JP5674499B2 - Sensing device - Google Patents

Description

  The present invention relates to an apparatus for processing a detection result of a sensor that measures a physical environment.

  In fields where the physical environment at multiple points needs to be managed, such as in the agricultural field, the detection results of the sensors installed at each point are collected and the surrounding physical environment is understood without visiting the site, making management work more efficient There is a technology to make it. In the technique described in Patent Document 1 below, the state of a farm is photographed by an image sensor and the image is transmitted to a server so that the growing state of the crop can be grasped remotely.

  On the other hand, the sensor may need to be calibrated periodically because the measured value may shift due to aging. In addition, the measured value may be shifted due to a sensor failure. Japanese Patent Application Laid-Open No. 2004-228561 describes a method for elucidating the cause of a defect by comparing a past measurement result and an abnormal measurement result in a maintenance system for maintaining a working machine such as a harvesting machine.

JP 2002-176855 A JP 2002-190871 A

  In general, since calibration of a sensor is performed periodically, if a failure occurs in the sensor, the failure is detected after the calibration is performed and dealt with. Therefore, there is a tendency that a period from when the sensor breaks down until the abnormality is removed tends to be long, and there is a possibility that the accuracy of the measurement data is lowered and the original management work is hindered.

  In this regard, the techniques described in Patent Documents 1 and 2 above are considered to have the same problem as described above with respect to the period from when an abnormality occurs until calibration or cause elucidation is performed. In addition, in the technique described in Patent Document 2, the cause of the failure is clarified in comparison with the past measurement result, but since the comparison target is only the same device, it is an unexpected occurrence that does not result from the failure of the device. Even if the measurement result fluctuates temporarily due to an environmental change or the like, it may be determined that the device is abnormal.

  The present invention has been made to solve the above-described problems, and an object of the present invention is to provide a sensing device capable of appropriately comparing the measurement results of a plurality of sensors.

  The sensing device according to the present invention classifies the measurement result of the sensor using the attribute of the installation location of the sensor. The classification result is presented to the user.

  According to the sensing device of the present invention, measurement results having the same tendency can be compared with each other by classifying sensor measurement results for each attribute. Thereby, a some sensor measurement result can be compared appropriately and abnormality can be discovered at an early stage.

1 is a functional block diagram of a sensing device 100 according to Embodiment 1. FIG. It is a figure which shows the structure and example of data of the sensor data. It is a figure which shows the structure of the attribute data 122, and a data example. The attribute data 122 is data describing the attribute of the place where each sensor is installed. It is a conceptual diagram which shows a mode that the sensing apparatus 100 classify | categorizes the measurement result of each sensor. It is a figure which shows the detail of the program module contained in the processing engine 131. FIG. It is a processing flow explaining the process which the sensing apparatus 100 classifies sensor data. It is a module block diagram of the processing engine 131 with which the sensing apparatus 100 which concerns on Embodiment 2 is provided. It is a module block diagram of the processing engine 131 with which the sensing apparatus 100 concerning Embodiment 3 is provided. It is a module block diagram of the processing engine 131 with which the sensing apparatus 100 concerning Embodiment 4 is provided. FIG. 10 is a functional block diagram of a sensing device 100 according to a fifth embodiment. FIG. 10 is a functional block diagram of a sensing device 100 according to a sixth embodiment.

<Embodiment 1>
FIG. 1 is a functional block diagram of a sensing device 100 according to Embodiment 1 of the present invention. The sensing device 100 is a device that receives and accumulates sensor data describing the results of the measurement of the physical environment by the sensor. The sensing device 100 includes a calculation unit 110, a storage unit 120, a memory 130, a display 140, a printer 150, a keyboard 160, and a mouse 170. Prepare. In addition, a network interface (not shown) is provided, and sensor data can be received from each sensor via the network.

  The calculation unit 110 is a calculation device such as a CPU (Central Processing Unit), and executes each program stored in the memory 130. In addition, each functional unit of the sensing device 100 is controlled. In the following description, for convenience of description, each program may be described as an operation subject. However, it is added that an arithmetic device such as the arithmetic unit 110 actually executes these programs.

  The storage unit 120 is a storage device such as a hard disk device, and stores sensor data 121 and attribute data 122. Details of these data will be described later with reference to FIGS.

