JP6857007B2 - Fire monitoring system - Google Patents

Description

The present invention relates to a fire monitoring system that determines a fire by a neural network from an image of a monitoring area captured by a sensor such as a fire detector or a surveillance camera and gives an alarm.

Conventionally, a system for detecting a fire using a sensor that monitors a specific physical quantity such as a smoke detector or a heat detector has been put into practical use.

On the other hand, conventionally, various devices and systems have been proposed in which a fire is detected by performing image processing on an image of a surveillance area captured by a surveillance camera.

In such a fire monitoring system, early detection of a fire is important from the viewpoint of initial fire extinguishing and evacuation guidance.

For this reason, in the conventional device (Patent Document 1), as phenomena caused by smoke accompanying a fire from an image, a decrease in transmittance or contrast, a convergence of a brightness value to a specific value, and a narrowing of the brightness distribution range to disperse the brightness. The decrease in brightness, the change in the average brightness due to smoke, the decrease in the total amount of edges, and the increase in intensity in the low frequency band are derived, and these are comprehensively judged to enable smoke detection.

Japanese Unexamined Patent Publication No. 2008-046916 Japanese Unexamined Patent Publication No. 7-245757 JP-A-2010-238028 Japanese Unexamined Patent Publication No. 6-325270

However, a fire detection system using a sensor that monitors a specific physical quantity has a problem that even if a monitoring standard is satisfied due to a non-fire event, it is regarded as a fire and the fire cannot be correctly discriminated.

Further, in a conventional fire monitoring system that detects a fire from an image of smoke associated with a fire, the characteristics of smoke such as transmission rate, contrast, and edge in the smoke image are determined in advance, and the image captured by the surveillance camera is used. It is necessary to generate smoke characteristics by processing, and there are various situations of smoke generation due to fire, and it is extremely difficult to find out what characteristics smoke has in it, which is a decisive factor. Since it is difficult to find the characteristic, a fire monitoring system that accurately judges smoke caused by a fire from a monitoring image and outputs a fire alarm is in the process of being put into practical use.

On the other hand, in recent years, for example, a large number of images of cats and dogs are labeled, and they are trained by a multi-layer neural network equipped with a convolutional neural network, so-called deep learning is performed, and new images are learned. A technique for presenting to a pre-existing multi-layer neural network and determining whether it is a cat or a dog is disclosed.

Further, deep learning is being studied to be used not only for image analysis but also for natural language processing and behavior analysis.

Such a multi-layer neural network is provided in a judgment device that judges a fire from the input information by using the physical quantity obtained from a sensor represented by a fire detector or the image of the monitoring area captured by the monitoring camera as input information and learning. At times, a large number of input information during fire and non-fire is prepared and trained by a multi-layer neural network, and at the time of monitoring, if the input information is input to a trained multi-layer neural network, the output is used. It is possible to construct a fire monitoring system that estimates whether or not it is a fire with high accuracy and outputs an alarm.

In this case, the multi-layer neural network is trained by using a large number of input information at the time of fire and non-fire prepared in advance at the manufacturing stage of the fire monitoring system as learning information with a teacher, and the trained multi-layer neural network is used. A judgment device equipped with a network is installed in the monitoring facility, and the sensor data of the fire detector or the like installed in the monitoring area or the image captured by the camera is input to the judgment device to monitor the fire.

However, the learning of the multi-layer neural network performed at the manufacturing stage is not the data acquired in the actual monitoring area, but the learning using the input information prepared as standard, and is input by the local sensor or surveillance camera. It remains possible that the fire cannot be estimated with sufficiently high accuracy when the actual physical quantity or surveillance image is input.

In the present invention, in cooperation with a fire alarm system equipped with a fire detector, a multi-layer neural network can be efficiently learned by input information at the time of fire and / or non-fire suitable for a monitoring area to improve the accuracy of fire determination. The purpose is to provide a fire monitoring system that can be improved.

(Fire detection by deep learning)
The present invention is a fire monitoring system that detects a fire based on input information by using a fire detector composed of a multi-layer neural network.
It is characterized by having a learning control unit that learns a fire detector from deep learning.

(Input information)
Physical quantity detected by the sensor and or the input information image of the monitored area captured by the imaging unit.

(Cooperation with receiver)
Learning control unit causes the learning input information the installed fire detector to the monitoring area corresponding to the fire monitoring result by a receiver connected as learning information. Further, when the image of the monitoring area is used as the input information, the moving image of the monitoring area captured by the imaging unit may be extracted at predetermined intervals to create the image of the monitoring area.

(Fire learning by input information corresponding to the detector that reported the fire)
When the signal from the fire alarm of the fire detector is input, the learning control unit inputs the input information from the predetermined time before the signal input by the fire alarm to the input information corresponding to the fire detector that issued the fire alarm. It makes learning the information as fire learning information.
Alternatively, the fire detector detects the temperature or smoke concentration and sends the detected analog value to the receiver to judge the fire.
When the detected analog value reaches a predetermined fire judgment level and the receiver determines that there is a fire, the learning control unit fires from the receiver when the detected analog value exceeds a predetermined fire sign level lower than the fire judgment level. Of the input information up to the time when it is determined that the fire is reported, the input information corresponding to the fire detector that reported the fire is learned as fire learning information.

(Fire learning at the time of fire determination by input information from the specified time before the fire alarm)
When the fire signal is input based on the fire alarm of the fire detector by the receiver and then the fire signal is input based on the fire assertion operation, the learning control unit starts the fire transfer signal from a predetermined time before. of the input information to the time of input of the causes learning input information corresponding to the fire detector which is fire alert as fire learning information.

(Fire learning at the time of fire determination by input information from exceeding the fire sign level)
The fire detector detects the temperature or smoke concentration and sends the detected analog value to the receiver to judge the fire .
In the learning control unit, after the detection analog value reaches a predetermined fire judgment level and a fire transfer signal based on the fire alarm of the fire detector is input from the receiver, the fire determination transfer signal based on the fire determination operation is input. If it is, among the input information to the time of input of a fire Utsuriho signals from when the detection analog value exceeds a predetermined fire indication level lower than the fire determination level, the input information corresponding to the fire detector which is fire alert make learning as fire learning information.

(Non-fire learning at the time of restoration by input information from the specified time before the fire alarm)
When the recovery transfer signal based on the recovery operation is input after the fire transfer signal based on the fire alarm of the fire detector is input by the receiver, the learning control unit inputs the fire transfer signal from a predetermined time before. of the input information until causes learn the input information corresponding to the fire detector which is fire alert as a non-fire learning information.

(Non-fire learning at the time of recovery by input information from exceeding the fire sign level)
The fire detector detects the temperature or smoke concentration and sends the detected analog value to the receiver to judge the fire .
In the learning control unit, the detection analog value reached a predetermined fire judgment level , a fire transfer signal based on the fire alarm of the fire detector was input from the receiver, and then a recovery transfer signal based on the restoration operation was input. In the case, among the input information from the time when the detected analog value exceeds the predetermined fire sign level lower than the fire judgment level to the time when the fire transfer signal is input, the input information corresponding to the fire detector that issued the fire is non-fire. make learning as learning information.

