JP2011227677A - Notification device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To propose a notification device of which the reliability is improved by reducing erroneous determination of an authorized person as a suspicious person when existence of a suspicious person and occurrence of an emergency state are automatically detected and notified.SOLUTION: The notification device detects the existence of the suspicious person in a monitoring zone and notifies it to a remote monitoring center. This device includes: a monitoring information acquisition section for acquiring monitoring information by which a moving object in the monitoring zone can be recognized; a person number detecting part by which the number of persons in the monitoring zone is detected based on the acquired monitoring information; a suspicious person detecting part by which suspicious person detection processing to detect the suspicious person existing in the monitoring zone is executed; and a notification part by which, when the suspicious person is detected, it is notified. The suspicious person detecting part executes the suspicious person detection processing when the number of persons in the monitoring zone is less than a predetermined number, and prohibits the execution of the suspicious detection processing when the number of persons in the monitoring zone is equal to or more than the predetermined number.

Description

  The present invention relates to a reporting device that detects and reports the presence of a suspicious person who may impair the maintenance of a monitored area or the occurrence of an emergency, and in particular, a technique that can reduce erroneous determination of a legitimate person as a suspicious person About.

Conventionally, as a reporting device for users in stores and offices to report the existence of suspicious persons such as robbery to the outside, remote monitoring of images captured by surveillance cameras along with emergency signals by operating so-called emergency buttons What is sent to the center is widely known.
Although this type of reporting device is based on the user's voluntary operation, it rarely reports an emergency situation accidentally, but when an emergency such as a burglary occurs, the operation of the emergency button is regulated by the burglary. The user may be upset and forget to operate the emergency button, and there is a problem that the certainty of the notification is lacking.

The present applicant has filed a patent application according to Patent Document 1 as a countermeasure to such a problem.
In this patent document 1, a phrase whose pronunciation is predicted when an emergency occurs and a sound pressure level corresponding to the danger level of the phrase are stored in association with each other and input to a microphone in a store or the like. An informing device is disclosed that determines whether or not an emergency has occurred based on a combination of a phrase and sound pressure by monitoring the phrase and sound pressure in the voice.

Japanese Patent Laid-Open No. 2006-11872

  According to the reporting device proposed in Patent Document 1, it is possible to report the occurrence of an emergency such as a burglary to the outside while eliminating the need for an emergency button operation by the user. Both convenience and certainty of reporting can be achieved. Can be improved.

By the way, in general, suspicious persons such as robbers aim to seize money by conquering the surveillance area, and often appear aiming at a time zone in which there are few people who have a high probability of success.
On the other hand, since the notification device of Patent Document 1 is configured to constantly collect the sound of the monitoring area and determine the presence of a suspicious person or the occurrence of an emergency, the busy time is low in the possibility of a suspicious person appearing. Even if it is a belt, it will be continuously monitored.

  However, in such a busy time period, the environmental sound due to the hustle is generally large, and there is a risk that the suspicious person may be erroneously determined even in a situation where only a legitimate person exists in the notification device of Patent Document 1. . For example, it may be erroneously determined as an emergency by a message that flows at a relatively large volume, such as a noise caused by a large number of visitors or a broadcast in a store.

  The present invention has been made in view of the above problems, and reduces the possibility of misjudging a legitimate person as a suspicious person when automatically detecting and reporting the presence of a suspicious person or the occurrence of an emergency. The purpose is to propose a reporting device with improved reliability.

  In order to achieve the above object, a notification device according to the present invention is a notification device that detects the presence of a suspicious person in a monitoring area and remotely notifies the monitoring apparatus, and acquires monitoring information that can be recognized by a moving object in the monitoring area. A monitoring information acquisition unit, a number detection unit for detecting the number of people in the monitoring area based on the acquired monitoring information, and a suspicious person for executing a suspicious person detection process for detecting a suspicious person existing in the monitoring area A suspicious person detection unit that performs a suspicious person detection process when the number of persons in the monitoring area is less than a predetermined number. The suspicious person detection process is prohibited when the number of persons in the monitoring area is a predetermined number or more.

