JP2011081715A - Monitoring system and monitoring method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a monitor system which can automatically monitor various meters by automatically generating monitoring information from automatically collected measurement values of the meters. <P>SOLUTION: A monitoring system 1 reads a meter 20 by a camera 10 to monitor the variation in measurement of the meter 20, which is recognized by image processing. An image processor 11 extracts an image of the meter 20 as a monitoring object. from image information picked up by the camera 10 and recognizes a measurement shown in a display surface of the meter 20. A management device 12 collects measurements recognized by the image processor 11, via a network 13 and generates time-series information to monitor the variation in measurement of the meter 20. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention particularly relates to a monitoring technique for monitoring a measurement value of a measuring instrument using a camera.

  Conventionally, in facilities such as factories and buildings, an equipment called a panel on which various measuring instruments for monitoring various devices such as air conditioners (hereinafter sometimes referred to as meters) are installed is arranged. Yes. Specifically, it is a control panel, a distribution board, and the like, and various meters such as a watt hour meter and a thermometer are incorporated.

  In recent years, in order to promote energy saving measures, it has been required to collect and aggregate the measurement values (power amount, temperature value, voltage value, current value, etc.) of various meters and measure the energy saving effect. For this reason, a method capable of efficiently collecting the measurement values of various meters without requiring human labor as much as possible is desirable.

  Although it is possible to attach a device that can automatically collect the measurement values of various meters to the panel, the existing panel will be remodeled and difficult to realize easily for reasons such as cost. Thus, as a method for automatically collecting meter measurement values without requiring modification of the panel, a meter recognition system that reads the value of an existing meter using a camera has been proposed (for example, Patent Document 1). reference).

JP 2006-277608 A

  By using the meter recognition system described above, it is possible to realize a method of automatically collecting meter measurement values without requiring modification of the panel. However, for example, information for monitoring the energy saving effect cannot be acquired simply by collecting the measurement values of the meter.

  Accordingly, an object of the present invention is to provide a monitoring system capable of automatically creating monitoring information from automatically collected meter measurement values and automatically monitoring various meters.

  An aspect of the present invention is a monitoring system that monitors, for example, fluctuations in measured values of meters recognized by image processing by reading various meters attached to, for example, a distribution board in a factory with a camera.

  A monitoring system according to an aspect of the present invention includes a camera that captures a display surface showing measurement values of a measuring instrument, image processing means that performs image processing for recognizing the measurement values from video information from the cameras, A management unit that generates time-series information for monitoring a measurement target based on the measurement value recognized by the image processing unit, and that monitors a measurement value variation of the measuring device based on the time-series information. It is.

  According to the present invention, monitoring information can be automatically created from, for example, a server from automatically collected meter measurement values, and various meters can be automatically monitored.

The block diagram for demonstrating the structure of the monitoring system regarding embodiment of this invention. 6 is a flowchart for explaining a procedure of image processing according to the embodiment. The conceptual diagram corresponding to the procedure of the image processing shown in FIG. 2 regarding this embodiment. 6 is a flowchart for explaining a procedure of image processing according to the embodiment. The conceptual diagram corresponding to the procedure of the image processing shown in FIG. 4 regarding this embodiment. It is a figure for demonstrating the view angle deviation correction regarding the modification of this embodiment. It is a figure for demonstrating cooperation with the illumination control regarding the modification of this embodiment. The figure for demonstrating the function which learns the correlation with the angle of a needle | hook and a measured value regarding the modification of this embodiment. The figure for demonstrating the function which searches a meter from the video information from a camera regarding the modification of this embodiment. The figure for demonstrating the function which identifies the kind of meter regarding the modification of this embodiment.

  Embodiments of the present invention will be described below with reference to the drawings.

[System configuration]
FIG. 1 is a block diagram illustrating a main part of a monitoring system according to the present embodiment.

  As shown in FIG. 1, the monitoring system 1 according to this embodiment includes a camera 10, an image processing device 11, a management device 12, and a network 13. The camera 10 is a surveillance camera such as a wide-angle camera having a wide-angle lens or a PTZ (pan / tilt / zoom) PTZ camera. The PTZ camera has a head swing function, a zoom function, and a field angle control function. In the present embodiment, the camera 10 is disposed, for example, in a machine room or facility management room of a factory or building, and is attached to a position where the meter 20 can be imaged on the surface of the panel equipment 2 such as a switchboard or distribution board. .