(Functional configuration of multi-layer neural network)
Multi-layer neural networks
It consists of a feature extraction unit and a recognition unit.
Feature extraction unit, a neural network narrowing seen tatami with a plurality of convolution layer for generating feature information input information is entered features of the input information is extracted,
Recognition unit is supplied with characteristic information outputted from the neural network narrowing seen tatami, a neural network having a plurality of total binding layer for outputting a fire characteristic value.

(Learning by backpropagation)
The learning control unit learns by backpropagation (error backpropagation) based on the error between the value output when the learning information is input to the multi-layer neural network of the fire detector and the predetermined expected value.

(Initialization learning using normal monitoring input information in the monitoring area)
Learning control unit causes the initial learning of the input information of the normal monitoring state of the receiver as a learning information of a non-fire.

(Timing of initialization learning)
Initialization timing learning, when starting up the apparatus, when a predetermined operation is performed, when there is no substantially change in input information, see containing a predetermined time interval, one or more either one of,
In the case of each predetermined time interval, initialization learning may be performed in a plurality of patterns for the initial time and the time interval .

(Clarification of fire judgment grounds)
Fire detector, Ru is displayed about the factors determining the fire addition to the detection of a fire.

(Basic effect)
The present invention has a learning control unit that learns a fire detector from deep learning in a fire monitoring system that detects a fire based on input information using a fire detector composed of a multi-layered neural network. Therefore, it is possible to estimate a fire with high accuracy from the output of the sensor and the image of the monitoring area, which cannot be determined by artificial analysis as to whether it is a fire or a non-fire.

(Effect of input information)
The physical quantity detected by the sensor and or, since the input information image of the monitored area captured by the imaging unit, a number from the input information such as an image of the imaged surveillance area by the physical quantity or the imaging unit, which is detected by the sensor It is possible to efficiently learn the multi-layered neural network of the fire detector by acquiring the learning information of the fire detector, and then by inputting the input information to the trained fire detector, the fire is estimated with high accuracy. Allows alarm.

(Effect of cooperation with receiver)
Further, the learning control unit, due to so as to learn the input information the installed fire detector to the monitoring area corresponding to the fire monitoring result by a receiver connected as learning information, standard fire and at the manufacturing stage multilayered neural network of the fire detector learned by non-fire images, based on the fire monitoring result of the receiver, if the fire alarm is output for example, the multilayered the time image as a fire image By learning the neural network and learning the multi-layered neural network with the image as a non-fire image at that time when a fire alarm is output but it is non-fire, it can actually be used as an imaging unit. Learning from images of fire and non-fire corresponding to the monitoring area monitored by the surveillance camera is efficiently performed, and it is possible to estimate a fire with high accuracy from the surveillance image captured by the surveillance camera and make an alarm.

This point also applies to physical quantities such as temperature and smoke concentration detected by the sensor, and it is possible to estimate a fire with high accuracy from the detection signal of the sensor and make an alarm.

(Effect of fire learning by input information corresponding to the detector that reported the fire)
In addition, when the signal from the fire alarm of the fire detector is input, the learning control unit corresponds to the fire detector that issued the fire among the input information from the predetermined time before the signal input by the fire alarm. input information makes learning as fire learning information, or fire detector may send to the receiver detected analog value by detecting the temperature or smoke density is determined fire, learning control unit detects analog values predetermined When the fire judgment level is reached and the receiver determines that the fire is a fire, the input information from the time when the detected analog value exceeds the predetermined fire sign level lower than the fire judgment level to the time when the receiver determines that the fire is a fire. Of these, since the input information corresponding to the fire detector that reported the fire is learned as fire learning information, the input information at the location where the fire occurred can be obtained as learning data, and it is possible to automatically obtain the learning data. Become.

Sensor case, for example, when the physical quantity from 5 minutes before the fire alert is learned as a physical quantity of a fire, every predetermined time intervals, for example every five seconds to input information of physical quantity detected by the sensor of the fire detector or the like Assuming that the data has been detected, 60 sets of sensor data will be obtained. Further, when the image of the surveillance camera is used as the input information, for example, when the image from 5 minutes before the fire alarm is learned as the image of the fire , if the recorded image is recorded at 30 frames / second, it takes 5 minutes. 9000 images can be obtained from the recorded images of. As described above, by issuing a fire detector once, learning with multiple input information corresponding to the fire detector can be easily realized, so that a fire can be estimated with higher accuracy from the input information and an alarm can be given. And.

(Effect of fire learning at the time of fire determination by input information from a predetermined time before the fire alarm)
In addition, when the fire signal is input based on the fire alarm of the fire detector by the receiver and then the fire signal is input based on the fire determination operation, the learning control unit starts the fire transfer from a predetermined time before. of the input information to the time of input of the broadcast signal, since the input information corresponding to the fire detector which is fire alert was set to be learned as a fire learning information, fire information input in a state where the fire is reliably generated Since it can be learned as input information of time, it becomes possible to learn with error-free training data without the trouble of labeling the training data as fire or non-fire.

(Effect of fire learning by input information from exceeding the fire sign level)
Further, the fire detector sends a detection analog value by detecting the temperature or smoke density is determined fire, learning control unit detects an analog value reaches a predetermined fire determination level, the fire detector from the receiver If a fire definitively Utsuriho signal based on fire assertive operation after the fire Utsuriho signal based on fire alert has been input is entered, the fire from the time when the detected analog value exceeds a predetermined fire indication level lower than the fire determination level of the input information to the time of input of the Utsuriho signal, since the input information corresponding to the fire detector which is fire alert was set to be learned as a fire learning information, a physical quantity detected by the analog type fire detector monitoring temperature and smoke density regions, since it learns the input information from the time it reaches a predetermined fire sign level as a sign of lower fire level fire until the input of the fire Utsuriho signals, from the initial stages of a fire It is possible to read and learn a large amount of input information until it is determined to be a fire as the input information of the fire, and it is possible to detect a sign of a fire.

(Effect of non-fire learning at the time of recovery by input information from a predetermined time before the fire alarm)
When the recovery transfer signal based on the recovery operation is input after the fire transfer signal based on the fire alarm of the fire detector is input by the receiver, the learning control unit inputs the fire transfer signal from a predetermined time before. of the input information until, due to so as to learn the input information corresponding to the fire detector which is fire alert as a non-fire learning information, for example, inputs the input information from five minutes before the fire alert non-fire When learning as information, the input information in the state where a fire has not occurred can be learned as the input information at the time of non-fire, so there is no need to label the training data as fire or non-fire, and there is no error in learning. You will be able to learn from the data.

