WO2019215924A1

WO2019215924A1 - Operation data classification system, operation data classification method, and program

Info

Publication number: WO2019215924A1
Application number: PCT/JP2018/018380
Authority: WO
Inventors: 俊二菅谷
Original assignee: 株式会社オプティム
Priority date: 2018-05-11
Filing date: 2018-05-11
Publication date: 2019-11-14

Abstract

[Problem] To provide an operation data classification system, an operation data classification method, and a program with which it is possible to classify to which operation newly acquired data relates, and generate a timeline for a person or device that performs the operation. [Solution] This operation data classification system comprises: an acquisition module 211 which acquires a plurality of sets of basic data relating to various operations; an extraction module 212 which extracts feature quantities of the plurality of sets of basic data; a storage module 231 which associates and stores the feature quantities with the operations; a determination module 213 which determines which feature quantity of the stored operation data is similar to a feature quantity of unknown basic data; a classification module 214 which, on the basis of the determination result, classifies to which operation the unknown basic data relates; and a timeline generation module 215 which generates a timeline for each person or device on the basis of the operation times associated with the basic data.

Description

Work data classification system, work data classification method, and program

The present invention obtains data related to various operations such as image data from cameras, logs from machines, logs of smartphone terminals held by users, etc., and classifies which operation is related to unknown data, The present invention relates to a work data classification system, a work data classification method, and a program capable of generating a timeline for the work.

A method has been proposed that allows each sensor data to be properly classified using only sensor data collected in time series from a large number of sensors even in an environment where noise exists and cannot be measured separately. (Patent Document 1).

JP2016-099888

However, in the method of Patent Document 1, even in an environment where noise exists and the noise cannot be separated and measured, each sensor data is appropriately classified using only sensor data collected in time series from many sensors. It is possible to make it possible, but it is not possible to automatically classify what kind of work each sensor data is. Therefore, it is not possible to generate a timeline for the classified work.

When actually assuming construction sites, construction sites, store sites, production sites, farm work, office work, etc., work of the target person or device from images taken by the camera and data acquired from the sensors of each device It is desirable to classify the contents automatically, and further to automatically generate a timeline consisting of work contents and work time.

The present invention can acquire a plurality of basic data related to various operations, classify which basic data the unknown basic data is based on from the feature amount, and generate a timeline for the classified operations. It is an object to provide a work data classification system, work data classification method, and program.

The present invention provides the following solutions.

The invention according to the first feature is
An acquisition means for acquiring a plurality of basic data relating to various operations performed by a person or a device;
Extraction means for extracting feature quantities of the plurality of basic data;
Storage means for storing the feature quantity and the work in association with each other;
Determining means for determining which feature quantity of the feature data of the unknown basic data is similar to the feature quantity in the feature data of the stored basic data;
Based on the result of the determination, classification means for classifying what the unknown basic data is data about what work by the person or device;
Timeline generating means for determining the content of work from the classified data and generating a timeline or a percentage of work time for each person or device based on the work time associated with the basic data;
A work data classification system is provided.

According to the first aspect of the invention, in the work data classification system, an acquisition unit that acquires a plurality of basic data relating to various operations performed by a person or a device, and an extraction unit that extracts feature amounts of the plurality of basic data And storage means for associating and storing the feature quantity and the work, and determination means for determining which feature quantity in the feature quantity of the stored basic data is similar to the feature quantity of the stored basic data And, based on the result of the determination, classifying means for classifying what the unknown basic data is data relating to what work by the person or equipment, and determining the content of the work from the classified data, Timeline generating means for generating a timeline or a ratio of work time for each person or device based on the work time associated with the basic data.

The invention according to the first feature is a category of the work data classification system, but the work data classification method and program have the same operations and effects.

The invention according to the second feature is a work data classification system which is the invention according to the first feature,
Schedule storage means for storing a schedule of work performed by the person or device;
A schedule comparison means for comparing the timeline generated by the timeline generation means with the stored schedule;
A work data classification system is provided.