  The meter 20 is various measuring instruments such as a watt hour meter that measures the amount of power and a water meter that measures the amount of water. The meter 20 has various display methods such as an analog meter that shows a measured value with a needle as shown in FIG. 1 and a digital meter that shows a measured value with a number.

  The image processing apparatus 11 is composed of a computer (hardware and software) installed in, for example, a factory or a building management center, and measures the meter 20 from video information captured by the camera 10 as will be described later. Image processing for recognizing the value is executed.

  The management device 12 is a monitoring server, for example, and is installed in a remote place from the location of the panel equipment 2 to be monitored. The management device 12 is connected to the image processing device 11 via a network 13 such as the Internet or a LAN, and collects measurement values for each meter 20 recognized by the image processing device 11. In the present embodiment, for the sake of convenience, the network 13 is configured to include a communication device for transmitting information that is a recognition result (measurement value) output from the image processing device 11 to the management device 12. The management device 12 creates time-series information from the measured values for each meter 20 received via the network 13, and monitors the fluctuation state of the measurement target (for example, power consumption).

[System monitoring operation]
FIG. 2 is a flowchart for explaining a procedure of image processing (digital image processing) by the image processing apparatus 11 in the system 1 of the present embodiment. FIG. 3 is a conceptual diagram corresponding to each image processing procedure.

  First, when a wide-angle camera is used as the camera 10, the system 1 images a plurality of meters 20 at the same time. Further, when using a PTZ camera as the camera 10, the system 1 takes an image for each meter 20, and circulates each meter 20 to take an image. In any camera, the image processing apparatus 11 performs image processing on video information captured by the camera 10 and recognizes a measurement value displayed on each meter 20.

  Here, as shown in FIG. 3A, it is assumed that each camera 20 that displays a measurement value with characters (numbers) is imaged by the camera 10. As shown in FIG. 3A, the image processing apparatus 11 executes luminance expansion processing (density processing) included in the image processing filter processing (step S1). In FIG. 3A, the mask area is an area where the meter image does not originally exist, and is an area for determining the occurrence of the angle of view of the camera 10. Further, as shown in FIG. 3B, the image processing apparatus 11 executes an edge extraction process for extracting the contour of the meter (step S2).

  Next, as shown in FIG. 3C, the image processing apparatus 11 performs binarization processing on the edge extracted image (step S3). As shown in FIG. 3D, the image processing apparatus 11 performs a labeling process for recognizing the graphic component of each meter on the binarized image (step S4). Thereafter, the image processing apparatus 11 executes meter extraction processing including rectangular processing shown in FIG. 3E and character region extraction processing shown in FIG. 3F (step S5).

  By the meter extraction process as described above, the image processing apparatus 11 recognizes a character string that is a measurement value from a character area obtained by imaging the display surface of each meter 20, as shown in FIG. Character recognition processing is executed (step S6). The measurement value (character string information) of each meter 20 recognized by the image processing apparatus 11 is transmitted to the management apparatus 12 via the network 13. The management device 12 creates time-series information from the measurement values for each meter 20, and monitors the fluctuation state of the measurement target (for example, power consumption).

  FIG. 4 is a flowchart for explaining an image processing procedure in the system 1 according to the present embodiment assuming a case where the camera 20 captures an image of the meter 20 that displays a measurement value with a needle as shown in FIG. FIG. 5 is a conceptual diagram corresponding to each image processing procedure shown in FIG.

  Even in this case, the image processing apparatus 11 sequentially executes a luminance expansion process, an edge extraction process, a binarization process, a labeling process, and a meter extraction process in the same manner as described above (steps S11 to S15). That is, the image processing apparatus 11 executes edge extraction processing for extracting the contour of the meter after the luminance expansion processing (step S12). Next, as shown in FIG. 5A, the image processing apparatus 11 executes a binarization process on the edge-extracted image, and executes a labeling process on the binarized image (step S13). , S14). Thereafter, as shown in FIG. 5B, the image processing apparatus 11 executes meter extraction processing for removing other than the needle image (step S15).

  Next, as shown in FIG. 5C, the image processing apparatus 11 executes a needle extraction process (straight line extraction process) for extracting the needle figure of the meter 20 (step S16). The image processing apparatus 11 calculates the angle of the extracted needle and calculates a measurement value based on the angle (step S17). At this time, a correction process based on a view angle shift of the camera 10 is executed (step S18). Then, the image processing apparatus 11 outputs a measurement value based on the needle angle of the meter 20 as a read value (step S19). The measurement value (display information at the needle angle) of the meter 20 output as a read value by the image processing device 11 is transmitted to the management device 12 via the network 13.