(Effect of non-fire learning at the time of recovery by input information from exceeding the fire sign level)
Further, the fire detector may send to the receiver detected analog value by detecting the temperature or smoke density is determined fire, learning control unit detects an analog value reaches a predetermined fire determination level, the fire receiver When the recovery transfer signal based on the recovery operation is input after the fire transfer signal based on the fire alarm of the fire detector is input, when the detected analog value exceeds the predetermined fire sign level lower than the fire judgment level. of the input information to the input mode of the fire Utsuriho signals from, because it so as to learn the input information corresponding to the fire detector which is fire alert as learning information of the non-fire has been detected by the analog type fire detector When the temperature or smoke concentration in the monitoring area, which is a physical quantity, reaches a predetermined fire sign level that indicates a fire sign lower than the fire judgment level, but it is a non-fire due to an increase in temperature or smoke concentration due to a cause other than fire. In addition, since learning the non-fire input information from the time when the fire sign level is reached as the non-fire input information, a large amount of input information from the initial stage to the non-fire input information is used as the non-fire input information. You can learn, with respect to non-fire state of which could be regarded as a sign of fire reliably prevented possible to false alarms by estimating the non-fire with greater accuracy.

(Effect of functional configuration of multi-layer neural network)
Further, the multi-layer neural network is composed of a feature extraction unit and a recognition unit, and the feature extraction unit is a plurality of convolutional layers that generate feature information from which input information of a monitoring area is input and features of the input information are extracted. a neural network convolution with a recognition unit characteristic information outputted from the convolutional neural network is input, since as a neural network having a plurality of total binding layer for outputting a fire characteristic value, a convolution neural network By automatically extracting the features by, the features of the fire input information by preprocessing from the input information of the monitoring area, for example, in the image, input without the need for preprocessing such as extracting the contour etc. The characteristics of the information are extracted, and the subsequent recognition unit makes it possible to estimate the fire with high accuracy.

(Effect of learning by backpropagation)
Learning control unit because it has so as to learn by back propagation based on the error between the value and the predetermined expectation value is output when you enter learning information multilayered neural network of the fire detector (Back Propagation) , Input a large amount of input information as fire learning information in response to the fire alarm, give the estimated value of the fire as the expected value of the output, and minimize the error between the output value and the expected value by back propagation processing. Weights and biases in multi-layer neural networks are learned, and fires can be estimated with higher accuracy from input information and alerted.

Similarly, when a fire alarm is non-fire, a large amount of input information is input as non-fire learning information, an estimated non-fire value is given as the expected output value, and the output value and expectation are obtained by back propagation processing. Weights and biases in multi-layer neural networks are learned to minimize value errors, and non-fire can be estimated with higher accuracy from input information to ensure prevention of false alarms.

(Effect of initialization learning by normal monitoring input information in the monitoring area)
In addition, since the learning control unit initializes and learns the input information of the receiver 's normal monitoring state as non-fire learning information , the non-fire estimation accuracy for the input information of the monitoring area in the normal monitoring state is improved. , Learning is performed by the input information of the fire linked to the fire monitoring of the receiver or the input information of the non-fire, and the accuracy of the estimation for the fire and the non-fire of the multi-layered neural network is further improved.

(Effect of initialization learning timing)
The timing of the initialization learning, when starting up the apparatus, when a predetermined operation is performed, when there is no substantially change in input information, see containing a predetermined time interval, any one or more one of, In the case of each predetermined time interval, the initial time and time interval may be initialized in multiple patterns, so the non-fire condition in the installation environment is educated by performing initialization learning when the device is started up. By performing initialization learning when a predetermined operation is performed, it is possible to educate the non-fire state at any time. For example, when the interior is changed, the non-fire state can be immediately learned. It becomes possible to educate. In addition, by performing initialization learning when there is almost no change in the sensor output or the image captured by the camera, it is possible to automatically educate the non-fire state in a stable monitoring area, and further, at predetermined time intervals. Initialization learning is performed for each initial time and time interval in a plurality of patterns . For example, by shifting the timing of initialization learning by predetermined times, non-fire learning information can be obtained at different times. You can get and learn non-fire.

(Effect of clarifying the grounds for fire judgment)
Further, the fire detector, because you so that to display the factors that determines that fire addition to the detection of a fire, for example, in the surveillance camera images, and displays the image determined as a fire, even higher contribution rate determination fire By highlighting the area where the fire has occurred, it becomes easier to visually check the area determined by the fire detector as a fire, and it becomes possible to easily determine whether or not a fire has really occurred. Can help.

Explanatory drawing showing the outline of a fire monitoring system that monitors a fire with a surveillance camera and a fire detector. Explanatory drawing showing the functional configuration of the judgment device using the multi-layer neural network that estimates the fire from the image captured by the surveillance camera. Explanatory diagram showing the functional configuration of the multi-layer neural network shown in FIG. A flowchart showing learning control of a multi-layer neural network linked to fire monitoring of a fire receiver by the learning control unit of FIG. Explanatory drawing showing the time change of the detected analog value detected by the analog fire detector Explanatory drawing showing the outline of a fire monitoring system that monitors a fire with an analog fire detector that functions as a sensor. Explanatory diagram showing the functional configuration of a judgment device using a multi-layer neural network that estimates a fire from a detection signal from an analog fire detector. A time chart showing the time change of smoke concentration detected by the analog fire detector stored in the time series data generator of FIG.

[Overview of fire monitoring system]
FIG. 1 is an explanatory diagram showing an outline of a fire monitoring system that monitors a fire with a surveillance camera and a fire detector.

As shown in FIG. 1, surveillance cameras 16-1 and 16-2 that function as imaging means are installed in the monitoring areas 14-1 and 14-2 of facilities such as buildings, respectively, and the monitoring areas 14-1 are monitored cameras. According to 16-1, the surveillance area 14-2 is imaged as a moving image by the surveillance camera 16-2.

The surveillance areas 14-1 and 14-2 are described as the surveillance area 14 when there is no particular need to distinguish them, and the surveillance cameras 16-1 and 16-2 are described as the surveillance cameras 16 when there is no particular need to distinguish them.

The surveillance camera 16 captures an RGB color image at, for example, 30 frames / second and outputs it as a moving image. Further, one frame has, for example, a pixel arrangement of 4056 × 4056 pixels in length and width.

On-off type fire detectors 18-1 and 18-2 are installed in the monitoring areas 14-1 and 14-2, respectively, to detect the temperature or smoke concentration due to the fire and exceed a predetermined threshold level. The case is issued and a fire alarm signal is output.

The fire detectors 18-1 and 18-2 are described as the fire detector 18 when it is not necessary to distinguish them.

A determination device 10 and a receiver 12 of a fire alarm system are installed in the disaster prevention monitoring center and the manager's room of the facility with respect to the monitoring area 14. The determination device 10 and the receiver 12 may be integrated. A surveillance camera 16 installed in the surveillance area 14 is connected to the determination device 10 by a signal cable 20, and a moving image image captured by the surveillance camera 16 is input.