  The memory 130 is a storage device that holds data necessary for the operation of the arithmetic unit 110, and stores the processing engine 131, the processing engine management unit 132, the arithmetic information management unit 133, and the setting condition management unit 134.

  The processing engine 131 is a program that describes a procedure for processing sensor data received from each sensor by the sensing device 100. Details will be described later with reference to FIG. The processing engine management unit 132 is a program that issues an instruction to each program module included in the processing engine 131. The calculation information management unit 133 is a program that manages the processing result of the processing engine 131. The setting condition management unit 134 is a program that manages processing conditions set by the user of the sensing device 100.

  The display 140 is a display device that displays a processing result of the processing engine 131 on the screen. The printer 150 is a device that prints the processing results of the processing engine 131. These serve as an output unit that outputs the processing result of the sensing device 100.

  The keyboard 160 and the mouse 170 are input devices that a user of the sensing device 100 gives an operation instruction to the sensing device 100.

  FIG. 2 is a diagram illustrating a configuration of the sensor data 121 and a data example. Sensors installed in various places measure the surrounding physical environment (for example, temperature, surrounding image, etc.), and transmit sensor data describing the measurement results to the sensing device 100. The sensing device 100 receives the sensor data and accumulates it as sensor data 121 in the storage unit 120.

  The sensor data 121 includes a data ID field 1211, a sensor ID field 1212, a platform number field 1213, and a data field that holds measurement results of each sensor. Here, as an example, the date / time field 1214, the temperature field 1215, and the rainfall field 1216 are shown. However, in the case of using a sensor that measures other physical environments, a data field for holding each measurement result may be provided. Good.

  The data ID field 1211 is a serial number assigned for convenience by the sensing device 100 to identify each sensor data. The sensor ID field 1212 is an ID for identifying the sensor that has measured the sensor data and transmitted it to the sensing device 100. The platform number field 1213 is an ID of a measuring device equipped with a sensor identified by the value of the sensor ID field 1212. Since each measuring device is installed at a specific location, the value in this field corresponds to the location of each sensor. The same measuring device may have a plurality of sensors. Even if the value of the platform number field 1213 is different, the installation location of the measuring device itself may be the same.

  When each sensor transmits sensor data to the sensing device 100, the values of the sensor ID field 1212 and the platform number field 1213 are described in the sensor data. The sensing device 100 takes out those values and stores them in the fields.

  The date / time field 1214, the temperature field 1215, and the rainfall field 1216 hold the date / time when each sensor acquired the measurement result, the measurement result of the ambient temperature, and the measurement result of the ambient rainfall. Each sensor describes each measurement result and measurement date and time in the sensor data and transmits them to the sensing device 100. The sensing device 100 takes out those values and stores them in the fields.

  FIG. 3 is a diagram illustrating a configuration of the attribute data 122 and a data example. The attribute data 122 is data describing the attribute of the place where each sensor is installed. The attribute of the place where each sensor is installed is an attribute that can be a basis for collecting the same kind of measurement results of each sensor, such as the geographical position of the place and the type of physical environment measured by the sensor. For example, since it is assumed that temperature sensors installed in areas close to each other will present similar measurement results, it can be considered that they can be handled as measurement results belonging to the same category.

  Since the installation location of the sensor can be identified by the value of the platform number field 1213 described with reference to FIG. 2, the platform number field 1221 is also provided in the attribute data 122. Regarding the other data fields, it is preferable to provide a field that can appropriately classify the sensor installation location according to individual specifications such as the measurement target of each sensor. In general, it is considered that the attribute of the sensor installation location can be well described by the sensor position and the sensor type.

  The attribute data 122 includes a platform number field 1221, a latitude / longitude field 1222, and a sensor type field 1223.

  The platform number field 1221 is a data field corresponding to the platform number field 1213 of the sensor data 121. The latitude / longitude field 1222 holds a value indicating the geographical position of the sensor installation location identified by the value of the platform number field 1221. The sensor type field 1223 holds a value indicating the type of physical environment detected by the sensor at the sensor installation location identified by the value of the platform number field 1221.

  The configuration of the sensing device 100 has been described above. Next, the operation of the sensing device 100 will be described.