According to the second aspect of the invention, in the work data classification system according to the first aspect of the invention, the schedule storage means for storing a work schedule performed by the person or the device, and the timeline generation means And a schedule comparison means for comparing the stored timeline with the stored schedule.

The invention according to the third feature is a work data classification system which is the invention according to the first feature or the second feature,
The determination means performs machine learning of past feature amounts, and determines which feature amount in the feature amount of the basic data is similar to the feature amount of the unknown basic data. Provide a data classification system.

According to the invention relating to the third feature, in the work data classification system which is the invention relating to the first feature or the second feature, the determination means performs machine learning of a past feature amount, and the unknown basis It is determined which of the feature quantities of the basic data is similar to the feature quantity of the basic data.

The invention according to the fourth feature is a work data classification system which is the invention according to any one of the first feature to the third feature,
The schedule comparison unit provides an operation data classification system that emphasizes and outputs when the generated timeline is delayed with respect to the schedule or when the difference is large.

According to the invention according to the fourth feature, in the work data classification system according to any one of the first feature to the third feature, the schedule comparison means is configured to generate the timeline generated with respect to the schedule. Is output when it is delayed or when the difference is large.

The invention according to the fifth feature is
Obtaining a plurality of basic data on various tasks performed by a person or device;
Extracting features of the plurality of basic data;
Storing the feature quantity and the work in association with each other;
Determining which feature quantity of the unknown basic data feature quantity is similar to the feature quantity of the stored basic data; and
Classifying the unknown basic data based on the result of the determination as to what work by the person or device;
Determining the content of work from the classified data, and generating a timeline or percentage of work time for each person or device based on the work time associated with the basic data;
A work data classification method comprising:

The invention according to the sixth feature is
Work data classification system
Obtaining a plurality of basic data relating to various operations performed by a person or device;
Extracting features of the plurality of basic data;
Storing the feature quantity and the work in association with each other;
Determining which feature quantity of the unknown basic data feature quantity is similar to the feature quantity of the stored basic data;
Classifying the unknown basic data based on the result of the determination as to what work by the person or device,
Determining the content of work from the classified data, and generating a timeline or a percentage of work time for each person or device based on the work time associated with the basic data;
Provide a program to execute.

According to the present invention, acquiring a plurality of basic data relating to various operations, classifying which basic data the unknown basic data is based on from the feature amount, and further generating a timeline for the classified operations It is possible to provide a work data classification system, a work data classification method, and a program that can be used.

FIG. 1 is a schematic diagram of a preferred embodiment of the present invention. FIG. 2 is a diagram illustrating the functional blocks of the apparatus 100 and the computer 200 and the relationship between the functions. FIG. 3 is a flowchart of the timeline generation process. FIG. 4 is a flowchart when the computer 200 classifies the unknown basic data according to the number of determinations. FIG. 5 is an example of a flowchart for determining whether or not the feature amount of unknown basic data is similar to the feature amount of stored work data in the computer 200. FIG. 6 is a flowchart in the case of performing determination processing by machine learning of a combination of past work and feature values of basic data. FIG. 7 is a diagram illustrating the functional blocks of the apparatus 100 and the computer 200 and the relationship between the functions when performing schedule comparison processing. FIG. 8 is a flowchart for performing the schedule comparison process. FIG. 9 is an example of a table showing the data structure of work and basic data stored in the storage unit 230. FIG. 10 is an example of a table showing the data structure of unknown basic data. FIG. 11 is an example of a table including determination results of stored basic data having a feature quantity similar to unknown basic data. FIG. 12 is an example of the timeline output display of the present invention. FIG. 13 is an example of a detailed display of work status. FIG. 14 is an example of a work time ratio display. FIG. 15 is an example of the schedule comparison display of the present invention.

Hereinafter, the best mode for carrying out the present invention will be described with reference to the drawings. This is merely an example, and the technical scope of the present invention is not limited to this.