A sensor line 22 is pulled out from the receiver 12 to the monitoring area 14, and a fire detector 18 is connected to each of the sensor lines 22.

The determination device 10 includes a multi-layer neural network, inputs a moving image image sent from the surveillance camera 16 in frame units, and when a fire image is input, outputs a fire determination signal to the receiver 12, for example, a fire. Output a fire sign alarm, etc. that indicates a sign. Further, the moving image from the surveillance camera 16 is continuously recorded by the recording device provided in the determination device 10.

When the receiver 12 receives the fire alarm signal from the fire detector 18, the receiver 12 outputs a fire alarm, and also outputs a fire transfer signal including a signal for identifying which fire detector has been issued to the determination device 10. Output. When a fire alarm is output from the receiver 12, the person in charge of management or the person in charge of disaster prevention goes to the installation site of the fire detector 18 that has issued a fire alarm to check for the presence of a fire, and if a fire is confirmed, receives the fire alarm. A fire determination operation is performed on the machine 12. When the fire determination operation is performed by the receiver 12, the temporarily suspended district acoustic alarm is canceled, and the fire determination transfer signal is output to the determination device 10.

If there is no fire in the on-site confirmation of the fire alarm, after removing the cause of the non-fire, the receiver 12 performs a recovery operation to cancel the fire alarm state and return to the normal monitoring state. If a fire recovery operation is performed without performing the fire determination operation after the fire alarm is output by the receiver 12 in this way, the recovery transfer signal is output from the receiver 12 to the determination device 10.

The determination device 10 is recorded in the recording device corresponding to the alarm location of the fire detector based on the fire monitoring result by the fire transfer signal, the fire determination signal, and the recovery signal output from the receiver 12. A moving image captured by the monitoring camera 16 in the monitoring area 14 up to the output of the fire alarm is read from the recording device, and this is used as a fire image or a non-fire image to control the learning of the multi-layer neural network of the determination device 10. For example, when the fire detector 18-1 issues a report, the moving image captured by the surveillance camera 16-1 is read from the recording device.

[Judgment device]
(Functional configuration of judgment device)
FIG. 2 is an explanatory diagram showing a functional configuration of a determination device using a multi-layer neural network that estimates a fire from an image captured by a surveillance camera.

As shown in FIG. 2, the determination device 10 includes a fire detector 24, a recording device 26 functioning as a storage unit, a learning image holding unit 28, and a learning control unit 30, and the fire detector 24 includes an image input unit 32. , A multi-layer neural network 34 and a determination unit 36. Here, the functions of the fire detector 24, the learning image holding unit 28, and the learning control unit 30 are realized by executing a program by the CPU of the computer circuit corresponding to the processing of the neural network.

An image of the monitored area fire detector 24 is captured by the surveillance camera 16, via the image input unit 32 inputs to the multi-layered neural network 34, the determination unit 36 determines whether a fire or non-fire from the output value , When a fire is determined, a fire determination signal is output to the receiver 12.

The recording device 26 records a moving image in the monitoring area captured by the monitoring camera 16, and can partially read the recorded moving image according to a playback instruction from the outside.

Learning control unit 30, the fire Utsuriho signal E1 from the receiver 12, based on the fire assertion Utsuriho signal E2 and recovery Utsuriho signal E3, when the fire alarm is output by the receiver 12, needed from the recording device 26 The moving image of the portion to be used is read out and temporarily stored and held in the learning image holding unit 28, the images are sequentially read out frame by frame from the moving image held in the learning image holding unit 28, and the multi-layer neural network is passed through the image input unit 32. It is input to the network 34 as a supervised image, and the weight and bias of the multi-layer neural network 34 are learned by a learning method such as a back propagation method (error back propagation method).

When an image of the surveillance area captured by the surveillance camera 16 is input to the multi-layer neural network 34 that has been trained using the supervised image, an estimated value corresponding to the fire image is output. This estimated value has an expected value of 1 in the case of a fire image used for learning, an expected value of 0 in the case of a non-fire image, and a value in the range of 0 to 1 when an actual image is input. , Is input to the determination unit 36 and compared with a predetermined threshold value, for example, 0.5, and when the threshold value or more is exceeded, a fire determination signal is output to the receiver 12, and the receiver 12 outputs, for example, a fire sign alarm.

In addition, a monitor device was provided in the determination device 10, and when a fire was determined, the image of the monitoring area where the fire was determined was displayed on the screen, which was captured by the surveillance camera 16, and the fire sign warning from the receiver 12 was known. The person in charge of management or the person in charge of disaster prevention may be able to confirm the fire. In this case, if a fire determination switch is provided in the operation unit of the determination device 10 and the fire determination switch is operated when a fire is confirmed from the monitor image , a fire transfer signal is output in the same manner as when the receiver 12 is operated. , The fire alarm may be output from the receiver 12.

Further, although the determination device 10 uses the recorded information corresponding to the alarm location of the fire detector as the input information, it is preferable that the fire detector 24 is also independent for each detector. That is, the learning method is the same for all the fire detectors 24, but each fire detector 24 is given different input information, and the fire determination is also performed by different determinations. As a result, learning specialized in the installation environment is performed.

[Multi-layer neural network]
FIG. 3 is an explanatory diagram showing the functional configuration of the multi-layer neural network shown in FIG. 2, the outline is shown in FIG. 3 (A), and the details are schematically shown in FIG. 3 (B).

As shown in FIG. 3A, the multi-layer neural network 34 of the present embodiment is composed of a feature extraction unit 38 and a recognition unit 40. The feature extraction unit 38 is a convolutional neural network, and the recognition unit 40 is a fully connected neural network.

The multi-layer neural network 34 is a neural network that performs deep learning, is a neural network that has a deep layer that connects a plurality of intermediate layers, and performs expression learning that is feature extraction.

A normal neural network requires work by artificial trial and error to extract features for estimating a fire from an image, but in a multi-layer neural network 34, a convolutional neural network is used as a feature extraction unit 38. Then, the pixel value of the image is input, the optimum feature is extracted by learning, and the optimum feature is input to the fully connected neural network of the recognition unit 40 to identify whether it is a fire or a non-fire.

As schematically shown in FIG. 3B, the fully connected neural network of the recognition unit 40 is composed of an input layer 46, a connecting layer 48, a repetition of the intermediate layer 50 and the connecting layer 48, and an output layer 52. ..

(Convolutional neural network)
FIG. 3B schematically shows the structure of the convolutional neural network constituting the feature extraction unit 38.

Convolutional neural networks have slightly different characteristics from ordinary neural networks and incorporate biological structures from the visual cortex. The visual cortex contains a receptive field, which is a collection of small cells that are sensitive to a small area of the visual field, and the behavior of the receptive field can be imitated by learning weighting in the form of a matrix. This matrix is called the weight filter (kernel) and is sensitive to similar subregions of an image, similar to the biological role played by receptive fields.

The convolutional neural network can express the similarity between the weight filter and the subarea by the convolution operation, and through this operation, the appropriate feature of the image can be extracted.