  FIG. 4 is a conceptual diagram showing how the sensing device 100 classifies the measurement results of each sensor. The sensor 401 measures the surrounding physical environment and transmits sensor data to the sensing device 100. While the normally operating sensor 401 transmits a normal measurement result, the malfunctioning sensor 402 may transmit an abnormal measurement result.

  When the measurement results of each sensor are individually observed, it is difficult to determine whether or not the measurement results are abnormal. This is because even if the measurement result changes abruptly, there is a possibility that the physical environment itself measured by the sensor has changed abruptly, not the failure of the sensor itself. In this case, the sensor itself is operating normally and should not be treated as abnormal.

  Therefore, in the present invention, it is assumed that sensors having similar attributes are collected and classified for each attribute (classification 400, 410, 420), and sensors belonging to the same classification will present similar measurement results.

  For example, when the same kind of sensors installed in close proximity are classified into the same classification, the sensors belonging to the same classification should present the same measurement result. When only some sensors present measurement results different from those of other sensors belonging to the same category, some of the sensors may be broken. On the other hand, when the physical environment itself measured by the sensor changes abnormally, it is assumed that other sensors belonging to the same classification will also change.

  The sensing device 100 utilizes the above-described method, classifies each sensor data according to the attribute of the sensor installation location, and enables each sensor data to be compared for each classification. As a result, the user can grasp the measurement result while comparing each sensor data with the other sensor data, so that when a sensor abnormality occurs, it can be immediately discovered.

  FIG. 5 is a diagram showing details of program modules included in the processing engine 131. The processing engine 131 includes a data receiving unit 1311, an attribute reading unit 1312, an attribute classification unit 1313, a comparison processing unit 1314, and an output unit 1315. These functional units are configured as program modules included in the processing engine 131.

  The data receiving unit 1311 receives sensor data from each sensor via the network, and stores it as sensor data 121 in the storage unit 120. The attribute reading unit 1312 reads the attribute of the installation location of each sensor data from the attribute data 122. The attribute classification unit 1313 classifies the attributes read by the attribute reading unit 1312 using an arbitrary method. Based on the processing result of the attribute classification unit 1313, the comparison processing unit 1314 performs processing that allows the user to compare sensor data belonging to the same classification. For example, processing such as aligning sensor data belonging to the same classification along the measurement time can be considered. The output unit 1315 outputs the processing result of the comparison processing unit 1314 to the display 140 or the printer 150.

FIG. 6 is a process flow illustrating a process in which the sensing device 100 classifies sensor data. Hereinafter, each step of FIG. 6 will be described.
(FIG. 6: Step S600)
The user of the sensing device 100 instructs the sensing device 100 to classify and compare the sensor data using the keyboard 160 or the like. The calculation unit 110 receives the instruction and starts this processing flow. It is assumed that the sensing device 100 has previously acquired sensor data from each sensor and has accumulated it as sensor data 121.
(FIG. 6: Step S601)
The attribute reading unit 1312 reads the sensor data 121 and acquires a corresponding record of the attribute data 122 using the value of the platform number field 1213 of each sensor data as a key. The attribute reading unit 1312 recognizes the attribute information described in the record acquired from the attribute data 122 as an attribute at the installation location of the sensor data. The attribute reading unit 1312 performs the same processing for all the records held in the sensor data 121, and associates each sensor data with the attribute of the installation location.

(FIG. 6: Step S602)
The attribute classification unit 1313 classifies the attribute of each sensor data read by the attribute reading unit 1312 in step S601 in a form suitable for determination of sensor abnormality. As a result of this step, each sensor data belongs to one of the categories according to the category to which the attribute value belongs. As the classification criteria, for example, the following examples can be considered.
(FIG. 6: Step S602: Classification example 1)
The attribute classification unit 1313 may create one classification for each attribute, or may combine a plurality of attributes into one classification.
(FIG. 6: Step S602: Classification example 2)
The attribute classification unit 1313 calculates the similarity between attribute values for attributes that are difficult to classify clearly, such as latitude / longitude, and sensor data having attribute values with high similarity belong to the same classification. You may do it. Specifically, sensor data that fall within a certain range of calculated similarity may belong to the same classification.
(FIG. 6: Step S602: Classification example 3)
The attribute classification unit 1313 may calculate a distance between attribute values, and may cause sensor data having attribute values that are close to each other to belong to the same classification. Alternatively, the attribute values may be classified by a known method such as clustering.
(FIG. 6: Step S602: Classification example 4)
The attribute classification unit 1313 does not necessarily need to use all attributes as classification criteria. For example, when classifying the installation location attribute of the temperature sensor, it is only necessary that the installation location assumed to have similar weather conditions belong to the same classification, such as geographical location, altitude, terrain information, climate information, etc. It is sufficient to classify any of these attributes as a criterion. Note that at the time when the attribute reading unit 1312 reads the attribute from the attribute data 122, only the attribute necessary as the classification criterion may be read.