  In such a configuration, the notification device acquires monitoring information that can be recognized by a moving object in the monitoring area, detects the number of people in the monitoring area from the monitoring information, and monitors the change in the number of people. If the number is less than the predetermined number, a suspicious person detection process for detecting a suspicious person existing in the monitoring area is executed to report the presence of the suspicious person. In addition, if the number of people in the monitoring area is equal to or greater than the predetermined number, the suspicious person detection process is prohibited from being executed.

  According to this configuration, since the suspicious person detection process is executed when the number of persons in the monitoring area is less than the predetermined number, the suspicious person detection process is executed when the monitoring area such as a store or an office is in a busy time zone. Is prohibited, and erroneous determination caused by a noise source such as a hustle and bustle can be suppressed to improve detection accuracy of a suspicious person.

  Further, in the notification device of the present invention, the monitoring information acquisition unit acquires image data of the monitoring area and / or distance data to an object in the monitoring area as the monitoring information, and the number detection unit A moving object in the monitoring area may be detected from the monitoring information, and the number of persons in the monitoring area may be detected based on the number of moving objects.

  Accordingly, the number of moving objects is detected based on the image data and / or the distance data, and the number of moving objects is detected as the number of people in the monitoring area, so that the number of persons can be easily detected by a simple process.

  Further, the notification device of the present invention further includes a storage unit that stores a preset suspicious person characteristic as a suspicious pattern, and the suspicious person detection unit detects a moving object from the monitoring information as the suspicious person detection process. If the extracted moving object has an appearance or behavior that matches a preset suspicious pattern, it may be determined as a suspicious person.

  Thus, it is only necessary to extract a moving object in a monitoring area where the number of persons is less than a predetermined number, and determine whether or not the moving object is a human body, that is, a suspicious person. The suspicious person can be determined. As a process for determining whether or not the moving object is a suspicious person, various determination criteria such as, for example, the existence position, the movement locus, the staying time, and the fact that each part of the face is not concealed can be used.

  The notification device of the present invention further includes a voice acquisition unit that acquires voice data of the monitoring area, and the suspicious person detection unit performs voice recognition processing on the voice data as the suspicious person detection process, and the voice data A suspicious person existing in the monitoring area may be detected based on a character string included in the.

  As a result, it is only necessary to perform voice recognition processing on voice data acquired in a monitoring area where the number of persons is less than a predetermined number, and determine whether or not there is a suspicious person from the extracted character string. Suspicious person determination can be easily performed by a simple process while suppressing the above. As processing for determining a suspicious person from a character string included in voice data, for example, various determination criteria such as presence / absence of a dangerous word / phrase stored in advance or determination based on a combination of a word / sound pressure can be used.

  Further, in the notification device of the present invention, the monitoring area is divided into a plurality of divided areas, and the number of persons detecting unit detects the number of persons in a specific divided area set in advance among the plurality of divided areas, and the suspicious area The person detection unit executes the suspicious person detection process when the number of persons in the specific divisional area is less than a predetermined number, and detects the suspicious person when the number of persons in the specific divisional area is equal to or larger than the predetermined number. Execution of processing may be prohibited. In this case, the notification device includes a storage unit that stores area information obtained by dividing the monitoring area into a plurality of divided areas.

  In this way, it is possible to monitor the number of persons in a specific area where the traffic of outsiders is free, and to monitor suspicious persons when this specific area is particularly quiet. It is possible to effectively monitor the appearance of a suspicious person while preventing erroneous determination.

  Further, in the notification device of the present invention, a storage unit that stores the number of users in the monitoring area in advance, and the suspicious person detection process is performed by multiplying the number of users in the monitoring area by a predetermined coefficient. And a threshold value determination unit that determines a predetermined number used for the determination.

  This makes it possible to calculate the number of people allowed to perform suspicious person detection processing dynamically from the number of users that differ depending on the monitored area, making it easy to set up when the reporting device is installed, and forgetting to set up when installing A decrease in security can be prevented.

  According to the present invention, since the suspicious person detection process is executed when the number of persons in the monitoring area is smaller than the predetermined number, the suspicious person detection process is executed when the monitoring area such as a store or an office is in a busy time zone. Is prohibited, and erroneous determination caused by a noise source such as a hustle and bustle can be suppressed to improve detection accuracy of a suspicious person.