[Outline of work data classification system]
FIG. 1 is a schematic diagram of a preferred embodiment of the present invention. The outline of the present invention will be described with reference to FIG. The work data classification system includes an apparatus 100, a computer 200, and a communication network 300.

In FIG. 1, the number of devices 100 is not limited to one and may be plural. Here, a WEB camera is illustrated as the device 100A, and a wearable device is illustrated as the device 100B. Further, the computer 200 is not limited to a real device, and may be a virtual device. The computer 200 may be the same device as the device 100.

The apparatus 100 includes a sensor unit 10, a control unit 110, a communication unit 120, and a storage unit 130 as shown in FIG. As shown in FIG. 2, the computer 200 includes a control unit 210, a communication unit 220, a storage unit 230, and an input / output unit 240. The control unit 210 implements the acquisition module 211 in cooperation with the communication unit 220 and the storage unit 230. Further, the control unit 210 implements an extraction module 212, a determination module 213, and a classification module 214 in cooperation with the storage unit 230. In addition, the control unit 210 implements the timeline generation module 215 in cooperation with the storage unit 230 and the input / output unit 240. The storage unit 230 implements the storage module 231 in cooperation with the control unit 210. The communication network 300 may be a public communication network such as the Internet or a dedicated communication network, and enables communication between the apparatus 100 and the computer 200.

The apparatus 100 includes the sensor unit 10 capable of acquiring images, machine logs, and data of various sensors, and is an apparatus capable of data communication with the computer 200. Here, as an example, a WEB camera is illustrated as the apparatus 100A, and a wearable device is illustrated as the apparatus 100B. However, a digital camera, a digital video, a security camera, an in-vehicle camera, a 360-degree camera, an industrial apparatus, an agricultural apparatus, a drone , A wearable device, etc., and a device having necessary functions. Various work data may be stored in the storage unit 130.

The computer 200 is a computing device capable of data communication with the device 100. Here, a desktop computer is illustrated as an example, but in addition to a mobile phone, a portable information terminal, a tablet terminal, a personal computer, electrical appliances such as a netbook terminal, a slate terminal, an electronic book terminal, and a portable music player Or a wearable terminal such as a smart glass or a head-mounted display.

In the work data classification system of FIG. 1, first, the storage module 231 of the computer 200 stores in the storage unit 230 the work associated with the features of a plurality of basic data (step S01). The work associated with the feature quantities of the plurality of basic data may be acquired from another computer or a storage medium, or may be created by the computer 200. Further, a dedicated database may be provided in the storage unit 230. In the present invention, the basic data is a generic term that represents images, machine logs, and data of various sensors acquired from the apparatus 100. The basic data always includes information for specifying the time of work.

FIG. 9 is an example of a table showing the data structure of the work stored in the storage unit 230 and the basic data. Here, two basic data are stored as operation X, of which the feature amount of data A is v, the data type is moving image, the feature amount of data B is w, and the data type is It indicates that the sensor is an acceleration sensor. Also, as the operation Y, three basic data are stored, of which the feature amount of the data C is x, the type of data is moving image, the feature amount of the data D is y, and the type of data is machine The feature amount of the log and data E is z, which indicates that the data type is GPS. Each data itself may be stored together with these data. FIG. 9 shows an example in which the storage destination of each data itself is described in the rightmost column of the table.

Returning to FIG. 1, the apparatus 100 transmits unknown basic data to the computer 200 (step S02), and the acquisition module 211 of the computer 200 acquires unknown basic data (step S03). When there are a plurality of unknown basic data, all of them are acquired as work data relating to one work. Here, the flow for transmitting unknown basic data from the device 100 has been described. However, the acquisition module 211 instructs the device 100 to transmit unknown basic data, and the device 100 receives the unknown basic data in response to the instruction. May be sent. The acquisition module 211 not only acquires the work data that the apparatus 100 has acquired in real time, but also acquires the work data that the apparatus 100 has acquired in the past and stored in the storage unit 130. Good.