  As described above, the system of the present embodiment captures the display surface of each meter 20 by the monitoring camera 10 such as a wide-angle camera or a PTZ camera, and performs image processing on the captured video information to each meter 20. Recognizes the measurement value displayed by the character string or the angle of the needle. The management device 12 collects measurement values recognized for each meter 20 and creates time-series information from each measurement value, thereby monitoring the variation state of the measurement target (for example, power consumption).

  As described above, when a wide-angle camera is used as the camera 10, a plurality of meters 20 can be imaged simultaneously. Further, when a PTZ camera is used as the camera 10, it is possible to take an image by patroling each meter 20 at regular intervals. That is, it is possible to set the timing for reading the measurement value from each meter 20 according to the type of the camera 10.

(First modification)
As a modification of the present embodiment, when a PTZ camera is used as the camera 10, not only monitoring at a fixed timing or cyclic interval but also learning a fluctuation range of each meter and automatically monitoring at an optimal cyclic interval It is the monitoring system 1 which performs.

  Specifically, for example, the meter 1 is an air conditioner-related measuring device, the meter 2 is an illumination-related measuring device, and the meter 3 is an equipment-related measuring device. The monitoring system 1 monitors, for example, the meter 1 that consumes power most intensely, and then monitors the facility-related meter 3 in detail, for example, in the summer daytime patrol. As a specific example, the monitoring system 1 circulates from meter 1 to meter 2 and meter 3, and further circulates from meter 1 to meter 1, meter 3, meter 1, and meter 1, and circulates from meter 2 to meter 3. .

  In addition, the monitoring system 1 finely monitors the meter 2 that consumes power most intensely, for example, in the autumn night patrol, and then closely monitors the facility-related meter 3. As a specific example, the monitoring system 1 circulates from meter 1 to meter 2, meter 3, and meter 2, and further circulates from meter 2 to meter 3, meter 2, meter 1, meter 2, and meter 3.

  In short, the system 1 according to the method of the present modification learns different power consumption characteristics and the like for each place such as an office, a factory, and a department store, and also learns characteristics for each season and time zone. Based on such learning results, the system 1 can diversify the monitoring operation of each meter by circulating the camera 10 using the PTZ camera in the optimal time interval and the order of the circulation. Become.

(Second modification)
FIGS. 6A to 6D are diagrams for explaining the correction of the angle of view of the camera 10 as a modification of the present embodiment.

  This modification example assumes an analog meter that indicates a measurement value corresponding to the angle of the needle as shown in FIG. 6A, and relates to a process for correcting a field angle deviation that occurs according to the imaging direction of the camera 10. That is, as shown in FIG. 6C, the angle of the needle shown on the display surface of the meter 20 is greater when the camera 10 is imaged from the front than when the camera 10 is imaged from the front. Different. FIG. 6B shows a case where the display surface of the meter is imaged from the diagonal left direction by the camera 10. FIG. 6D shows the case where the camera 10 images the display surface of the meter from the lower diagonal direction.

  In such a case, the image processing device 11 detects a field angle deviation of the camera 10 and corrects a read value (measured value) corresponding to the angle of the needle based on the amount of the field angle deviation. Thereby, when the display surface of the meter 20 is imaged from the oblique direction by the camera 10, the meter 20 varies depending on the imaging direction of the camera 10 in order to correct a reading value that generates an error according to the angle of view of the camera 10. Variations in readings (measured values) from can be compensated.

(Third Modification)
FIG. 7 is a modified example of the present embodiment and is a diagram for explaining a cooperation function with illumination control.

  The present modification is a system that performs illumination control for controlling illumination of the display surfaces of the meters 20a and 20b in cooperation with the monitoring operation when monitoring a plurality of types of meters 20a and 20b. The meters 20a and 20b are, for example, an air conditioning meter 1 and an illumination meter 2, respectively.

  When the camera 10 reads the measurement value from the display surface of the meter 20a, the image processing device 11 turns on the illumination device 15 that illuminates the display surface of the meter 20a to be monitored via the illumination control device 14. At this time, if the meter 20b is not monitored, the image processing device 11 turns off the illumination device 15 that illuminates the display surface of the meter 20b via the illumination control device 14.

  On the other hand, when reading a measurement value from the display surface of the meter 20b by the camera 10, the image processing device 11 turns on the illumination device 15 that illuminates the display surface of the meter 20b to be monitored via the illumination control device 14. At this time, when the meter 20a is not monitored, the image processing device 11 turns off the illumination device 15 that illuminates the display surface of the meter 20a via the illumination control device 14.