(FIG. 6: Step S603)
The comparison processing unit 1314 performs processing so that the user can compare sensor data that belong to the same category in step S602.
(FIG. 6: Step S603: Supplement)
The purpose of this step is to make it easier for the user to identify sensor data belonging to the same category that has an abnormal value, so it is easy to visually identify the trend of sensor data within the category and It is desirable to perform a process that can be identified. For example, sensor data belonging to the same classification may be arranged in a tabular form so that the acquisition times are aligned. In addition, it is conceivable to obtain a difference between the data average value in the classification and each sensor data, to obtain a function that approximates the temporal change of each sensor data, and to create a scatter diagram.
(FIG. 6: Step S604)
The output unit 1315 outputs the processing result of step S603 to the display 140 or the printer 150.

<Embodiment 1: Summary>
As described above, the sensing device 100 according to the first embodiment classifies the attribute of the installation location of the sensor data describing the measurement result, and presents at least the classification result to the user on the display 140 or the like. As a result, since the user can compare sensor data belonging to the same classification, even if the sensor abnormality is difficult to grasp only by a single sensor data, it is abnormal compared to other sensor data. Whether or not there is an abnormality can be easily identified.

  In addition, the sensing device 100 according to the first embodiment calculates a tendency (average value, approximate function, etc.) of sensor data belonging to the same classification, and outputs the tendency together with the classification result on the display 140 or the like. Thereby, the user can easily grasp the sensor data belonging to the same classification that does not match the tendency of the other sensor data.

<Embodiment 2>
In the first embodiment, the processing result of the comparison processing unit 1314 is presented to the user on the display 140 or the like, so that the user can easily identify the sensor abnormality. On the other hand, it is also conceivable to automatically identify sensor data having a tendency that is clearly different from other sensor data belonging to the same classification, and estimate that the data is abnormal.

  Therefore, in the second embodiment of the present invention, a configuration example in which abnormal sensor data is estimated based on the processing result of the comparison processing unit 1314 and presented to the user will be described. Since the other configuration is the same as that of the first embodiment, the following description focuses on the differences.

  FIG. 7 is a module configuration diagram of the processing engine 131 provided in the sensing device 100 according to the second embodiment. In the second embodiment, the processing engine 131 newly includes an abnormality detection unit 1316 in addition to the configuration described in FIG. 5 of the first embodiment. The abnormality detection unit 1316 is configured as a program module included in the processing engine 131 like other function units.

  The abnormality detection unit 1316 receives the processing result of the comparison processing unit 1314, identifies sensor data having a tendency different from other sensor data among the sensor data belonging to the same classification, and the sensor that acquired the sensor data has an abnormality. Judge that it has occurred. The abnormality detection unit 1316 delivers the abnormality determination result and the processing result of the comparison processing unit 1314 to the output unit 1315. The output unit 1315 presents these to the user on the display 140 or the like.

  As an example of processing in which the abnormality detection unit 1316 detects abnormal sensor data, an example in which an image of a crop is acquired as sensor data and an example in which the communication amount abnormally varies will be described below.