As shown in FIG. 3B, the convolutional neural network first performs convolution processing on the input image 42 by the weight filter 43. For example, the weight filter 43 is a matrix filter having a predetermined weighting of 3 × 3 in the vertical and horizontal directions, and by performing a convolution calculation while aligning the filter center with each pixel of the input image 42, 9 pixels of the input image 42 can be obtained. convolution to one pixel of feature maps 44a to be subregions, a plurality of feature maps 44a is generated.

Subsequently, the pooling calculation is performed on the feature map 44a obtained by the convolution calculation. The pooling operation is a process of removing features unnecessary for identification and extracting features necessary for identification.

Subsequently, the convolution calculation and the pooling calculation using the weight filters 45a and 45b are repeated in multiple stages to obtain the feature maps 44b and 44c, the feature map 44c of the last layer is input to the recognition unit 40, and the normal full coupling is performed. A recognition unit 40 using a neural network estimates whether it is a fire or a non-fire.

Note that the pooling calculation in the convolutional neural network does not perform the pooling calculation because the features unnecessary for distinguishing between fire and non-fire are not always clear and the necessary features may be deleted. You may do so.

[Learning of multi-layer neural network]
(Backpropagation)
A neural network composed of an input layer, a plurality of intermediate layers, and an output layer is provided with a plurality of units in each layer and connected to a plurality of units in other layers, and weights and bias values are set for each unit, and a plurality of units are set. The vector product of the input value and the weight of is obtained, the bias value is added to obtain the sum, and this is passed through a predetermined activation function and output to the unit of the next layer to reach the final layer. Forward propagation is performed in which the value propagates to.

To change the weights and biases of such neural networks, a learning algorithm known as backpropagation is used. In backpropagation, supervised learning when a data set of input value x and expected output value (expected value) y is given to the network, and unsupervised learning when only input value x is given to the network. In this embodiment, supervised learning is performed.

When performing backpropagation in supervised learning, for example, a function of mean squared error is used as an error to compare the estimated value y * and the expected value y, which are the results of forward propagation that has passed through the network. To do.

In backpropagation, the magnitude of the error between the estimated value y * and the expected value y is used to propagate the value from the rear to the front of the network while correcting the weight and bias. The corrected amount for each weight and bias is treated as a contribution to the error, calculated by the steepest descent method, and the value of the error function is minimized by changing the weight and bias values.

The procedure for learning by backpropagation for a neural network is as follows.
(1) The input value x is input to the neural network, forward propagation is performed, and the estimated value y * is obtained.
(2) Calculate the error with an error function based on the estimated value y * and the expected value y.
(3) Backpropagation is performed on the network while updating the weight and bias.

This procedure is repeated with different combinations of input values x and expected values y to minimize the value of the error function until the error in the weight and bias of the neural network is minimized as much as possible.

[Learning control linked to receiver fire monitoring]
(Learning with fire images)
The learning control unit 30 of the determination device 10 shown in FIG. 2 receives a fire transfer signal E1 based on the fire alarm of the fire detector 18 by the receiver 12 and outputs a fire alarm, and the site by the manager or the like. When a fire is confirmed by confirmation and the fire determination signal E2 based on the fire determination operation of the receiver 12 is input based on this, for example, from a predetermined time 5 minutes before to the input of the fire transfer signal E1. The image of the monitoring area of the above is read from the recording device 26, input as a fire image to the multi-layer neural network 34 of the fire detector 24 as a fire image via the image input unit 32, the estimated value y * is obtained, and the expected value y = 1 of the fire. Back propagation is repeated while changing the supervised image so as to minimize the value of the error function of and, and learning is performed to change the weight and bias by minimizing the error function.

In this case, assuming that the recorded image is recorded on the recording device 26 as a moving image of 30 frames / second, 9000 fire images can be obtained from the recorded image of 5 minutes, and learning with a large number of fire images such as 9000 is easy. By realizing this, it is possible to estimate a fire with higher accuracy from the surveillance image captured by the camera and make an alarm.

When a fire image is generated from a 5-minute video, the change between each frame image having a 1/30 second cycle is very small. Therefore, for example, a frame image thinned out with a 1-second cycle is a supervised image. May be. In this case, 300 fire images can be obtained from the 5-minute moving image, and a sufficient number of images can be obtained for learning the multi-layer neural network 34 by backpropagation.

(Learning with non-fire images)
Further, if the learning control unit 30 is non-fired by the site confirmation by the manager or the like after the fire transfer signal E1 based on the fire alarm of the fire detector 18 is input by the receiver 12. , The recovery operation is performed by the receiver 12, and the recovery transfer signal E3 based on the recovery operation is input. Therefore, in this case, from a predetermined time before, for example, 5 minutes before, until the fire transfer signal E1 is input. The image of the monitoring area is read from the recording device 26 and input to the multi-layer neural network 34 of the fire detector 24 as a non-fire image via the image input unit 32 to obtain an estimated value y *, and the expected non-fire value y = Back propagation is repeated while changing the supervised image so as to minimize the value of the error function from 0, and learning is performed to change the weight and bias.

In this case as well, assuming that the recorded image is recorded on the recording device 26 as a moving image of 30 frames / second, 9000 non- fire images can be obtained from the 5-minute recorded image, and a large number of non-fire images such as 9000 can be obtained. By easily realizing learning by, it is possible to estimate non-fire with higher accuracy from the surveillance image captured by the camera and prevent false reports.

Further, when a non-fire image is generated from a 5-minute video, the change between each frame image having a 1/30 second cycle is very small. Therefore, for example, a frame image thinned out with a 1-second cycle is used as a non-fire image. May be.

(Initialization learning of judgment device)
In the multi-layer neural network 34 of the determination device 10 shown in FIG. 2, weights and biases are randomly initialized at the manufacturing stage in a factory or the like, and standard fire images and non-fire images prepared in advance are used. It is initialized by learning by backpropagation, and in this state, it is installed in the facility to be monitored as shown in FIG.

In this case, the image of the surveillance area 14 captured by the surveillance camera 16 varies depending on the surveillance area 14, and is different from the standard supervised image used in the initialization learning. At the time of start-up after installation, a moving image in a normal monitoring state, that is, a non-fire moving image, which is captured by a surveillance camera 16 and recorded on a recording device 26, is recorded on the recording device 26, and a predetermined state is determined by a playback operation from the moving image. A recording for an hour, for example, 5 minutes is read and held as a non-fire image in the learning image holding unit 28, and this non-fire image is input to the multilayer neural network 34 to obtain an estimated value y *, and an expected non-fire value is obtained. It is desirable to perform learning to change the weight and bias by repeatedly performing back propagation while changing the teacher image so as to minimize the error with y = 0 and minimizing the error function.

As the image with a teacher used for this learning, it is desirable to read out a video different depending on the time zone such as morning, noon, and night from the video of the day in the monitoring area from the recording device 26 and train it as a non-fire image. ..