  As described above, according to the present modification, the illumination control for controlling the illumination of the display surfaces of the meters 20a and 20b is linked in cooperation with the monitoring operation of the meters 20a and 20b, so that the camera 10 with respect to the meter to be monitored is linked. A reliable imaging operation can be realized. When neither meter 20a nor 20b is monitored, the image processing apparatus 11 turns off the lighting device 15 corresponding to each. Thereby, useless power consumption required for illumination can be saved. Further, when the reading of the monitored meter is finished, the corresponding lighting device is turned off. This has the effect of preventing the reflection of illumination when reading for another meter is started, as well as saving unnecessary power consumption.

(Fourth modification)
In the monitoring system 1 of the present modification, the management device 12 has a function of learning a past measurement value variation of the meter and predicting a future meter measurement value. Specifically, the meter 20 such as a watt hour meter increases in the order of 0, 1, 2, 3, 4, 5, 6, 7, 8, and 9, for example, a measured value of “56418”. On the other hand, it is determined that the next measurement value is “56419”. Moreover, the management apparatus 12 has a function which estimates the measured value of the child meter which belongs to the parent meter by arrangement when reading the measured value of the parent meter.

  In short, with the system 1 according to the present modification, the management device 12 can correct an error in the measurement value read by the image processing device 11 using a function for predicting or estimating the measurement value. . Therefore, even when a variation occurs in the imaging operation of the camera 10 or the image processing of the image processing device 11, the management device 12 can compensate for the variation by using the function of predicting or estimating the measurement value. Become.

(Fifth modification)
FIGS. 8A and 8B are diagrams for explaining a function of learning a correlation between a meter needle angle and a measured value, in particular, a modification of the present embodiment.

  The system 1 according to this modification includes an image processing device 11 that learns the correlation between the reading value (measured value) of the meter 20 and the state of the meter 20 that is video information from the camera, and stores this learning information. It is. The image processing apparatus 11 recognizes the reading value (measurement value) of the meter 20 using the stored learning information at the time of the next meter reading operation.

  Specifically, as shown in FIGS. 8A and 8B, the scale plate (measurement value display surface) is different for each meter. Therefore, the image processing apparatus 11 stores learning information indicating the correlation between the meter needle angle and the scale value (measurement value) as table information, and uses the table information when image processing is performed on video information from the camera. refer. That is, the image processing apparatus 11 reads the measurement value indicated by the meter 20 from the correlation between the angle of the meter needle that is the state of the meter 20 and the measurement value.

  In short, in the system 1 of this modification, the image processing apparatus 11 refers to the learning information (table information) indicating the correlation between the angle of the meter needle and the measurement value, and thus different scale plates for each meter. It is possible to reliably recognize the measurement value read from the.

(Sixth Modification)
FIGS. 9A and 9B are diagrams illustrating a function of searching for a meter from video information from a camera according to a modification of the present embodiment.

  As shown in FIG. 9A, the system 1 according to this modification includes an image processing apparatus 11 that stores learning information for specifying a meter by designating the meter in video information from the camera 10. It is a configuration. As shown in FIG. 9B, the image processing apparatus 11 has a feature (template, identification information) that the learning information has a black edge with the meter, the upper 3/4 has a white scale plate, and the lower 1/4 has a black plate. ) To get. Next, when the video information from the camera 10 is input, the image processing apparatus 11 automatically specifies a meter from the video information using features obtained from the learning information.

  Note that the image processing apparatus 11 may store, as learning information for specifying a meter, information indicating a place where the meter is arranged instead of a feature (template, identification information). In this case, when image information corresponding to the location from the camera 10 is input, the image processing apparatus 11 automatically specifies a meter from the video information using information indicating the location obtained from the learning information. To do.

  In short, with the system 1 according to the present modification, the image processing apparatus 11 refers to the learning information for identifying the meter, and reliably identifies and extracts the meter in the video information from the camera 10. It is possible to reliably execute the meter extraction process.

(Seventh Modification)
FIGS. 10A and 10B are diagrams for explaining a function of specifying the type of meter to be monitored as a modification of the present embodiment.

  As shown in FIG. 10A, the system 1 according to this modification has a configuration including an image processing device 11 that identifies the type of meter to be monitored (measuring target of the meter) based on the meter layout. . In addition, as shown in FIG. 10B, the system 1 of this modification identifies the type of meter to be monitored (measuring object of the meter) based on the label attached to the meter or the shape of the meter. The image processing apparatus 11 is included.