It is the schematic which shows the whole structure of the notification system by the notification apparatus of this invention. It is a block diagram which shows the structure of the notification apparatus of this invention. It is a flowchart which shows the monitoring process of the notification apparatus of this invention.

Embodiments of the present invention will be specifically described below with reference to the drawings.
In the present embodiment, a case where an area in which a store area and an office area such as a financial institution or a store are provided as a monitoring area is monitored is illustrated, but the scope of the present invention is not limited to this.

FIG. 1 is a configuration diagram showing a notification system 1 using the notification device of the present invention.
FIG. 1 shows a monitoring area 2 composed of a store area 2a and an office area 2b having an entrance to the outside, a monitoring camera 3 installed in the monitoring area 2, and a notification device 4 to which the monitoring camera 3 is connected. The relationship is schematically shown on the plan view.

  Here, in the present embodiment, the store area 2a is a region where a store clerk who is a user of the monitoring area 2 and a customer who is a visitor from the outside can exist together, and the office area 2b is This is an area where only the store clerk can exist. As shown in FIG. 1, in the store area 2a, a store entrance 21 is provided as an entrance to the outside, and a common entrance 22 is provided between the store area 2a and the office area 2b. In addition, the office area 2b is provided with an office entrance 23 as an entrance to the outside in addition to the above-described common port 22.

In the example of FIG. 1, the monitoring camera 3 is installed at a position where the entire store area 2a can be imaged, and the monitoring camera 3 is also installed at a position where the entire office area 2b can be imaged. The monitoring camera 3 and the notification device 4 are connected by a LAN (not shown), and the notification device 4 is further connected to a remote monitoring center 5 via a communication line network 6.
Although not particularly shown, a security sensor such as a heat ray sensor or an open / close sensor or an emergency button may be connected to the LAN connecting the notification device 4 and the monitoring camera 3.

  The monitoring camera 3 is a monitoring unit that generates monitoring information of the monitoring area 2. The surveillance camera 3 stores a camera number for identifying each surveillance camera 3 in advance, and an image signal input from the image sensor is transmitted at a predetermined photographing interval (for example, a cycle of 0.2 seconds, that is, 5 fps (frame / second)). Then, the image data is converted into a digital signal, compression-encoded, and image data compliant with a predetermined standard (for example, JPEG standard) is generated. The generated image data is output to the LAN and transmitted to the notification device 4. The image data transmitted at this time includes a camera number. Note that the surveillance camera 3 may compress and encode the audio signal input from the microphone to generate audio data, multiplex the image data and audio data, and output the multiplexed audio data to the LAN.

The notification device 4 receives the image data input from the monitoring camera 3 and detects the number of people present in the monitoring area 2 from this image data. Then, if the number of persons in the monitoring area 2 is less than a preset threshold value, a suspicious person detection process for determining whether or not there is a suspicious person in the monitoring area 2 is executed. As a result of the suspicious person detection process, when it is determined that a suspicious person exists in the monitoring area 2, a suspicious person detection signal is sent to the monitoring center 5 through the communication network 6 and received from the monitoring camera 3. The image data is transmitted to the monitoring center 5. The image data transmission process is continuously performed until a stop input is received from a user of the monitoring center 5 or the monitoring area 2.
On the other hand, the reporting device 4 prohibits execution of the suspicious detection process if the number of persons in the monitoring area 2 is equal to or greater than the threshold value.

  In this embodiment, a suspicious person refers to a person who may impair the maintenance of the monitoring area 2, and specifically, a person whose appearance and behavior match a predetermined suspicious pattern. For example, people who are extremely depressed or wearing sunglasses or masks to hide part of their face and do not know the human phase, who stepped on or stayed in a specific area, who shouted loudly or others around them Someone who screams.

  The monitoring center 5 is a facility provided with a center device 51 operated by a security company or the like. The center device 51 is composed of one or a plurality of computers, and realizes the function of the monitoring center 5 related to the present invention. In the monitoring center 5, various devices are controlled by the center device 51, the suspicious person detection signal received from the reporting device 4 is recorded, and the image data transmitted from the reporting device 4 is displayed on the display 52, so that the monitoring staff Monitors a plurality of monitoring areas 2 to be monitored.

