KR20110073902A - Context cognition surveillance method for multi-sensor and/or node collaboration - Google Patents

Abstract

PURPOSE: A sensor node state recognizing unmanned monitoring method for a multi sensor is provided to transmit minimum sensor node state information by improving an unmanned monitoring system. CONSTITUTION: An object type by sensor data collection of a multi sensor is determined(S1). State recognition information is formed for tracing a mobile object by collaboration of a neighbor monitoring node(S2). An abnormal behavior pattern of an object is extracted about an interest region. State recognizing information of the monitoring node is spread to the neighbor monitoring node(S4).

Description

Context cognition surveillance method for multi-sensor and / or node collaboration}

The present invention relates to a sensor node unattended monitoring method, and more specifically, to multiple sensors (vibration, infrared, radar, magnetic field, temperature, humidity, illuminance, etc.) including low power characteristics and self-learning functionalized images and sounds that a sensor node should have. It is a situational engine that fuses the information collected from the system and recognizes the occurrence of events by itself, and uses the environmental information collected through the multi-sensor fusion, and by using the environmental information collected through the multi-sensor fusion, the moving object by the collaboration and fusion between sensor nodes with the static recognition of the unmanned monitoring objects. Sensor node situation or unmanned monitoring for multi-sensor and node collaboration that provides more active situation or unmanned surveillance service through current-based situation response capability by inferring current situation and predicting future situation by recognizing spatial and temporal change It is about a method.

A conventional unmanned monitoring sensor node is disclosed in Patent Registration No. 874499, "Image acquisition capable sensor network node and image acquisition and transmission method using the same".

Patent No. 844499, "Image Acquisition Sensor Network Node and Image Acquisition and Transmission Method Using Them", uses an infrared sensor and the like to transmit a photographing command to a camera module according to a sensor detection change, and remotely via a wireless transmission / reception module. It is possible to detect objects when blind spots occur in the sensor and camera by transmitting sensing data to the central collection device, and there is a defect that requires astronomical installation and maintenance costs when applying broadband.

In the conventional unmanned surveillance system, the monitoring data is collected from the sensor nodes by the server to analyze the status, and the appropriate situation level is determined from the analyzed result to re-send the alarm action to the alarm node.

This server-oriented bidirectional monitoring system can cause excessive traffic and communication bottlenecks in the USN-based unmanned surveillance network, and there is a possibility of network failure when transmitting situation data from many sensor nodes, as well as periodic / aperiodic sensor data. Since excessive power consumption for transmission and throughput may be reduced due to excessive communication, an increase in bandwidth and construction cost is required to improve this situation, requiring high cost and high difficulty design.

On the other hand, the wireless sensor network is a network in which a large number of sensor nodes operate in cooperation, and the effect of a specific sensor node on the entire network has to be minimized. However, in the server-centric unmanned surveillance network, server instability can cause a huge impact on the entire network, it is difficult to reflect the characteristics of the wireless sensor network and there is a problem that is not easy to implement the application service.

Therefore, the inventors of the present invention set an arbitrary threshold value for detecting the movement of an object for configuring a collaborative unmanned surveillance network, and detects the object when the threshold value is over the threshold and propagates the sensing information to the neighbor node. Using the received information and the pattern table, the path of the object is predicted, and if the movement of the object is predicted toward itself, the threshold is adjusted to increase the detection probability of the object, thereby minimizing the missing-rate. Based on the fact that the wireless network-based collaborative object tracking can be done, the research has been completed.

The present invention has been made in view of the above-mentioned situation to solve various defects and problems caused in the conventional unmanned surveillance system, the object of which is to provide a simple sensor object-oriented multi-sensor and fusion object situation information of the existing heterogeneous sensor information; It provides a sensor node context awareness unattended monitoring method for multi-sensor and node collaboration that improves with collaboration-based object context information.

Another object of the present invention is a multi-sensor to tune the self-diagnostic and adaptive sensor parameters to determine the type and situation of the object by the step-by-step sensor data collection based on the multi-sensor to minimize the situational awareness information of the unmanned sensor node And it provides a sensor node situation awareness unattended monitoring method for node collaboration.

Another object of the present invention is to improve the conventional server-centered redundant unmanned surveillance system to a sensor node-based unmanned unmanned surveillance system to transmit the minimum sensor node status information to the server sensor node for multi-sensor and node collaboration It is to provide an unattended monitoring method for situations.

It is still another object of the present invention to unify the unmanned surveillance by unifying the object type and situation by multiple sensors of the sensor node, spatial and temporal situation by the sensor node collaboration, neighbor node propagation and server transmission of minimal situation information It is to provide a sensor node context-aware unmanned monitoring method for multi-sensor and node collaboration to build a system.

