired before a predetermined time, the detection unit 111 determines whether or not the object is a person based on the distance and size of the object from its own device.
The motion mode determination unit 112 determines the motion mode of a person based on the change in the detection position in the distance image of the object detected as a person.
The alarm unit 113 outputs an alarm based on an operation mode according to the operation mode of the object detected as a person.

The alarm unit 113 controls, for example, the amount of illumination light of the lighting device to be operated for the alarm output based on the distance of the object detected to be a person from its own device.
Alternatively, the alarm unit 113 controls the illumination direction of the operation target lighting device of the alarm output based on the position of the object detected to be a person.
Alternatively, the alarm unit 113 outputs an alarm at the time of crime prevention when the movement of the object detected as a person is a movement pattern indicating a suspicious person.
Alternatively, when the movement of the object detected as a person is not a movement pattern indicating a suspicious person, the alarm unit 113 controls the lighting state of the lighting device with the lighting state when guiding the person in a predetermined direction.

FIG. 5 is a first diagram showing a processing flow of the security device.
First, when the security device 1 is activated, the TOF sensor 11 is activated to generate a distance image. The TOF sensor 11 repeats the generation of the distance image. The detection unit 111 sequentially acquires the distance images generated by the TOF sensor 11 (step S101). The detection unit 111 compares the newly acquired distance image with the distance image previously acquired and stored in the memory or the like (step S102). The detection unit 111 determines whether or not there is a difference equal to or greater than a threshold value in the distance from the security device 1 indicated by each pixel in a predetermined range or more based on the difference between the corresponding pixels in each distance image (step S103). The range equal to or larger than the predetermined range in the distance image may be a range having a size that can be determined as a human being, or may be a range having a size that can be determined as a small animal. When the distance from the security device 1 indicated by each pixel in the range equal to or larger than the predetermined range in the previous distance image and the current distance image has a difference equal to or greater than the threshold value, the detection unit 111 determines the operation mode in the operation mode determination unit 112. To instruct.

When the distance from the security device 1 indicated by each pixel in a predetermined range or more has a difference of equal to or greater than a threshold value, the operation mode determining unit 112 determines whether the distance indicated by the distance information of the pixels in the range is short. (Step S104). When the distance indicated by the distance information is short, it indicates that the object indicated by the range is approaching. The operation mode determination unit 112 outputs the distance information of the pixels in the range to the alarm unit 113. The alarm unit 113 determines whether the distance indicated by the distance information is less than a predetermined light emitting distance (step S105). When the distance information is less than the light emitting distance, the operation mode determining unit 112 determines to emit light. The alarm unit 113 controls the light emission of the light 12 (step S106). The alarm unit 113 may control the light 12 so that the shorter the distance information, the stronger the emission intensity (illumination light amount) of the light 12.

In step S103, the detection unit 111 records the last acquired distance image when there is no difference of the distance from the security device 1 indicated by each pixel in the predetermined range or more (step S107). If the distance indicated by the distance information has not changed in step S104, the detection unit 111 determines that the distance indicated by the distance information is the distance to the road. The detection unit 111 determines whether or not the person's body is sideways and is moving left or right based on the distance image, and when it is determined that the detected person is a passerby, the detection unit 111 indicates the person. The distance may be specified based on the distance information of the pixels of the above, and this distance may be determined as the distance to the road. The detection unit 111 records the distance to the road in a storage unit such as a memory (step S108). In step S105, the alarm unit 113 determines that the light 12 is not emitted when the distance information is equal to or longer than the predetermined light emission distance, and controls to stop the issuance when the current light is emitting light. Step S109).

FIG. 6 is a second diagram showing a processing flow of the security device.
After the security device 1 controls the light emission, the detection unit 111 determines whether the range determined to have a difference in the distance image is a range having a size capable of determining a person (step S110). When the range determined to have a difference in the distance image is a range having a size that can be determined as a person, the detection unit 111 detects the range as a person. Note that the detection unit 111 has a range of a size that can be determined as a person in the distance image, and the pixels in the range that can be determined as a human shape have a distance greater than or equal to the threshold value between the previous distance image and the current distance image. If there is a difference, the range may be detected as a person. Further, when a person is detected in the distance image acquired last time, the detection unit 111 may determine whether the person is the same as the person detected last time based on the range converted in the distance image. Further, the detection unit 111 determines whether or not there are a plurality of ranges in which the difference in the distance image is detected (step S111). When there are a plurality of ranges in which the difference is detected in the distance image, the detection unit 111 determines that a plurality of people have been detected.