<Notification device>
Next, the configuration of the notification device 4 will be described with reference to FIG. FIG. 2 is a block diagram showing the configuration of the notification device 4.
The reporting device 4 is installed in the monitoring area 2 (in the present embodiment, the office area 2b) so as to be able to communicate with the monitoring camera 3 and the monitoring center 5.

  The reporting device 4 includes a monitoring information acquisition unit 41 as a communication I / F (interface) connected to the LAN, a communication unit 42 connected to the communication network 6, and an operation unit 43 on which a user performs an input operation. The storage unit 44 includes an HDD, a memory, and the like, and the control unit 45 includes an MPU, a microcomputer, and the like and controls each unit.

The monitoring information acquisition unit 41 is connected to the monitoring camera 3, receives the camera number and image data output from the monitoring camera 3 as monitoring information of the monitoring area 2, and outputs these data to the control unit 45.
The communication unit 42 is connected to the center device 51 via the communication line network 6 and communicates with the monitoring center 5. When the control unit 45 determines that there is a suspicious person in the monitoring area 2, the communication unit 42 transmits to the monitoring center 5 a suspicious person detection signal including its own address information and image data captured by the monitoring camera 3. To do.
The operation unit 43 is a switch capable of inputting setting information such as the number of users.

  The storage unit 44 is configured by a ROM, a RAM, or an HDD, stores address information for identifying itself, various programs, and the like, and further stores various information for operating the reporting device 4. Specifically, the storage unit 44 constitutes reference data serving as background information for extracting a moving object, threshold information for determining whether or not to perform suspicious person detection processing, and the monitoring area 2. Area information indicating area information is stored. Further, the storage unit 44 stores image data for a past predetermined time output from the monitoring camera 3.

Here, the reference data stored in the storage unit 44 is comparison reference information used for extracting a moving object in the monitoring area 2 in comparison with the current image data in the number detection process described later. It is the image data acquired by imaging the monitoring area 2 at the time of unattended.
The threshold information is information that defines the reference number of persons in the monitoring area 2 that is allowed to execute the suspicious person detection process, and is used when the reporting device 4 is installed or when the security planning of the monitoring area 2 is changed. Is a value calculated by.
Further, the area information is preset with information on the monitoring camera 3 and the entrances 21, 22, 23 corresponding to the store area 2 a and the office area 2 b constituting the monitoring area 2 and various areas set in the image data. Information.

  The control unit 45 includes a microcomputer including a CPU, a ROM, a RAM, and the like and peripheral circuits thereof, and controls each unit described above. For this purpose, the control unit 45 executes a suspicious person detection process and a number detection unit 451 for detecting the number of persons in the monitoring area 2 as functional modules realized by the microcomputer and a computer program executed on the microcomputer. A suspicious person detection unit 452, a threshold value determination unit 453 that determines a threshold value used for determination for executing the suspicious person detection process, and a communication control unit 454 that controls communication processing by the communication unit 42. Yes.

  The number detection unit 451 performs image processing on the image data acquired by the monitoring information acquisition unit 41 (image data input from the monitoring camera 3) to detect a moving object, and exists in the monitoring area 2 from the number of moving objects. Detect the number of people who are. For example, the number of people detected from the image data by the number of people detection unit 451 is the number of variable regions in the image data. The number-of-people detection unit 451 compares the reference data (image data when nobody is present) stored in the storage unit 44 with the current image data input from the monitoring information acquisition unit 41, and the variation in luminance value is greater than or equal to a predetermined value. In addition to extracting the fluctuating pixels, a group of substantially continuous fluctuating pixels is grouped as one fluctuating region, and each fluctuating region is labeled with a unique label. Then, the number of fluctuation regions is detected as the number of detected people.

  The number-of-people detection unit 451 performs the above-described processing on the image data captured by each monitoring camera 3, calculates the number of people for each image data, and thereby the number of people in the store area 2a and the number of people in the office area 2b based on the camera number. , And the total number of people in the monitoring area 2 is calculated.