The timing of initialization learning may be further performed when the device is started up. This makes it possible to first educate the non-fire condition in the installation environment.

Further, the timing of initialization learning may be performed when a predetermined operation is performed. As a result, it is possible to educate the non-fire state at any timing, and it is possible to immediately educate the non-fire state when, for example, the interior is changed.

Further, the timing of initialization learning may be performed when there is no or almost no image taken by the sensor output / camera. As a result, it is possible to automatically educate the non-fire state in a state where the monitoring area is stable.

Further, the timing of initialization learning may be shifted by a predetermined time. For example, the first time the initialization learning is performed at 6:00, 12:00, 18:00, and 24:00, and the second time the initialization learning is performed at 7:00, 13:00, 19:00, and 1:00. As a result, non-fire learning data can be obtained at different times, and non-fire learning can be performed including special conditions such as cooking, sunrise, and sunset.

By learning that the monitoring image of the monitoring area 14 is a non-fire image, the estimation accuracy of the non-fire with respect to the image of the monitoring area in the normal monitoring state is improved. After that, learning is performed using a fire image or a non-fire image linked to the fire monitoring by the receiver 12 described above, and the accuracy of estimating fire and non-fire of the multi-layer neural network 34 is further improved.

(Control operation by the learning control unit)
FIG. 4 is a flowchart showing learning control of a multi-layer neural network linked to fire monitoring of a fire receiver by the learning control unit of FIG.

As shown in FIG. 4, the learning control unit 30 has the recording device 26 record the moving image from the surveillance camera 16 installed in the surveillance area in step S1, and the fire transfer signal from the receiver 12 is recorded in step S2. When the input is determined, the process proceeds to step S3, and the recorded image from a predetermined time ago is read out from the recording device 26 and held by the learning image holding unit 28.

Subsequently, the process proceeds to step S4, and when the learning control unit 30 determines the input of the fire determination signal from the receiver 12, the process proceeds to step S5. After that, when the input of the recovery signal from the receiver 12 is determined, step S6 Proceed to, the recorded image for a predetermined time held in the learning image holding unit 28 is read out, input to the multi-layer neural network 34 as a fire image, and learning is performed to change the weight and bias by backpropagation.

On the other hand, the learning control unit 30 inputs the fire determination signal while determining whether or not the recovery signal is input in step S7 without determining the input of the fire determination signal in step S4. If the input of the recovery transfer signal is determined in step S7 without discrimination, the process proceeds to step S8, the recorded image for a predetermined time held in the learning image holding unit 28 is read out, and the multi-layer neural network is used as a non-fire image. Learning is performed by inputting to the network 34 and changing the weight and bias by back propagation.

[Fire learning by monitoring images from exceeding the fire sign level]
(Learning with fire images)
As another embodiment of learning control by the learning control unit 30 of the determination device shown in FIG. 2, an analog fire detector is installed in the caution area, and the temperature or smoke concentration is detected by the analog fire detector to detect the detected analog value. When the fire is judged by sending to the receiver 12 , the image from the time when the fire sign is judged to the time when the fire is judged is read from the recording device 26, and the multi-layer neural network 34 is learned by backpropagation. ..

As shown in FIG. 5, the fire sign level is set to the fire sign level TH1 which is lower than the fire judgment level TH2 when the temperature detected by the fire detector rises with the passage of time due to the fire that occurred at time t0. The image recorded during the time T from the time when the fire is reached at t1 to the time when the fire judgment level TH2 is reached at time t2 is back-propagated as a fire image.

Since all the images recorded during this time T are images due to fire and do not include non-fire images, they are input to the multi-layer neural network 34 as fire images and weighted and biased by back propagation. By learning to change, the accuracy of identifying a fire from the input image can be surely improved.

Specifically, the learning control unit 30 of the determination device 10 shown in FIG. 2 predicts that the detection analog value of the temperature or smoke concentration from the receiver 12 to the analog fire detector reaches a predetermined fire sign level TH1. An alarm is output, then the detected analog value reaches the fire judgment level TH2, a fire transfer signal based on the fire alarm is input, a fire alarm is output, and the fire is confirmed by on-site confirmation by the manager etc. Based on this, when the fire determination signal E2 based on the fire determination operation of the receiver 12 is input, the image of the monitoring area from the time when the fire sign is detected to the time when the fire signal is input is recorded by the recording device 26. Is read from, and is input as a fire image to the multi-layer neural network 34 of the fire detector 24 via the image input unit 32, and learning is performed to change the weight and bias by back propagation.

(Learning with non-fire images)
Further, the learning control unit 30 outputs a sign alarm when the detection analog value of the temperature or smoke concentration from the analog fire detector reaches a predetermined fire sign level TH1 from the receiver 12, and then the detected analog value is a fire. When the judgment level TH2 is reached, a fire transfer signal based on a fire alarm is input, a fire alarm is output, and if there is no fire at the site confirmation by the manager, etc., the receiver 12 performs a recovery operation. Since the recovery signal is input based on the restoration operation, the image of the monitoring area from the time when the fire sign is detected to the time when the fire signal is input is read from the recording device 26, and the fire detector 24 is read. The multi-layer neural network 34 is input as a non-fire image via the image input unit 32, and learning is performed to change the weight and bias by back propagation.

[Fire monitoring system that monitors fires with sensors]
(Overview of fire monitoring system)
FIG. 6 is an explanatory diagram showing an outline of a fire monitoring system that monitors a fire with an analog fire detector that functions as a sensor.

As shown in FIG. 6, analog fire detectors 60-1 and 60-2 that function as sensors are installed in the monitoring area 14 of a facility such as a building, and are connected to a transmission line 62 drawn from the receiver 12. It enables serial data transmission. The analog fire detectors 60-1 and 60-2 are described as analog fire detectors 60 when there is no particular need to distinguish them.

The analog fire detector 60 detects the smoke concentration by the smoke detector, outputs a smoke concentration detection signal, periodically performs A / D conversion by transmitting a batch AD conversion command from the receiver 12, and stores the smoke concentration data in the memory. Smoke concentration data is transmitted to the poll from the receiver 12 that specifies the sensor address, and if the smoke concentration exceeds a predetermined threshold level, it is judged as a fire and the fire interruption signal is received by the receiver. It is transmitted to 12 to output a fire alarm. The analog fire detector 60 may detect temperature, CO concentration, etc. in addition to smoke concentration.

The determination device 10 includes a multi-layer neural network, and the smoke concentration data detected by the analog fire detector 60 is input via the receiver 12 and stored in the storage unit as input information from the sensor.

Receiver 12 receives the fire interrupt signal of an analog fire detector 60 outputs a fire alarm, and the decision unit 10 outputs a fire Utsuriho signal. When a fire alarm is output from the receiver 12, the person in charge of management or the person in charge of disaster prevention goes to the installation site of the notified analog fire detector 60 to check for the presence of a fire, and if the fire is confirmed, The receiver 12 performs a fire determination operation. When the fire determination operation is performed by the receiver 12, the temporarily suspended district acoustic alarm is canceled, and the fire determination transfer signal is output to the determination device 10.