Then, the detection unit 111 determines whether the feature indicated by the range of one or a plurality of people in the distance image matches the feature stored in advance (step S112). For example, the detection unit 111 identifies the position of the face in the pixels in the range that can be determined to be the shape of a person in the distance image, and the distance indicated by each pixel in the range of the face is stored in advance as the facial feature information (of the face) of the person. It is determined whether or not the information matches the distance information of each pixel in the range. As a result, the detection unit 111 determines whether the detected person is a resident of the house. The detection unit 111 outputs the processing result to the alarm unit 113. Further, the detection unit 111 instructs the operation mode determination unit 112 to start determining the operation mode.

Upon receiving an instruction to start determining the operation mode, the operation mode determination unit 112 acquires each distance image acquired in the past for a predetermined time based on the current distance image. The operation mode determination unit 112 determines the movement pattern of a person based on the transition pattern of the position in the range determined as a person in the acquired plurality of continuous distance images (step S113). Further, the operation mode determination unit 112 identifies the center of the range of one or a plurality of people (step S114). The operation mode determination unit 112 outputs the processing results to the alarm unit 113.

The alarm unit 113 controls the irradiation direction of the light 12 at the center of the human range (step S115). Then, the operation mode determination unit 112 performs a process of determining whether or not the person is a suspicious person based on the movement pattern of the person (step S116). The alarm unit 113 may use the information on the distance to the road stored in step S108 to determine whether or not the person is a suspicious person. For example, when the alarm unit 113 converts a distance image in which a person cannot be detected into a distance image in which a person is detected, if the distance of the person is shorter than the distance to the road, the crime prevention target area passes through the gate from the road. It may be determined that the person is a suspicious person, not a person who has entered. When the detected person is not a suspicious person, the alarm unit 113 performs a guidance mode process for controlling the lighting state of the lighting device according to the lighting state when guiding the person in a predetermined direction (step S117). When the detected person is a suspicious person, the alarm unit 113 performs a security mode process for outputting an alarm at the time of crime prevention (step S118). The security device repeats the above process until the power is turned off.

Next, the details of the processing of the operation mode determination unit and the alarm unit will be described.
When the detected person is a resident, the operation mode determination unit 112 of the security device 1 calculates the distance amount (walking speed) per unit time indicated by the distance information of the pixels in the range indicated by the person, and averages them. The value may be determined as the walking speed of the resident. Then, this walking speed may be stored as a characteristic of the resident, and the information on the walking speed may be used for determining whether or not the person is a resident in the detection of a person in the distance image. As a result, the accuracy of determining whether the detected person is a resident may be improved.

In the guidance mode, the alarm unit 113 detects when a predetermined time has passed from the light emission control of the light in step S106 or when the resident is a resident of the house and the person is not a suspicious person. When a predetermined time elapses after the state becomes impossible, the light emission stop control may be performed. In the guidance mode, when the person is not a suspicious person and is not a resident of the house, the alarm unit 113 detects a state in which the distance indicating the person becomes longer after the distance indicating the person becomes closer to the security device 1. When the distance is long, the illuminance of the light emission may be reduced from 100% as the distance becomes long. The alarm unit 113 may be controlled so that the light emission intensity of the light 12 becomes stronger as the distance from the security device 1 to a person becomes shorter.

Further, when a plurality of people are detected in the guidance mode, the operation mode determination unit 112 detects the position of each person and determines whether or not the position deviates from the position corresponding to the range of the predetermined guidance path. .. The alarm unit 113 may determine that the mode shifts from the guidance mode to the crime prevention mode when it is determined that the number of people has decreased or when the position significantly deviates from the predetermined guidance path. When the control is changed from the guidance mode to the crime prevention mode, the alarm unit 113 may control the emission intensity of the light 12 to 100% regardless of the distance of a person.

Further, the alarm unit 113 may detect an object larger than a person at the distance of the road to be stored based on the distance image, and if it subsequently detects a person, it may determine that the person is a courier. In this case, the alarm unit 113 controls the light emission of the light 12 according to the distance of a person in the guidance mode.