In order to achieve the above object, the present invention provides a sensor node context-aware unattended monitoring method for multi-sensor and node collaboration. ; A situation information forming step (S2) of forming situation or information for continuous tracking of a moving object by cooperation with a neighbor monitoring node; A pattern extraction step (S3) of extracting an intensive action on an ROI and an abnormal behavior pattern of an object when processing image information of a monitoring node equipped with an image sensor; A situation propagation information propagation step (S4) of propagating information on the status of the monitoring node to the neighboring monitoring node; Characterized in that it comprises the information transmission step (S5) for transmitting the situation awareness information according to the wireless sensor network routing path to the monitoring server.

The present invention can improve the simple fusion object context information of heterogeneous sensor information into multi-sensor and sensor-based object context information centered on sensor nodes, and based on multi-sensors to minimize context awareness information of unmanned sensor nodes. In addition to tuning the self-diagnostic and adaptive sensor parameters to determine the type and situation of the object by the step-by-step sensor data collection, the sensor-centered unmanned unmanned surveillance system can be applied to the conventional server-oriented redundant unmanned surveillance system. As a result, it is possible to transmit only the minimum sensor node status information to the server.

Hereinafter, with reference to the accompanying drawings will be described in detail a preferred embodiment of the sensor node situation awareness unmanned monitoring method for multi-sensor and node collaboration.

1 is a flowchart illustrating an unmanned monitoring method of a sensor node situation for the multi-sensor and node collaboration according to the present invention, FIG. 2 is a flowchart illustrating the execution of a primary sensor data collection step according to the present invention, and FIG. 3 is a secondary sensor according to the present invention. 4 is a flow chart of the data acquisition step, Figure 4 is a flow chart of the situation information forming step according to the present invention, Figure 5 is a flow chart of the automatic focusing operation of the region of interest of the image sensor detection monitoring node, Figure 6 is a multi-sensor and As a schematic diagram of a situation or unmanned surveillance system for collaboration, the present invention provides a sensor node situation awareness unattended monitoring method for multi-sensor and node collaboration. Determination step S1; A situation information forming step (S2) of forming situation or information for continuous tracking of a moving object by cooperation with a neighbor monitoring node; A pattern extraction step (S3) of extracting an intensive action on an ROI and an abnormal behavior pattern of an object when processing image information of a monitoring node equipped with an image sensor; A situation propagation information propagation step (S4) of propagating information on the status of the monitoring node to the neighboring monitoring node; An information transmission step (S5) is performed to transmit the situation awareness information according to the wireless sensor network routing path to the monitoring server.

The object type and situation determination step (S1) is to collect sensor data in the sensor data collection step of the first and second stages.

The primary sensor data collection step includes a step of collecting data by a sensor (steps S11 and S11 '); A frequency conversion step (step S12); Determining whether or not the standard acoustic frequency pattern is matched (step S13); Giving a specific value (steps S14, S14 '); The monitoring level judgment logic consists of a situation determination step (S15) of determining the type and situation of the object.

In addition, the secondary sensor data collection step includes the step of collecting data by the sensor (steps S21-1 to S21-7); A specific value assigning step (S22-1 to S22-7) for giving a specific value to the collected data; A monitoring level decision logic hiring step (S23 step) of applying the monitoring level decision logic; A monitoring level determining step of determining a monitoring level (step S24); Data transmission steps (S25, S25 ', S25 "steps).

The situation-aware information forming step (S2) is a monitoring level determination step of comparing the monitoring level collected by the operation of the acoustic sensor, vibration sensor, image sensor, temperature / humidity / illumination sensor node with the monitoring level collected from the neighboring monitoring node. (31); A specific value assigning step (S32, S32 ') for assigning a specific value according to the determination result of the monitoring level determining step (S31 step); A transmission step (step S33) of transmitting the monitoring level to a neighboring monitoring node when the monitoring level X = 4 is obtained in the monitoring level determining step (S31); A camera operating step of operating the camera (step S34); Consists of temperature, humidity, roughness measuring step (S35 step) for measuring the temperature / humidity / roughness.

And the pattern extraction step (S3) comprises a monitoring node operation step (step S41); Coordinate information receiving step (step S42); Image sensor operation step (step S43); Abnormal behavior pattern step (step S44); Monitoring server, centralized monitoring node position coordinate receiving step (step S45); Image sensor focusing step (step S46).

Next, the operation of the sensor node context awareness unattended monitoring method for multi-sensor and node collaboration made as described above will be described in detail.

In order to determine the object type and the situation by the sensor data collection step by step of the multi-sensor, the monitoring node collects the sensor data in steps 1 and 2 as shown in Figs. 2 and 3, and according to the monitoring mode setting as shown in Table 1 Configure the sensor byte. Especially in unattended monitoring mode, type and situation of object is determined according to monitoring level as shown in Table 2.