  The detection of the number of people is not limited to this, and the moving region component may be extracted by tracking the fluctuation region in the image data of a predetermined predetermined frame and used for detecting the number of people. For example, when two people pass each other, there are cases where two people overlap each other in one image data. In such a case, the fluctuation region is tracked with a plurality of continuous image data, and a vector By extracting the components, what was detected as one fluctuation area (= 1 person) in a certain image should actually be detected as two fluctuation areas (= 2 persons) I understand. In addition, when the variable regions are grouped and labeled, a variable region model having a human body shape feature may be used. Thereby, it is possible to prevent a moving object (such as a pet) other than a person from being counted as the number of detected persons.

The suspicious person detection unit 452 executes suspicious person detection processing according to the number of persons output from the number of persons detection part 451. In the present embodiment, the suspicious person detection unit 452 determines whether or not there is a suspicious person in the monitoring area 2 if the number of persons in the monitoring area 2 is less than the reference number of persons defined in the threshold information of the storage unit 44. The suspicious person detection process to be determined is executed, and the suspicious person detection process is prohibited from being executed if the number of reference persons is exceeded.
As a result, when the store area 2a or the office area 2b is in a busy time period, the execution of the suspicious person detection process is prohibited, and it is erroneously determined that a legitimate person is a suspicious person due to a crowd of crowds. The detection accuracy is improved by performing detection processing only during a period when the possibility of appearance of a suspicious person is high.

  However, the present invention is not limited thereto, and the number of people in the store area 2a detected from the image data in the store area 2a is compared with the reference number. If the number of people in the store area 2a is less than the reference number, a suspicious person detection process is executed. If the number is greater than or equal to the number of persons, the execution of the suspicious person detection process may be prohibited. By doing in this way, when the inside of the store area 2a where the visitor enters and exits is particularly quiet, it is possible to monitor the presence or absence of a suspicious person, and effectively monitor the suspicious person appearing in the store area 2a. be able to.

  The detection method of the suspicious person by the suspicious person detection processing executed by the suspicious person detection unit 452 is not limited to the image, and any method may be used. However, in this embodiment, the monitoring information acquisition unit 41 is used as the image characteristic of the suspicious person. It is detected that there is a person who conceals each part (eyes, nose, mouth), which is a facial feature part whose human phase can be distinguished, in the image data acquired in (cannot detect each part of the face). The suspicious person detecting unit 452 can use various known methods for detecting a facial feature portion in the face area.

  Specifically, the suspicious person detection unit 452 extracts, as suspicious person detection processing, an edge component of the fluctuation region labeled by the number of persons detection unit 451, and an elliptical edge that approximates the contour shape of the face in the edge image data. When the distribution is detected and the color information in the elliptical region surrounded by the edge distribution includes a skin color (for example, a pixel whose hue (H) component is 0 to 30 in the HSV color system), the elliptical region is determined. Extract as a face area. Then, it is determined whether or not an edge (for example, a horizontal edge such as an eye or a mouth) that becomes a facial feature portion appears in the facial region, and a facial region where such a facial feature portion cannot be detected by a suspicious person. Judge that there is.

  In addition, the suspicious person detection unit 452 individually detects a face area wearing sunglasses and a mask from the luminance distribution of the face area. In the face area where the sunglasses and the mask are worn, dark pixels concentrate on the upper side, so that the luminance center of gravity in the vertical direction is lower than the center. Further, the luminance value for each line viewed in the horizontal direction is small, whereas the variance of the luminance value for each line viewed in the vertical direction is large. Furthermore, there is a property that the luminance value is extremely low depending on the part of the sunglasses. Therefore, the Y coordinate of the luminance center of gravity of the face area is below the center of the face area, and the ratio between the standard deviation in the X direction and the standard deviation in the Y direction is larger than a predetermined threshold, and the pixel having a low luminance value Is equal to or greater than the threshold value, it is determined that the image in the face area is an image of a face wearing sunglasses and a mask, and the face area is determined to be a suspicious person.