If the detected person is not a resident, the motion mode determination unit 112 determines whether or not the movement of the person has stopped based on the change in the distance to the person in the distance image. When a person stops, the operation mode determination unit 112 may determine that the person has stopped suddenly and instruct the alarm unit 113 to shift to the crime prevention mode. The alarm unit 113 controls the light emission intensity of the light 12 to 100% when the mode shifts to the crime prevention mode.

Based on the change in the distance to the person in the distance image, the motion mode determination unit 112 determines whether the speed of movement of the detected person has increased sharply when the person is not a resident. When the movement of a person suddenly increases (when the reduction rate of the distance to the person per unit time suddenly increases more than a predetermined reduction rate), the operation mode determination unit 112 determines that it is a suspicious person detection. The alarm unit 113 may be instructed to shift to the crime prevention mode. The alarm unit 113 controls the light emission intensity of the light 12 to 100% when the mode shifts to the crime prevention mode.

It is assumed that the motion mode determination unit 112 moves within a predetermined range based on the position of the gate of the distance image, because the distance to the person in the distance image is substantially equal to the distance of the road to be stored. For example, when a person makes a round trip on the road, the movement of such a person can be detected. When the position of the person is moving within a predetermined range based on the position of the gate of the distance image, the operation mode determination unit 112 determines that the suspicious person is detected, and instructs the alarm unit 113 to shift to the crime prevention mode. May be good. The alarm unit 113 controls the light emission intensity of the light 12 to 100% when the mode shifts to the crime prevention mode.

It is assumed that the motion mode determination unit 112 detects a state in which the distance to a person in the distance image is substantially equal to the distance of the road to be stored and the position of the person does not change for a predetermined time or more. For example, when a person is stopped on the road, the movement of such a person can be detected. When the distance to the person is almost equal to the distance of the road to be stored and the position of the person does not change for a predetermined time or more, the operation mode determination unit 112 determines that the person is suspicious and warns the transition to the crime prevention mode. You may instruct unit 113. The alarm unit 113 controls the light emission intensity of the light 12 to 100% when the mode shifts to the crime prevention mode.

Based on the change in the distance to the person in the distance image, the motion mode determination unit 112 determines whether the speed of movement of the detected person has decreased sharply when the detected person is not a resident. When the movement of a person suddenly decreases (when the reduction rate of the distance to a person per unit time suddenly decreases more than a predetermined reduction rate), the operation mode determination unit 112 determines that it is a suspicious person detection. The alarm unit 113 may be instructed to shift to the crime prevention mode. The alarm unit 113 controls the light emission intensity of the light 12 to 100% when the mode shifts to the crime prevention mode.

It is assumed that the motion mode determination unit 112 detects a state in which the distance to the person in the distance image is substantially equal to the distance of the road to be stored and the position of the person does not change for a predetermined time or more, but immediately disappears from the position. For example, suppose a person stops on the road but immediately exits the gate. In such a case, the operation mode determination unit 112 may determine that the suspicious person is detected and instruct the alarm unit 113 to shift to the crime prevention mode. The alarm unit 113 controls the light emission intensity of the light 12 to 100% when the mode shifts to the crime prevention mode.

When the operation mode determination unit 112 detects a person in the distance image and then detects another person behind the person, the operation mode determination unit 112 may instruct the alarm unit 113 to shift to the crime prevention mode. The alarm unit 113 controls the light emission intensity of the light 12 to 100% when the mode shifts to the crime prevention mode.

When the operation mode determination unit 112 detects a suspicious person based on the distance image, cannot detect the suspicious person, and then detects a person of the same size as a suspicious person again within a predetermined time. The alarm unit 113 may be instructed to shift to the crime prevention mode. The alarm unit 113 controls the light emission intensity of the light 12 to 100% when the mode shifts to the crime prevention mode.

The motion mode determination unit 112 detects a person based on the distance image, and if the height of the person changes, determines that the person is stretching and crouching, and shifts to the crime prevention mode. May be instructed to the alarm unit 113. The alarm unit 113 controls the light emission intensity of the light 12 to 100% when the mode shifts to the crime prevention mode.

The processing mode of the operation mode determination unit 112 and the alarm unit 113 described above is an example, and a person's operation mode may be determined and alarm information may be output by other processing. The above-mentioned operation mode determination unit 112 further adds a determination as to whether or not the distance of a person is equal to or less than a predetermined distance, and instructs the alarm unit 113 to shift to the crime prevention mode when the distance is equal to or less than the predetermined distance. You may.