If there is no fire in the on-site confirmation of the fire alarm, after removing the cause of the non-fire, the receiver 12 performs a recovery operation to cancel the fire alarm state and return to the normal monitoring state. If a fire recovery operation is performed without performing the fire determination operation after the fire alarm is output by the receiver 12 in this way, the recovery transfer signal is output from the receiver 12 to the determination device 10.

The determination device 10 is stored in a storage unit corresponding to the location of the fire detector that transmitted the fire interruption signal based on the fire monitoring result by the fire transfer signal, the fire determination signal, and the recovery signal output from the receiver 12. For example, time-series data is generated from the smoke concentration data detected by the analog fire detector 60 in the monitoring area 14 up to the output of the stored fire alarm, and this is used as learning information for the multi-layer neural of the determination device 10. Controls input to the network for learning.

For example, when the analog fire detector 60-1 transmits a fire interrupt signal, the time-series data of the analog fire detector 60-1 is read from the storage unit. Further, when a sensor other than the analog fire detector is further provided, the sensor data arranged in the monitoring area is read from the storage unit.

For the input information of fire monitoring that monitors fires by sensors, in addition to time-series data, multi-term data from multiple sensors may be adopted, or time-series data from multiple sensors that are a combination thereof should be adopted. You can do it.

(Judgment device)
FIG. 7 is an explanatory diagram showing a functional configuration of a determination device using a multi-layer neural network that estimates a fire from a detection signal from an analog fire detector.

As shown in FIG. 7, the determination device 10 includes a fire detector 24, a time-series data generation unit 64 including a storage unit, a learning data holding unit 68, and a learning control unit 30, and the fire detector 24 is a time. It is composed of a series data input unit 66, a multi-layer neural network 34, and a determination unit 36.

The multi-layer neural network 34 of the present embodiment is only the fully connected neural network constituting the recognition unit 40 shown in FIG. 3A, and the convolutional neural network serving as the feature extraction unit 38 is excluded.

The time-series data generation unit 64 stores the smoke concentration data detected by the analog fire detector 60 via the receiver 12 in the storage unit. The smoke concentration data stored in the storage unit of the time-series data generation unit 64 is, for example, data showing a change in smoke concentration with the passage of time as shown in FIG.

The smoke concentration data of FIG. 8 is an example of a time change of the smoke concentration due to a fire. The smoke concentration starts to increase at time t0, reaches a predetermined fire sign level TH1 at time t1, and then reaches a fire judgment level at time t2. It shows the case where the fire alarm is output when TH2 is reached and the fire assertion operation is performed.

The learning control unit 30 is a time-series data generation unit 64 when a fire alarm is output from the receiver 12 based on the fire transfer signal E1, the fire determination transfer signal E2, and the recovery transfer signal E3 from the receiver 12. Is instructed to generate time-series data based on the smoke concentration sensor data shown in FIG. 8 stored in the storage unit, and the multi-layer neural network 34 is connected to the multi-layer neural network 34 via the time-series data input unit 66 in the time-series of fire. It is input as data, and the weight and bias of the multi-layer neural network 34 are trained by the back propagation method.

The time-series data generation unit 64 generates time-series data, for example, assuming that the smoke concentration data for each predetermined unit time Δt from time t0 to time t1 in FIG. 8 to reach the fire sign level TH1 is S1 to S18. While shifting by unit time Δt, time series data (S1 to S10), (S2 to S11), ... (S9 to S18) for each period T1, T2, ... T9 for a predetermined time are generated. It is stored in the training data holding unit 68.

The learning of the multi-layer neural network using the time series data (S1 to S10), (S2 to S11), ... (S9 to S18) is based on the case of learning from the time series data (S1 to S10). performing value Sl to S 10 in parallel to the input layer of the multi-layered neural network 34. Similarly, the remaining time series data (S2 to S11), ... (S9 to S18) are also learned by sequentially inputting them in parallel to the input layer.

Further, when the learning control unit 30 is non-fired by the site confirmation by the manager or the like after the fire transfer signal E1 based on the fire alarm of the analog fire detector 60 is input by the receiver 12. Is a non-fire time-series data as in the case of the fire time-series data shown in FIG. 8, since the recovery operation is performed by the receiver 12 and the recovery transfer signal E3 based on the recovery operation is input. Is input to the multi-layer neural network 34 of the fire detector 24 as non-fire time-series data via the time-series data input unit 66, and the weight and bias of the multi-layer neural network 34 are learned by the backpropagation method. Let me.

After learning the multi-layer neural network 34 of the fire detector 24, the time-series data generation unit 64 generates time-series data for a predetermined time every predetermined unit time Δt, and the time-series data input unit 66 It will be input to the multi-layer neural network 34 via the above to monitor the fire.

The determination device 10 uses time-series data corresponding to the analog fire detector that transmitted the fire interrupt signal as input information, but it is preferable that the fire detector 24 is also independent for each detector. That is, the learning method is the same for all the fire detectors 24, but each fire detector 24 is given different input information, and the fire determination is also performed by different determinations. As a result, learning specialized in the installation environment is performed.

[Modification of the present invention]
(Clarification of fire judgment grounds)
In the above embodiment, the result of determining the presence or absence of a fire is notified, but in addition to this, the factor determined to be a fire may be displayed. For example, in monitoring a camera image, an image determined to be a fire is displayed, and an area in which the contribution rate of the fire determination is high is highlighted. As a result, it becomes easy to visually confirm the area determined by the fire detector as a fire, it becomes possible to easily determine whether or not a fire has really occurred, and it can be used as an aid in making a response decision according to the situation. ..

(Arson monitoring)
The above embodiment takes fire monitoring in a caution area as an example, but in addition to this, a fire configured by a multi-layered neural network for arson monitoring performed by installing sensors such as a surveillance camera and a flame detector outdoors. A detector may be provided so that the fire detector can be learned from deep learning and the arson can be monitored.

(Feature extraction)
In the above embodiment, the image is input to the convolutional neural network to extract the features due to the fire, but the preprocessing for extracting the features such as contour and shading from the input image is performed without using the convolutional neural network. It is also possible to extract a predetermined feature and input the extracted image into a fully connected neural network functioning as a recognition unit to estimate whether it is a fire or a non-fire. This makes it possible to reduce the processing load of image feature extraction.

(About learning method)
In the above embodiment, learning by backpropagation is performed, but the learning method of the multi-layer neural network is not limited to this.

(Composite of image and sensor)
In the above embodiment, the fire monitoring by the image and the fire monitoring by the sensor are different forms, but the image data and the sensor data may be handled in parallel as the input information. For image data, for example, a black-and-white value per pixel is treated as an input term, and for sensor data, for example, a detection value for each sensor is treated as an input term. In this case, the term of the intermediate layer from which the features of the image are extracted in the intermediate layer and the term of the intermediate layer affected by the sensor data affect the terms of the intermediate layer after the next stage for determining fire detection. It is desirable as an educational result to be able to give, but it is not limited to this if fire monitoring can be enabled.