(Processing example 1 of the abnormality detection unit 1316: crop image)
It is assumed that the spectrum sensor measures the image spectrum of the crop and transmits it to the sensing device 100. Even for the same crop in the same region, the growth stage may differ depending on the planting time, and the spectrum data of each crop varies individually. In particular, the difference in spectral data is large during the growth period. For this reason, even if there is sensor data different from other sensor data, it is not easy to determine whether the sensor data is abnormal or whether the crop growth itself is different from the others. In this case, it is only necessary to acquire the spectrum data in time series, identify the position where each spectrum data is best matched while shifting the reference time, and search for the spectrum data that is extremely different from the others based on the position. If the time-dependent changes of each spectral data are the same and the deviation is within a predetermined range, it is considered that the growth stages of some of the crops are only different from others. If the deviation does not fall within the predetermined range and is extremely large, it is considered that an abnormality has occurred in the sensor that measured the spectrum data of the crop.

(Processing example 2 of the abnormality detection unit 1316: traffic)
Each sensor is assumed to be a movement sensor configured to be able to change the installation location. Each sensor transmits sensor data including its own position information to the sensing device 100. The sensing device 100 keeps movement sensors that are close to each other from belonging to the same classification. The abnormality detection unit 1316 monitors fluctuations in the amount of received data from the movement sensors belonging to the same classification, and determines that the movement sensor is abnormal when the change in the data amount is significantly different from the reference value. Alternatively, an average movement speed of movement sensors belonging to the same category is set as a reference value, and if the movement speed of any movement sensor is significantly different from the reference value, it is determined that an abnormality has occurred in the movement sensor. You can also.

<Embodiment 2: Summary>
As described above, the sensing device 100 according to the second embodiment determines that sensor data having a tendency different from the tendency of other sensor data belonging to the same classification is abnormal based on the processing result of the comparison processing unit 1314. Then, the abnormality detection result is presented to the user. Thereby, the user can identify abnormal data that is not noticed only by his / her own judgment.

<Embodiment 3>
In the first and second embodiments, it has been described that the sensing device 100 classifies the sensor data 121 according to the attribute. In the third embodiment of the present invention, a configuration example in which the user selects an attribute serving as a reference for classifying sensor data will be described. Since other configurations are the same as those in the first and second embodiments, the following description will focus on the differences.

  FIG. 8 is a module configuration diagram of the processing engine 131 provided in the sensing device 100 according to the third embodiment. In the third embodiment, the processing engine 131 includes a new attribute designation input unit 1317 in addition to the configurations described in the first and second embodiments. Here, an example in which an attribute designation input unit 1317 is provided in addition to the configuration described with reference to FIG.

  The attribute designation input unit 1317 is configured as a program module included in the processing engine 131 like other function units. The attribute designation input unit 1317 receives a designation input that designates an attribute serving as a reference for classifying sensor data from an input device such as the keyboard 160 and outputs the designation input to the attribute reading unit 1312. The attribute reading unit 1312 reads the attribute designated by the user, and the attribute classification unit 1313 classifies the sensor data based on the attribute. Other attributes may be used supplementarily or ignored.

  One or more attributes may be selected by the user. Alternatively, instead of specifying the attribute itself, a criterion for classifying data may be input in a natural language. For example, a character string such as “a neighboring lettuce growing in a similar environment” may be input to the attribute designation input unit 1317. In this case, the attribute designation input unit 1317 derives an attribute item that can be a classification criterion from the received character string, and outputs the attribute item to the attribute reading unit 1312.

  In the case of the above character string example, a plurality of attributes corresponding to “similar environments” can be classification criteria. For example, the position information of the sensor, the tendency of temperature change, the sensor whose measurement target is lettuce, the integrated temperature from planting to the present time, etc. can be classification criteria. The attribute reading unit 1312 may adopt all of these classification criteria, or may adopt only the classification criteria that are presumed to well reflect the conditions specified by the user.

<Embodiment 3: Summary>
As described above, the sensing device 100 according to the third embodiment includes the attribute designation input unit 1317 for a user to designate and input an attribute that is a reference for classifying sensor data. As a result, the user can classify the sensor data according to the desired criteria, and thus can easily grasp the significance of the classification result performed by the sensing device 100.

<Embodiment 4>
In the first to third embodiments, the example in which the sensing device 100 includes the attribute data 122 in advance has been described. However, there may be cases where attribute information cannot be obtained for all sensor installation locations. Therefore, in a fourth embodiment of the present invention, a configuration example in which sensor data is analyzed to extract attribute information of the sensor installation location will be described. Since other configurations are the same as those in the first to third embodiments, the following description will focus on differences.