The suspicious person detection process is not limited to this, and it is only necessary to determine that the appearance or behavior of a person matches a preset suspicious pattern.
For example, it may be determined whether the person is a suspicious person based on the position of the person. In this case, an important monitoring area is set in the image data, and the person existing in the important monitoring area is determined as the suspicious person. Further, it may be determined whether the person is a suspicious person based on the movement locus of the person. In this case, for example, the movement locus of the person (moving object) detected by the tracking processing of the labeled fluctuation region is set in the image data. The person may be determined to be a suspicious person when it is linear from the area of the store entrance 21 to an important monitoring area such as a cash register. Further, it may be determined whether or not the person is a suspicious person based on the residence time of the person. In this case, the person who exists in the important monitoring area set in the store area 2a or in the image data for a predetermined time is determined as the suspicious person. To do. In addition, a person who matches the suspicious pattern can be extracted by various known methods.

The threshold value determination unit 453 calculates the reference number of people in the monitoring area 2 that allows execution of the suspicious person detection process, and stores the calculated number in the threshold value information of the storage unit 44. When the number of users (the number of shop assistants) in the monitoring area 2 is input from the operation unit 43 and stored in the storage unit 44, the threshold value determination unit 453 multiplies this number by a predetermined coefficient (for example, 2). In addition, the standard number of people is calculated.
The reference number is a reference value for the number of people in the area regarding whether or not to perform suspicious person determination processing, and is set to a large value in view of the fact that suspicious persons such as robbers appear when there are few people as described above. That is not preferable. Therefore, in the present embodiment, the suspicious person detection process is executed when the number of shoppers who are users of the monitoring area 2 is less than the same number by making the reference number of persons twice the number of shop assistants. Like to do.
Needless to say, the setting of the reference number of people may be directly input from the operation unit 43 and stored in the storage unit 44 without depending on the threshold value determination unit 453.

  The communication control unit 454 transmits a suspicious person detection signal to the monitoring center 5 when the suspicious person detection unit 452 determines that there is a suspicious person in the monitoring area 2. The suspicious person detection signal includes the address information of the reporting device 4. Further, when the presence of a suspicious person is determined, the communication control unit 454 starts transmitting image data input from the monitoring camera 3 to the monitoring center 5. The camera number and the address information of the reporting device 4 are added as header information to the transmitted image data, and the current image data is continuously transmitted each time it is input from the monitoring camera 3. When the communication control unit 454 receives either a stop signal due to an image transmission stop input to the operation unit 43 or an image transmission stop signal transmitted from the monitoring center 5, the communication control unit 454 performs image data transmission processing to the monitoring center 5. To stop.

<Description of operation>
About the notification system 1 comprised as mentioned above, the operation | movement is demonstrated with reference to drawings. Here, the operation related to the reporting device 4 will be mainly described. FIG. 3 is a flowchart showing the operation of the monitoring program repeatedly executed by the reporting device 4.

  The notification device 4 accepts input of image data captured by the monitoring camera 3 from the monitoring information acquisition unit 41 at a predetermined cycle (for example, a cycle of 0.2 seconds, that is, 5 fps (frames / second)), and stores it in the storage unit 44. . When the monitoring program shown in FIG. 3 is executed, the number-of-people detection unit 451 detects the number of people in the monitoring area 2 by extracting the number of moving objects from the current image data captured by each monitoring camera 3 (Step S1). ST1). Then, the suspicious person detection unit 452 reads and compares the number of people in the monitoring area 2 detected by the number of people detection unit 451 and the reference number of persons stored in the threshold information of the storage unit 44, and the number of people in the monitoring area 2 is determined. It is determined whether the number is smaller than the reference number (step ST2).

  If the number of people in the monitoring area 2 is smaller than the reference number (step ST2-Yes), the suspicious person detection unit 452 executes a suspicious person detection process (step ST3). In such a suspicious person detection process, it is determined that the appearance or behavior of a person in the monitoring area 2 matches a preset suspicious pattern to determine the presence of a suspicious person. In the present embodiment, as image characteristics of a suspicious person, each part (eyes, nose, mouth), which is a facial feature part whose human phase can be identified, is concealed in the image data acquired by the monitoring information acquisition unit 41 (face A person is detected as a suspicious person.