In order to form situational awareness information for continuous tracking of moving objects by collaboration with neighboring monitoring nodes, the monitoring level of the current monitoring node is monitored until the monitoring level of neighboring monitoring node becomes X = 4 as shown in Table 2. Will repeat the operation. The current monitoring node is operated when the monitoring level X = 5 is equipped with an image sensor (camera), and X = 6 is equipped with a temperature / humidity / illuminance sensor to measure temperature, humidity, and illuminance. .

  Monitoring mode  Monitoring level                         Sensor Byte Monitoring contents
                           bit  8  7  6   5   4   3  2  One  0 0 0 0 0  X  X  X Video surveillance Acoustic
Detect vibration
Detect Infrared detection Radar detection Magnetic field sensing Temperature measurement Humidity measurement Roughness measurement Unmanned Surveillance 0 0 0 One  X  X  X  - Acoustic
Information Acoustic
Information Acoustic
Information Acoustic
Information Acoustic
Information Acoustic
Information Sound information Sound information Acoustic 0 0 One 0  X  X  X  - vibration
Information vibration
Information vibration
Information vibration
Information vibration
Information vibration
Information Vibration Information Vibration Information vibration 0 0 One One  X  X  X  - Infrared Information Infrared Information Infrared Information Infrared Information Infrared Information Infrared Information Infrared Information Infrared Information infrared ray 0 One 0 0  X  X  X  - Radar Information Radar Information Radar Information Radar Information Radar Information Radar Information Radar Information Radar Information Radar 0 One 0 One  X  X  X  - Magnetic Field Information Magnetic Field Information Magnetic Field Information Magnetic Field Information Magnetic Field Information Magnetic Field Information Magnetic Field Information Magnetic Field Information magnetic field 0 One One 0  X  X  X  - Temperature Information Temperature Information Temperature
Information Temperature
Information Temperature
Information Temperature Information Temperature Information Temperature Information Temperature 0 One One One  X  X  X  - Humidity Information Humidity Information Humidity
Information Humidity
Information Humidity
Information Humidity Information Humidity Information Humidity Information Humidity One 0 0 0  X  X  X  - Illuminance Information Illuminance Information Roughness
Information Roughness
Information Roughness
Information Illuminance Information Illuminance Information Illuminance Information Roughness X X X X  X  X  X  -  -  -  -  -  -  -  -  -  -

As a method for extracting an intensive action on a region of interest and an abnormal behavior pattern of an object when processing image information of a monitoring node equipped with an image sensor, as shown in FIG. 4, the image information is collected and the standard object behavior pattern DB When the abnormal behavior pattern is generated, the focusing operation is performed on the location coordinate of the intensive monitoring node for the image coordinate, and the object is continuously tracked by extracting the abnormal behavior pattern for the object.

 Monitoring mode  Monitoring level      Sensor Byte watch
level Situation judgment
measure
C A S P R M T H L 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0    no problem   - 0 0 0 0 0 0 One 0 One 0 0 0 0 0 0 0 One Surveillance Object P.R.M sensor operation 0 0 0 0 0 0 One 0 0 One 0 0 0 0 0 0 One Surveillance Object P.R.M sensor operation 0 0 0 0 0 One 0 0 One One 0 0 0 0 0 0 2 Surveillance Object P.R.M sensor operation 0 0 0 0 0 One One 0 One One One 0 0 0 0 0 3  Intruder Detection Intruder Identification 0 0 0 0 0 One One 0 One One 0 One 0 0 0 0 3  Moving object detection Mobile body check 0 0 0 0 0 One One 0 One One 0 0 One 0 0 0 3  Metal body detection Metal body check 0 0 0 0 One 0 0 0 One One One One 0 0 0 0 4 Intruder Movement Occurs Propagation to neighbor watchdog 0 0 0 0 One 0 0 0 0 0 0 One One 0 0 0 4 Metal body movement Propagation to neighbor watchdog 0 0 0 0 One 0 0 0 0 0 One 0 One 0 0 0 3 Metal holder surveillance Check for metal holders 0 0 0 0 One 0 0 0 0 0 One One One 0 0 0 4 Metallic Intruder Movement Propagation to neighbor nodes 0 0 0 0 One 0 One One 0 0 0 0 0 0 0 0 5 Video surveillance with video sensor Camera operation 0 0 0 0 One One 0 0 0 0 0 0 0 One One One 6 Measured when equipped with temperature / humidity / illuminance sensor Temperature / humidity / illumination measurement

The method for propagating the situation detection information of the monitoring node to the neighbor monitoring node includes the case in which the situation detection information of the current monitoring node 100 is propagated to the neighbor sensing node in step 2 of the monitoring node of FIG. The monitoring node 110 and the temperature / humidity / illumination sensor probe 120 may propagate to neighboring monitoring nodes depending on the situation after the sensor operation.