  FIG. 9 is a module configuration diagram of the processing engine 131 provided in the sensing device 100 according to the fourth embodiment. In the fourth embodiment, the processing engine 131 newly includes an attribute extraction unit 1318 in addition to the configurations described in the first to third embodiments. Here, an example in which an attribute extraction unit 1318 is provided in addition to the configuration described in FIG.

  The attribute extraction unit 1318 is configured as a program module included in the processing engine 131 like other function units. The attribute extraction unit 1318 analyzes the sensor data received by the data reception unit 1311 and extracts the attribute information of the sensor installation location from the information included in the measurement result. The sensor data 121 once stored in the storage unit 120 may be acquired and analyzed. The following is an example in which the attribute extraction unit 1318 extracts attribute information.

(Processing example 1 of attribute extraction unit 1318 1: crop image)
It is assumed that the spectrum sensor measures the image spectrum of the crop and transmits it to the sensing device 100. The attribute extraction unit 1318 analyzes the crop spectrum data included in the sensor data, and estimates the type of the crop. For example, the estimation may be performed using a classification method based on supervised classification. The attribute extraction unit 1318 acquires a wide range image of the area where the crop is growing, and grasps a wide range of vegetation situations on the image by using a vegetation index that can be calculated from the spectrum data. The attribute extraction unit 1318 can estimate the type of crop by comparing the connection method between pixels, spectrum information, and the like with the teacher data. Further, since the shape of the crop can be acquired if it is a spectrum image acquired at a closer position, it is possible to specify the crop type whose shape matches by a technique such as template matching. The specified crop type can be used as a classification standard.

(Processing example 2 of attribute extraction unit 1318: When the sensor position is changed)
When the position of the sensor is managed in the latitude / longitude field 1222, it is necessary to match the value of the field with the actual sensor position. Although both values can be manually adjusted by the user, when sensor position information is included in the sensor data, the value of the same field can be extracted as attribute information from the position information. For example, the attribute extraction unit 1318 acquires the value of the corresponding latitude / longitude field 1222 in the attribute data 122 using the sensor ID and platform number included in the sensor data as keys, and is included in the received sensor data. If the position information and the value of the latitude / longitude field 1222 are different, the value of the field can be updated using a new value.

<Embodiment 4: Summary>
As described above, the sensing device 100 according to the fourth embodiment extracts the attribute information of the sensor installation location from the sensor data received by the data receiving unit 1311 and stores the attribute information in the attribute data 122 or the attribute reading unit 1312. Output. The attribute reading unit 1312 reads from the attribute data 122 an attribute that matches the attribute extracted by the attribute extracting unit 1318. Thereby, even when the attribute information of the sensor installation location cannot be obtained in advance or when the attribute is changed, it is possible to achieve consistency with the actual installation environment.

<Embodiment 5>
FIG. 10 is a functional block diagram of the sensing device 100 according to Embodiment 5 of the present invention. In the sensing device 100 according to the fifth embodiment, the storage unit 120 newly stores work support data 123 in addition to the configurations described in the first to fourth embodiments.

  The work support data 123 is data describing the work content to be performed at the sensor installation location in accordance with the sensor data trend, that is, data describing the correspondence between the sensor data trend and the work content to be performed.

  In step S603, the comparison processing unit 1314 calculates a tendency of the sensor data in the same classification, and then reads a record that matches the tendency from the work support data 123. The output unit 1315 presents the work support content to the user on the display 140 or the like. Thereby, work support at the sensor installation location can be implemented. Hereinafter, an example will be described in which the spectrum sensor acquires a spectrum image and transmits it to the sensing device 100.

  If the comparison processing unit 1314 detects an abnormal tendency of the spectrum data and determines that the sensor is not due to a sensor failure, the physical environment itself measured by the sensor may be abnormal. Therefore, the comparison processing unit 1314 reads the corresponding work support information from the work support data 123 and presents it to the user on the display 140 or the like. Thereby, the user can assist the work for eliminating the abnormal situation.

  In this case, the work support data 123 describes the correspondence between abnormal conditions such as spectrum data on crops / soil, spectrum data in disease occurrence state, spectrum data in water shortage condition, and work contents to be performed in the abnormal state. Holds data.