  As a result of the suspicious person detection process, when it is determined that the suspicious person is present in the monitoring area 2 (step ST4-Yes), the communication control unit 454 transmits a suspicious person detection signal to the monitoring center 5 (step ST5). Transmission of image data input from the camera 3 is started to the monitoring center 5 (step ST6). The transmission of the image data is continuously performed until an image transmission stop signal is input from the operation unit 43 or the monitoring center 5. When an image transmission stop signal is input from the operation unit 43 or from the monitoring center 5 (step ST7—Yes), the transmission of image data to the monitoring center 5 is stopped and the process is terminated (step ST8).

On the other hand, if the number of people in the monitoring area 2 is greater than or equal to the reference number in step ST2 (step ST2-No), the execution of the suspicious person detection process is prohibited by skipping the processes in steps ST3 to ST8. ing. Therefore, when the number of people in the monitoring area 2 is equal to or more than the reference number, it is possible to prevent the determination that there is a suspicious person in the monitoring area 2 and to suppress erroneous determination.
As described above, the number of people in the store area 2a as a specific division area is detected in the number of people detection process in step ST1, and the number of people in the store area 2a is compared with the reference number of people in step ST2. If the number of persons is smaller than the reference number, the suspicious person detection process may be executed, and if the number is larger than the reference number, the execution of the suspicious person detection process may be prohibited. By doing in this way, a suspicious person can be monitored when the inside of the store area 2a in which a visitor frequently enters and leaves is particularly quiet.

  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 example in which the number of persons detection unit 451 detects the number of persons in the monitoring area 2 based on the image data has been described, but the present invention is not limited to this. In other words, any method may be used as long as the number of people in the monitoring area 2 can be determined. For example, a passage detection sensor capable of detecting the direction in which the person has passed at each of the entrances 21, 22, and 23 in the monitoring area 2. The person who entered the store area 2a and the office area 2b by inputting the detection signal of this traffic detection sensor to the monitoring information acquisition unit 41 as monitoring information and detecting the number of people passing through each doorway and the direction of traffic The difference between the number of persons and the number of persons leaving the field may be obtained to calculate the number of persons present. As such a traffic detection sensor, various methods such as a method of providing a plurality of infrared sensors along the traffic direction and detecting the traffic direction according to the detection order are known.

  Further, in the suspicious person detection process, the level for determining as a suspicious person may be varied depending on the degree that the number of people in the monitoring area 2 is smaller than the reference number. That is, when the second reference number smaller than the reference number is set in the threshold information, and the number of people in the monitoring area 2 is less than the reference number and greater than or equal to the second reference number, an obvious suspicious person (for example, sunglasses and the like) A person who is less likely to be a suspicious person in addition to an obvious suspicious person (for example, a person in the face area wearing a mask) is detected. Even a person who does not have an edge as a facial feature in the face area is determined as a suspicious person.

  Moreover, although this embodiment demonstrated the example which the monitoring information acquisition part 41 acquires image data as monitoring information of the monitoring area 2, and performs a person number detection and a suspicious person detection using this image data, it is not limited to this. . That is, the monitoring information is not limited to image data, and may be information that can identify and recognize a moving object in the monitoring area 2. For example, instead of the monitoring camera 3 or in addition to the monitoring camera 3, reflected light reflected by an object on the optical path by performing spatial scanning at a predetermined period while irradiating a laser beam in the monitoring area 2 as monitoring means A laser scanning distance sensor that detects the position of the object by obtaining the distance data to the object existing in the area and the irradiation direction by receiving the light may be used.

  In this case, the position (distance data and irradiation direction) of the object in the monitoring area 2 is input to the monitoring information acquisition unit 41 from the laser scanning distance sensor as monitoring information. As the reference data, the position of the existing object measured during unmanned operation is stored, and the variable object is extracted and labeled from the difference between the reference data and the position of the object in the current monitoring area 2, and the number of people is detected by the number of labels. The

  In the suspicious person detection process, it may be determined whether or not the person is a suspicious person based on the existence position, moving locus, and residence time of the labeled variable object (person) as in the above-described embodiment. That is, an important monitoring area is set within the scanning range of the laser scanning type distance sensor, and the entrance of a person into this area is monitored, and a predetermined suspicious pattern (entering into the important monitoring area, moving to the important monitoring area) A person who matches the trajectory and the residence time in the important monitoring area) is determined as a suspicious person.