  The image processing apparatus 11 stores in advance table information (identification information) in which information indicating a meter layout, meter shape, and label is associated with a meter type (meter measurement target). When the image processing apparatus 11 extracts information indicating the layout map, the shape of the meter, and the label from the video information from the camera 10, the image processing apparatus 11 refers to the table information and identifies the type of meter to be monitored (measurement target of the meter). .

  In short, in the system 1 of the present modification, the image processing apparatus 11 can identify the type of meter to be monitored (measurement target of the meter) from the video information from the camera 10, and thus recognizes the measurement value of the meter. Sometimes it becomes possible to improve the reading accuracy of the measurement value.

(Eighth modification)
As a modification of the present embodiment, the system 1 is configured not to constantly monitor the meter 20 but to perform a security monitoring operation in a time zone in which the meter 20 does not need to be monitored. That is, the system 1 of this modification is configured to use a security function together. Specifically, when a wide-angle camera is used as the camera 10, the camera 10 is installed at an angle of view that can capture not only a meter but also a specific area of a machine room, a hallway, and the like. When a PTZ camera is used, an angle of view capable of capturing an image of a specific area of a machine room or a corridor is set as one of presets.

  In the case of the system 1 that also uses the security function as described above, for example, when monitoring the meter 20 of the air conditioning system in the office, the power consumption is intense during the daytime. read. On the other hand, since the power consumption is low at night, the system 1 can use various processes such as reading the measurement value in a round with a rough time interval for monitoring the meter 20 together with the security function.

  Note that the present invention is not limited to the above-described embodiment as it is, and can be embodied by modifying the constituent elements without departing from the scope of the invention in the implementation stage. In addition, various inventions can be formed by appropriately combining a plurality of constituent elements disclosed in the embodiment. For example, some components may be deleted from all the components shown in the embodiment. Furthermore, constituent elements over different embodiments may be appropriately combined.

DESCRIPTION OF SYMBOLS 1 ... Surveillance system, 2 ... Board equipment, 10 ... Camera, 11 ... Image processing apparatus,
12 ... management device, 13 ... network, 20 ... meter.

Claims (10)

A camera that shoots the display surface showing the measurement value of the measuring instrument;
Image processing means for performing image processing for recognizing the measurement value from video information from the camera;
Management means for generating time series information for monitoring a measurement target based on the measurement value recognized by the image processing means and monitoring measurement value fluctuations of the measuring instrument based on the time series information. A monitoring system characterized by The image processing means includes
The monitoring system according to claim 1, wherein the measurement value is recognized at different time intervals depending on a measuring instrument or a time zone. The image processing means includes
Based on the measured value fluctuation range of the measuring instrument,
The measurement value is recognized in a relatively short time interval for the measuring instrument having a large measurement value fluctuation range, and the relatively long time interval is provided for the measuring instrument having a small measurement value fluctuation range. The monitoring system according to claim 2, wherein the measurement value is recognized. The measuring instrument is an analog meter or a digital meter,
The image processing means includes
It is configured to execute either image processing for recognizing an analog measurement value of the analog meter or image processing for recognizing a digital measurement value of the digital meter from video information from the camera. The monitoring system according to any one of claims 1 to 3. The image processing means includes
Means for detecting an angle of view shift of the camera;
The monitoring system according to claim 1, further comprising: a unit that corrects an error of the measurement value according to the angle of view deviation. A lighting device having a region including a display surface of the measuring instrument as a lighting target region;
The monitoring according to claim 1, further comprising: an illumination control unit that controls lighting / extinction of the lighting device in cooperation with an imaging operation of the display surface by the camera. system. The management means includes
Means for storing in a storage means information relating to a measurement value variation for each measuring instrument based on the time series information;
The monitoring system according to any one of claims 1 to 6, further comprising a predicting unit that predicts a future measurement value by using the stored information. The image processing means includes
Storing learning information obtained by learning the correlation between the video information from the camera and the measured value,
The monitoring system according to any one of claims 1 to 7, wherein the monitoring system is configured to execute image processing for recognizing the measurement value using the learning information. The image processing means includes
The monitoring system according to any one of claims 1 to 8, further comprising means for specifying the measuring instrument based on identification information of the measuring instrument included in video information from the camera. A monitoring method applied to a monitoring system that uses a camera that captures a display surface showing a measurement value of a measuring instrument,
Image processing for recognizing the measurement value from video information from the camera;
Processing for generating time-series information for monitoring a measurement object based on the measurement value recognized by the image processing;
And a process of monitoring a measurement value fluctuation of the measuring instrument based on the time series information.

Images