  The comparison processing unit 1314 searches for work support information that most closely matches the trend of the spectrum data in which an abnormality has occurred. At this point in time, the sensor installation location where the abnormality has occurred and the surrounding physical environment have been specified. Therefore, the search target in the work support data 123 may be narrowed down using the crop name, the region name, or the like. In addition, for example, a method of specifying environmental information such as that the high temperature period continues from the measurement result of the temperature sensor and searching for work support information corresponding to the high temperature failure may be considered.

  In addition, it is also possible to specify the degree of damage based on the measurement result and read out work support information corresponding to the degree of damage. For example, an example is possible in which the progress of pest damage is specified based on a spectrum image, and a measure corresponding to the progress is retrieved from the work support data 123.

<Embodiment 5: Summary>
As described above, the sensing device 100 according to the fifth embodiment reads out work support information corresponding to the tendency of sensor data and presents it to the user. Thereby, the user can promptly implement an appropriate measure according to the measurement result of the sensor installation location.

<Embodiment 6>
FIG. 11 is a functional block diagram of the sensing device 100 according to Embodiment 6 of the present invention. The sensing device 100 according to the sixth embodiment includes a new attribute information creation unit 135 in addition to the configurations described in the first to fifth embodiments. Here, the example provided with the attribute information creation part 135 in addition to the structure demonstrated in FIG. 5 was shown. Since other configurations are the same as those of the first to fifth embodiments, the following description will focus on the differences.

  In the fourth embodiment, the example in which the attribute information of the sensor installation location is extracted from the sensor data has been described. However, the attribute information cannot always be extracted from the sensor data. Therefore, the attribute information creation unit 135 obtains the attribute information of the sensor installation location using the information that the external data source has, and creates the attribute data 122. Examples of external data sources include geological maps, past weather statistics, elevation data, satellite images, maps, and the like. Hereinafter, an example in which land conditions are used as the attribute data 122 will be described.

(Example of creating attribute data 122 from an external data source: Part 1)
In general, it is assumed that the same kind of sensors installed in geographically close positions obtain similar measurement results. However, even if geographically close, the measurement results may actually differ because the climate varies depending on factors such as elevation differences and topography. Since it is difficult to extract information such as altitude difference and topography from sensor data, it is desirable to provide information using information from an external data source. The attribute information creation unit 135 acquires the altitude data of the sensor installation location from an external data source and records it in the attribute data 122. The attribute classification unit 1313 causes sensor data measured at a location close to the altitude to belong to the same classification. Furthermore, even if the altitude is the same, the climate may change due to the shadows of the mountains, etc., so an example of additionally using the sunshine condition calculated from the topography as an attribute is also conceivable.

(Example of creating attribute data 122 from an external data source: Part 2)
The attribute information creation unit 135 can also extract the land use situation using the spectrum image taken by the satellite and record the land use situation in the attribute data 122. Land use status refers to the use of land, such as urban areas, upland fields, bare land, and roads. Furthermore, altitude information and land use status can be combined and used as attribute data 122. For example, an urban area with a low altitude, a field crop with a high altitude, and the like.
(Example of creating attribute data 122 from an external data source: Part 3)
The attribute information creation unit 135 can acquire past meteorological data, distinguish, for example, a region that becomes hot and a region that becomes low in the summer and use it as the attribute data 122.

<Embodiment 7>
In the above first to sixth embodiments, the method of classifying the attribute of the sensor installation location has been described. Other classification methods are as follows.
(Example of classification method 1: Supplement between different sensors)
In the above first to sixth embodiments, in principle, the measurement results of the same type of sensors belong to the same class, but different types of correlated sensors belong to the same class and can be used complementarily. . For example, the measurement results of air temperature sensors and water temperature sensors placed in close proximity are thought to vary with the same tendency, so by mutually converting using an appropriate function, the mutual tendency can be comprehended. can do. In addition, since it is considered that the temperature decreases as the altitude increases, it is also possible to use the measurement results of the altimeter and the measurement results of the thermometer in a complementary manner.