In addition to the above-described embodiment, in the suspicious person detection process, a suspicious person may be detected based on voice data input from a microphone installed in the monitoring area 2 (the store area 2a and the office area 2b).
In this case, the monitoring information acquisition unit 41 also functions as a voice acquisition unit that acquires voice data collected by the microphone, and the suspicious person detection unit 452 performs voice recognition processing on the voice data as a suspicious person detection process. It is determined whether or not there is a suspicious person in the monitoring area based on the character string included in the data. Specifically, the suspicious person detection unit 452 extracts a voice feature amount from the voice data, cuts out the voice by pattern matching and replaces it with a character string of a code, and an emergency has occurred in which this character string is stored in the storage unit in advance. It is determined that there is a suspicious person when it coincides with a dangerous phrase (eg, “help” or “make money”) whose pronunciation is sometimes predicted. In this case, the dangerous word / phrase is associated with the determination sound pressure, and the presence of the suspicious person is determined when the sound data of the monitoring area 2 includes the dangerous word / phrase and the sound pressure of the word / phrase is equal to or higher than the determination sound pressure. Is preferred.

DESCRIPTION OF SYMBOLS 1 Reporting system 2 Monitoring area 2a Store area 2b Office area 3 Monitoring camera 4 Notification apparatus 41 Monitoring information acquisition part 42 Communication part 43 Operation part 44 Storage part 45 Control part 451 Number of persons detection part 452 Suspicious person detection part 453 Threshold determination Unit 454 Communication control unit 5 Monitoring center 51 Center device 52 Display 6 Communication network

Claims (6)

A reporting device that detects the presence of a suspicious person in a monitoring area and remotely reports it,
A monitoring information acquisition unit for acquiring monitoring information of a moving object in the monitoring area;
Based on the acquired monitoring information, a number detection unit that detects the number of people in the monitoring area,
A suspicious person detecting unit for executing a suspicious person detecting process for detecting a suspicious person existing in the monitoring area;
A reporting unit for reporting when the suspicious person is detected;
With
The suspicious person detection unit executes the suspicious person detection process when the number of persons in the monitoring area is less than a predetermined number, and the suspicious person detection process of the suspicious person detection process when the number of persons in the monitoring area is a predetermined number or more. A reporting device characterized by prohibiting execution.
The monitoring information acquisition unit acquires image data of the monitoring area and / or distance data to an object in the monitoring area as the monitoring information,
The notification device according to claim 1, wherein the number detection unit detects a moving object in the monitoring area from the monitoring information and detects the number of persons in the monitoring area based on the number of moving objects.
Furthermore, a storage unit for storing the characteristics of the suspicious person set in advance as a suspicious pattern is provided,
The suspicious person detection unit
The notification device according to claim 1, wherein as the suspicious person detection process, a moving object is extracted from the monitoring information, and the suspicious person is determined to be a suspicious person when an appearance or behavior of the extracted moving object matches a preset suspicious pattern.
Furthermore, a voice acquisition unit that acquires voice data of the monitoring area is provided,
The suspicious person detection unit
4. The notification device according to claim 1, wherein the voice data is subjected to voice recognition processing as the suspicious person detection processing, and a suspicious person existing in the monitoring area is detected based on a character string included in the voice data.
The monitoring area is divided into a plurality of divided areas,
The number of persons detection unit detects the number of persons in a specific division area set in advance among the plurality of division areas,
The suspicious person detection unit performs the suspicious person detection process when the number of persons in the specific divisional area is less than a predetermined number, and when the number of persons in the specific divisional area is equal to or larger than the predetermined number, The notification device according to claim 1, wherein execution of the person detection process is prohibited.
further,
A storage unit for previously storing the number of users in the monitoring area;
A threshold value determination unit for determining a predetermined number used for determination for executing the suspicious person detection process by multiplying a predetermined coefficient by the number of users in the monitoring area;
The notification device according to any one of claims 1 to 5.

Images