(Infrared lighting and infrared image imaging)
In the above embodiment, the surveillance area is imaged by the surveillance camera in the state of using the illumination of the surveillance area and / or in the state of natural light, but the infrared light from the infrared illumination device is irradiated to the surveillance area and the infrared region is sensitive. An infrared image is taken by a surveillance camera and the multi-layered neural network of the judgment device is learned by the back program partition, and the infrared image of the surveillance area is input to the multi-layered neural network to judge whether it is fire or non-fire. You may do so.

By inputting the infrared image of the monitoring area into the determination device in this way, fire monitoring using the monitoring image becomes possible without being affected by the lighting state of the monitoring area, the change in brightness during the day and night, and the like.

(Other)
Further, the present invention is not limited to the above-described embodiment, includes appropriate modifications that do not impair its purpose and advantages, and is not further limited by the numerical values shown in the above-described embodiment.

10: Judgment device 12: Receiver 14 , 14-1, 14-2 : Monitoring area 16 , 16-1, 16-2 : Surveillance camera 18 , 18-1, 18-2 : Fire detector 20: Signal cable 22 : Sensor line 24: Fire detector 26: Recording device 28: Learning image holding unit 30: Learning control unit 32: Image input unit 34: Multi-layer neural network 36: Judgment unit 38: Feature extraction unit 40: Recognition unit 42: Input images 43, 45a, 45b: Weight filters 44a, 44b, 44c: Feature map 46: Input layer 48: Bonding layer 50: Intermediate layer 52: Output layer 60 , 60-1, 60-2 : Analog fire detector 62: Transmission line 64: Time-series data generation unit 66: Time-series data input unit 68: Learning data holding unit

Claims (14)

In a fire monitoring system that detects a fire based on input information using a fire detector composed of a multi-layer neural network.
Have a learning control unit from the deep learning to learn the fire detector,
The learning control unit is a fire monitoring system characterized in that input information corresponding to a fire monitoring result by a receiver connected to a fire detector installed in a monitoring area is learned as learning information.
In the fire monitoring system according to claim 1,
Fire monitoring system wherein the physical quantity detected by the sensor and or, where an image of the monitored area captured by the imaging unit and the input information.
In the fire monitoring system according to claim 2.
When the image of the monitoring area is used as the input information, the fire monitoring system is characterized in that the moving image of the monitoring area captured by the imaging unit is extracted at predetermined intervals to create an image of the monitoring area.
In the fire monitoring system according to any one of claims 1 to 3.
The learning control unit, when the signal by the fire alarm of the fire detector is input, among the input information from a predetermined time before until the input signal by the fire alert, fire detector and fire alert A fire monitoring system characterized in that the input information corresponding to is learned as fire learning information.
In the fire monitoring system according to any one of claims 1 to 3.
The fire detector detects the temperature or smoke concentration and sends the detected analog value to the receiver to judge the fire.
When the detected analog value reaches a predetermined fire judgment level and the receiver determines that a fire has occurred, the learning control unit exceeds a predetermined fire sign level lower than the fire judgment level. A fire monitoring system characterized in that, among the input information from to the time when a fire is determined by the receiver, the input information corresponding to the fire detector that has issued a fire is learned as fire learning information.
In the fire monitoring system according to any one of claims 1 to 3.
When the fire determination signal based on the fire determination operation is input after the fire transfer signal based on the fire alarm of the fire detector is input by the receiver, the learning control unit starts from a predetermined time before. A fire monitoring system characterized in that, of the input information up to the time of input of a fire transfer signal, the input information corresponding to the fire detector that issued a fire is learned as fire learning information.
In the fire monitoring system according to any one of claims 1 to 3.
It said fire detector is detected analog value to determine a fire is sent to the receiver by detecting the temperature or smoke density,
In the learning control unit, after the detection analog value reaches a predetermined fire judgment level and a fire transfer signal based on the fire alarm of the fire detector is input from the receiver, the fire determination transfer based on the fire determination operation is performed. If the broadcast signal is input, among the input information from when the detecting analog value exceeds a predetermined fire indication level lower than the fire determination level until the input of the fire Utsuriho signal, the fire alarm and fire fire monitoring system, characterized in that to learn the input information corresponding to the detector as a fire learning information.
In the fire monitoring system according to claim 1 , 2, 3 or 6.
When the recovery transfer signal based on the recovery operation is input after the fire transfer signal based on the fire alarm of the fire detector is input by the receiver, the learning control unit starts the fire transfer from a predetermined time before. A fire monitoring system characterized in that, of the input information up to the time of input of a signal signal, the input information corresponding to the fire detector that issued a fire is learned as non-fire learning information.
In the fire monitoring system according to claim 1 , 2, 3 or 7.
It said fire detector is to determine the fire is sent to the receiver detected analog value by detecting the temperature or smoke density,
The learning control unit, the detection analog value reaches a predetermined fire determination level, recovery Utsuriho signal fire Utsuriho signal based on fire alert the fire detector from the receiver is based on recovery operations after being entered If There is input, among the input information from when the detecting analog value exceeds a predetermined fire indication level lower than the fire determination level until the input of the fire Utsuriho signal, the fire detector which is fire alert A fire monitoring system characterized in that the input information corresponding to is learned as non-fire learning information.
In the fire monitoring system according to any one of claims 1 to 9.
The multi-layer neural network
It consists of a feature extraction unit and a recognition unit.
The feature extraction unit, a neural network narrowing seen tatami with a plurality of convolutional viewed layer for generating feature information features have been extracted of the input information the input information is input,
The recognition unit, the characteristic information is input, the fire monitoring system is characterized in that a neural network having a plurality of total binding layer for outputting a fire characteristic value output from the neural network narrowing seen the mat.
In the fire monitoring system according to any one of claims 1 to 10.
The learning control unit is characterized in that is learned by back propagation based on the error between the value and the predetermined expectation value is output when you enter learning information to the multilayered neural network of the fire detector Fire monitoring system.
In the fire monitoring system according to any one of claims 1 to 11.
The learning control unit is a fire monitoring system characterized in that input information of a normal monitoring state of the receiver is initialized and learned as non-fire learning information.
In the fire monitoring system according to claim 12,
Timing of the initialization learning, when starting up the apparatus, when a predetermined operation is performed, when there is no substantially change in the input information, see containing a predetermined time interval, any one or more of,
A fire monitoring system characterized in that, in the case of each predetermined time interval, the initialization learning may be performed in a plurality of patterns for the initial time and the time interval.
In the fire monitoring system according to any one of claims 1 to 13.
It said fire detector is a fire monitoring system according to claim Rukoto be displayed about the factors determining the fire addition to the detection of a fire.

Images