According to the above-mentioned processing of the crime prevention device, it is possible to perform the processing for crime prevention without determining the person who is allowed to enter the crime prevention target area as a suspicious person or an intruder. In other words, if the person detected in the crime prevention target area is a suspicious person, the alarm information is output in the crime prevention mode, and for the person who is allowed to enter the crime prevention target area, the output of the warning information is canceled in the guidance mode. Can be done. When a suspicious person is detected, even if the position is far away, if it can be detected that the person is in a predetermined position, the light 12 is turned on with a predetermined emission intensity, which can contribute to crime prevention. In addition, when the person approaching the guide destination is a predetermined person such as a resident, the light 12 can be controlled as a guide light, and a suspicious person is instilled with anxiety in order to control the light 12 in the crime prevention mode. be able to. When the mode shifts to the security mode, the security device 1 may transmit the alarm information to a predetermined notification destination such as a resident's mobile terminal or a security company's terminal. Further, the security device 1 may sound an alarm sound from an alarm means other than the light emission by the light 12, for example, a speaker.

The minimum configuration of the function of the security device is the same as that in FIG.
FIG. 7 is a diagram showing a processing flow of the security device having the minimum configuration.
The security device 1 may include at least a detection unit 111, an operation mode determination unit 112, and an alarm unit 113.
Then, the security device 1 detects that the object is a person based on the distance and size of the object included in the acquired distance image from its own device (step 401). Then, the motion mode determination unit 112 determines the motion mode of the person based on the change in the detection position in the distance image of the object detected as the person (step 402). Further, the alarm unit 113 outputs an alarm based on the operation mode according to the operation mode of the object detected as a person (step 403).

The above-mentioned security device 1 has a computer system inside. Each of the above-mentioned processes is stored in a computer-readable recording medium in the form of a program, and the above-mentioned processing is performed by reading and executing this program by a computer. Here, the computer-readable recording medium refers to a magnetic disk, a magneto-optical disk, a CD-ROM, a DVD-ROM, a semiconductor memory, or the like. Further, this computer program may be distributed to a computer via a communication line, and the computer receiving the distribution may execute the program.

The above program may be for realizing a part of the above-mentioned functions. Further, a so-called difference file (difference program) may be used, which can realize the above-mentioned functions in combination with a program already recorded in the computer system.

1 ... Security device 11 ... TOF sensor 12 ... Light 111 ... Detection unit 112 ... Operation mode determination unit 113 ... Alarm unit

Claims (8)

A detection unit that detects that the object is a person based on the distance and size of the object included in the acquired distance image from its own device.
An operation mode determination unit that determines an operation mode of the person based on a change in the detection position of the object detected as the person in the distance image.
An alarm unit that outputs an alarm based on an operation mode according to the operation mode of the object detected to be the person, and an alarm unit.
Security device equipped with. The security device according to claim 1, wherein the alarm unit controls the amount of illumination light of the lighting device to be operated for the alarm output based on the distance of the object detected to be the person from the own device. The security device according to claim 1 or 2, wherein the alarm unit controls the lighting direction of the lighting device to be operated by the alarm output based on the position of the object detected to be the person. The alarm unit according to any one of claims 1 to 3, which outputs the alarm at the time of crime prevention when the movement of the object detected to be the person is a movement pattern indicating a suspicious person. Security device. The alarm unit controls the lighting state of the lighting device by the lighting state when guiding the person in a predetermined direction when the movement of the object detected to be the person is not a movement pattern indicating a suspicious person. The security device according to any one of items 1 to 4. When there is a difference between the acquired distance image and the distance image previously acquired before a predetermined time, the detection unit determines whether the object is the person based on the distance and size of the object from its own device. The security device according to any one of claims 1 to 5, which determines whether or not the device is used. Detecting that the object is a person based on the distance and size of the object included in the acquired distance image from its own device,
The movement mode of the person is detected based on the change in the detection position in the distance image of the object detected as the person.
A crime prevention method that outputs an alarm based on an operation mode according to an operation mode of the object detected to be the person. A person detection means for detecting that an object is a person based on the distance and size of the object included in the acquired distance image from its own device.
An operation mode detecting means for detecting an operation mode of the person based on a change in the detection position of the object detected as the person in the distance image.
An alarm processing means that outputs an alarm based on an operation mode according to the operation mode of the object detected to be the person.
A program that functions as.

Images