FIG. 10 is an example of a table showing the data structure of unknown basic data. Here, the work consists of three basic data, of which the feature amount of data F is s, the data type is GPS, the feature amount of data G is t, the data type is machine log, and data H The feature amount is u, indicating that the data type is a moving image. Each data itself may be stored together with these data. FIG. 10 shows an example in which the storage destination of each data itself is described in the rightmost column of the table.

1 again, the extraction module 212 of the computer 200 extracts the feature quantity of the unknown basic data acquired in step S03 (step S04). The feature amount here may be, for example, according to the type of data, or may be determined by analyzing the content of the data, and depending on the system. It may be used as For example, if the data type is a moving image, the image obtained by analyzing the moving image may be used as the feature amount. If the data type is the acceleration sensor, the operation amount may be used as the feature amount. For example, an analysis of machine operation and time may be used as the feature amount, and in the case of GPS data, the location, height, and weather may be used as the feature amount with reference to map data, weather data, and the like.

Next, the determination module 213 of the computer 200 determines which basic data feature quantity stored in the storage unit 230 is similar to the characteristic quantity of unknown basic data (step S05). Details of the determination method will be described later.

FIG. 11 is an example of a table including determination results of stored basic data having a feature quantity similar to unknown basic data. Here, the feature amount of data F is s and similar to data E, the feature amount of data G is t and similar to data D, the feature amount of data H is u and similar to data A This is shown as an example.

Next, the classification module 214 of the computer 200 classifies which work the acquired unknown basic data is related to (step S06). 9 and 11, the data F is similar to the data E of the work Y, the data G is similar to the data D of the work Y, and the data H is similar to the data A of the work X. I understand. As a classification method, for example, when classification is performed with priority on the number of determinations of basic data, based on the result that two of three sensors are similar to work Y and one is similar to work X, data F, data G, The work based on data H can be classified as work Y. Or, when classification is performed with priority given to the type of data, for example, if the classification result based on moving images is prioritized, the data H gives priority to the determination result that the work is similar to the data A of the work X, and the data The work by F, data G, and data H can be classified as work X. Here, the case where it is determined that there is stored basic data having a similar feature amount in the unknown basic data is described. However, when it is determined that there is no similar stored basic data, the classification module 214 performs the determination. , Unclassifiable or unclassified.

Finally, the timeline generation module 215 of the computer 200 determines the content of the work based on the result classified in step S06, and generates a timeline based on the work time associated with the acquired basic data ( Step S07). When generating the timeline, a timeline is generated for each person or device performing the work X. Alternatively, the ratio of the work time for each person or device performing the work X is generated. In the present invention, the timeline includes work contents and work time. Here, when generating a timeline or generating a work time ratio, information for specifying a work time included in the basic data is used. In order to generate a timeline of work time for each person or device, information on which device is the basic data acquired is used together. The generated timeline information may be output by display or voice.

FIG. 12 is an example of the timeline output display of the present invention. Here, an example is shown in which the work status of a certain construction site from 8:00 to 18:00 on February 2, 2018 is displayed as a timeline. There are hydraulic excavator A, hydraulic excavator B, crawler dump A, bulldozer A, vibration roller A, dump truck A, dump truck B, worker A, worker B, worker C as the person or equipment who performed the work. The timeline for each work is displayed. Here, for example, when the user of the work classification system selects the public road movement that is the work from 13:00 to 17:00 of the dump truck A, as shown in a balloon 1201, “13:00 departure 15:30 point The output may be performed so that a detailed history within the work can be understood, such as “Z 17:00 arrival”.

FIG. 13 is an example of detailed display of work status. In the screen of FIG. 12, it is assumed that the user of the work classification system selects the slope formation that is the work of the excavator A from 8:00 to 12:00. As shown in a balloon 1301 in FIG. 13, an image, a schematic diagram, etc. based on information included in the acquired basic data so that the work position and work status of each time from 8:00 to 12:00 can be understood. You may display with.