FIG. 6 is a configuration diagram of an unmanned surveillance system for multi-sensor and collaboration. As shown in FIG. 6, only the minimum information on status is transmitted to the monitoring server 200 according to a wireless sensor network routing path. As shown in FIG. 6, the monitoring node forms situation information according to the characteristics of the detected sensor, and gradually propagates through neighboring neighboring monitoring nodes to establish a one-way monitoring system as the final monitoring server 200.

Therefore, in the conventional server-based unmanned surveillance network, it is possible to configure a node-centric wireless network that can improve the problem that the server instability can cause a huge effect on the entire network, it is easy to implement the application service.

While the present invention has been described as a preferred embodiment, the present invention is not limited thereto, and various modifications can be made without departing from the gist of the invention.

1 is an execution flowchart of a sensor node situation awareness unattended monitoring method for multi-sensor and node cooperation according to the present invention;

2 is an execution flowchart of the primary sensor data collection step according to the present invention;

3 is a flow chart of execution of the secondary sensor data collection step according to the present invention,

4 is an execution flowchart of the situation information forming step according to the present invention;

5 is a flowchart of a region of interest automatic focusing operation of an image sensor detection monitoring node;

6 is a block diagram of a situation-aware unattended monitoring system for multi-sensor and collaboration according to the present invention.

Description of the Related Art

100: monitoring node 110: camera-mounted monitoring node

120: monitoring node with temperature / humidity / light sensor 200: monitoring server

300: transmitter

Claims (6)

An object type and situation determination step (S1) of determining an object type and a situation by sensor data collection by stages of multiple sensors; A situation information forming step (S2) of forming situation or information for continuous tracking of a moving object by cooperation with a neighbor monitoring node; A pattern extraction step (S3) of extracting an intensive action on an ROI and an abnormal behavior pattern of an object when processing image information of a monitoring node equipped with an image sensor; A situation propagation information propagation step (S4) of propagating information on the status of the monitoring node to the neighboring monitoring node; Sensor node situation awareness unattended monitoring method for multi-sensor and node collaboration, characterized in that the information transmission step (S5) for transmitting information whether the situation according to the wireless sensor network routing path to the monitoring server. The method according to claim 1, wherein the object type and situation determination step (S1) is a sensor node situation for multi-sensor and node collaboration, wherein sensor data is collected by a sensor data collection step of first and second steps. Unmanned Surveillance Method. The method of claim 2, wherein the primary sensor data collecting step comprises: collecting data by the sensor (S11, S11 '); A frequency conversion step (step S12); Determining whether or not the standard acoustic frequency pattern is matched (step S13); Giving a specific value (steps S14, S14 '); Sensor level status unattended monitoring method for the multi-sensor and node collaboration, characterized in that it comprises a situation determination step (S15) of judging the type and situation of the object by the monitoring level decision logic. The method of claim 2, wherein the second sensor data collecting step comprises: collecting data by the sensor (steps S21-1 to S21-7); A specific value assigning step (S22-1 to S22-7) for giving a specific value to the collected data; A monitoring level decision logic hiring step (S23 step) of applying the monitoring level decision logic; A monitoring level determining step of determining a monitoring level (step S24); Sensor node status unattended monitoring method for multi-sensor and node collaboration, characterized in that the data transmission step (S25, S25 ', S25 "step). The method of claim 1, wherein the contextual information forming step (S2) includes a monitoring level collected by a neighboring monitoring node and a monitoring level collected by an operation of an acoustic sensor, a vibration sensor, an image sensor, and a temperature / humidity / illumination sensor node. Monitoring level determining step 31 for comparing; A specific value assigning step (S32, S32 ') for assigning a specific value according to the determination result of the monitoring level determining step (S31 step); A transmission step (step S33) of transmitting the monitoring level to a neighboring monitoring node when the monitoring level X = 4 is obtained in the monitoring level determining step (S31); A camera operating step of operating the camera (step S34); Unattended monitoring method of sensor node status for multi-sensor and node collaboration, characterized in that the temperature, humidity, roughness measuring step, temperature, humidity, roughness measuring step (step S35). The method of claim 1, wherein the pattern extraction step (S3) comprises: a monitoring node operation step (step S41); Coordinate information receiving step (step S42); Image sensor operation step (step S43); Abnormal behavior pattern step (step S44); Monitoring server, centralized monitoring node position coordinate receiving step (step S45); Sensor sensor situation recognition unmanned monitoring method for multi-sensor and node collaboration, characterized in that the image sensor focusing step (step S46).

Images