(Example of classification method 2: crop image and other sensor measurement values)
A wide range of satellite images of farmland can be taken and parameters such as vegetation indices can be used to estimate the state of crop growth. Since the growth state of the crop and the physical environment such as temperature, season, and weather trend have a certain degree of correlation, they can be used complementarily. Similarly, the shape of the crop can be obtained from the close image of the crop, but since the shape of the crop represents the growing state, it can be used in the same manner as described above.

  As mentioned above, the invention made by the present inventor has been specifically described based on the embodiment. However, the present invention is not limited to the embodiment, and various modifications can be made without departing from the scope of the invention. Needless to say.

  In addition, each of the above-described configurations, functions, processing units, etc. can be realized as hardware by designing all or a part thereof, for example, with an integrated circuit, or the processor executes a program for realizing each function. By doing so, it can also be realized as software. Information such as programs and tables for realizing each function can be stored in a storage device such as a memory or a hard disk, or a storage medium such as an IC card or a DVD.

  100: Sensing device, 110: arithmetic unit, 120: storage unit, 121: sensor data, 1211: data ID field, 1212: sensor ID field, 1213: platform number field, 1214: date and time field, 1215: temperature field, 1216: Rainfall field, 122: attribute data, 1221: platform number field, 1222: latitude / longitude field, 1223: sensor type field, 123: work support data, 130: memory, 131: processing engine, 1311: data receiving unit, 1312 : Attribute reading unit, 1313: attribute classification unit, 1314: comparison processing unit, 1315: output unit, 1316: abnormality detection unit, 1317: attribute designation input unit, 1318: attribute extraction unit, 132: processing engine management unit, 133: Calculation information management unit, 134: setting condition managing unit, 135: attribute information generating unit, 140: Display, 150: printer 160: Keyboard, 170 Mouse.

Claims (7)

A receiver that receives sensor data describing the measurement results of a plurality of sensors that measure the physical environment at different installation locations ;
A storage unit for storing attribute data describing each attribute of location of the plurality of sensors,
An attribute reading unit that reads out the attribute of the installation location corresponding to each sensor data received by the receiving unit from the attribute data stored in the storage unit;
An attribute classification unit for classifying the sensor data based on the attribute read by the attribute reading unit;
A comparison processing unit that calculates a tendency of measured values of the physical environment indicated by the sensor data belonging to the same classification obtained as a result of the classification by the attribute classification unit,
Although the sensor data belongs to the same classification classified by the attribute classification unit, the sensor data indicating the measured value that does not match the calculated trend value of the physical environment is specified, and the sensor corresponding to the specified sensor data is identified. An anomaly detector that determines that an anomaly has occurred;
In addition to the result of the attribute classification unit classifying the sensor data , the comparison processing unit calculates the tendency of the sensor data and outputs the result of comparison with each other, and
A sensing device comprising: The sensor is
It is configured so that the installation location can be changed,
The sensor data is
Describes the location of the sensor;
The abnormality detection unit
When the change rate of the installation location of the sensor is greater than or equal to a reference value, or when the change in the amount of communication from the sensor set at the same location is greater than or equal to a reference value, an abnormality has occurred in the sensor The sensing device according to claim 1, wherein The sensing device according to claim 1, wherein the attribute classification unit calculates the similarity of the attribute read by the attribute reading unit and classifies the sensor data based on the similarity. The sensing apparatus according to claim 1, further comprising an attribute designation input unit that receives an attribute designation that designates the attribute used when the attribute classification unit classifies the sensor data. An attribute extraction unit that analyzes the characteristics of the measurement result described by the sensor data and acquires an attribute of the installation location of the sensor corresponding to the measurement result;
The attribute reading unit
The sensing device according to claim 1, wherein an attribute that matches the attribute extracted by the attribute extraction unit is read from the attribute data stored in the storage unit. The storage unit
Storing work support data describing a correspondence relationship between the measured value of the physical environment and the work content to be performed at the location according to the measured value of the physical environment ;
The comparison processing unit
Read the work content corresponding to the trend of the measurement value of the physical environment from the work support data,
The output unit includes a sensing device according to claim 1, wherein the outputting the work support data. The sensing apparatus according to claim 1, further comprising an attribute information creation unit that obtains an attribute of the installation location of the sensor from an external data source or calculates using information obtained from the external data source.

Images