FIG. 14 is an example of a work time ratio display. Instead of timeline generation, work time ratio generation may be performed. Here, the window 1401 shows the ratio of work during the work time of February 2, 2018, 9 hours. In this example, the percentage of each work during the work time is displayed as a percentage for each of the excavator A, the excavator B, the crawler dumper A, and the worker A as the person or equipment that performed the work. Since not all work of a person or equipment can be specified, the total percentage of work time does not necessarily need to be 100%. In FIG. 14, the operations of all the people or devices are collectively displayed. However, individual display may be performed for each person or for each device.

As described above, according to the present invention, a plurality of basic data relating to various works is obtained, and the work related to the unknown basic data is classified from the feature amount, and the timeline for the classified work is further classified. Alternatively, it is possible to provide a work data classification system, a work data classification method, and a program capable of generating a work time ratio.

[Description of each function]
FIG. 2 is a diagram illustrating the functional blocks of the apparatus 100 and the computer 200 and the relationship between the functions. The apparatus 100 includes a sensor unit 10, a control unit 110, a communication unit 120, and a storage unit 130. The computer 200 includes a control unit 210, a communication unit 220, a storage unit 230, and an input / output unit 240. The control unit 210 implements the acquisition module 211 in cooperation with the communication unit 220 and the storage unit 230. Further, the control unit 210 implements an extraction module 212, a determination module 213, and a classification module 214 in cooperation with the storage unit 230. Further, the control unit 210 implements a timeline generation module 215 in cooperation with the storage unit 230 and the input / output unit 240. The storage unit 230 implements the storage module 231 in cooperation with the control unit 210. The communication network 300 may be a public communication network such as the Internet or a dedicated communication network, and enables communication between the apparatus 100 and the computer 200.

The apparatus 100 includes the sensor unit 10 capable of acquiring images, machine logs, and data of various sensors, and is an apparatus capable of data communication with the computer 200. Here, as an example, a WEB camera is illustrated as the apparatus 100A, and a wearable device is illustrated as the apparatus 100B. However, a digital camera, a digital video, a security camera, an in-vehicle camera, a 360-degree camera, an industrial apparatus, an agricultural apparatus, a drone , A wearable device, etc., and a device having necessary functions. Various basic data may be stored in the storage unit 130.

The apparatus 100 includes a sensor capable of acquiring data of images, machine logs, and various sensors as the sensor unit 10. Further, the obtained data is assumed to have the accuracy necessary for extracting the feature amount.

The control unit 110 includes a CPU (Central Processing Unit), a RAM (Random Access Memory), a ROM (Read Only Memory), and the like.

As the communication unit 120, a device for enabling communication with other devices, for example, a WiFi (Wireless Fidelity) compatible device compliant with IEEE 802.11 or an IMT-2000 standard such as a third generation or fourth generation mobile communication system. Compliant wireless device etc. It may be a wired LAN connection.

The storage unit 130 includes a data storage unit such as a hard disk or a semiconductor memory, and stores captured images, necessary data such as imaging conditions, and the like.

The computer 200 is a computing device capable of data communication with the device 100. Here, a desktop computer is illustrated as an example, but in addition to a mobile phone, a portable information terminal, a tablet terminal, a personal computer, electrical appliances such as a netbook terminal, a slate terminal, an electronic book terminal, and a portable music player Or a wearable terminal such as a smart glass or a head-mounted display. Further, the computer 200 is not limited to a real device, and may be a virtual device. The computer 200 may be the same device as the device 100.

The control unit 210 includes a CPU, RAM, ROM, and the like. The control unit 210 implements the acquisition module 211 in cooperation with the communication unit 220 and the storage unit 230. Further, the control unit 210 implements an extraction module 212, a determination module 213, and a classification module 214 in cooperation with the storage unit 230. Further, the control unit 210 implements a timeline generation module 215 in cooperation with the storage unit 230 and the input / output unit 240.

A device for enabling communication with other devices as the communication unit 220, for example, a WiFi compatible device compliant with IEEE802.11 or a wireless device compliant with the IMT-2000 standard such as a third generation or fourth generation mobile communication system Etc. It may be a wired LAN connection.

The storage unit 230 includes a data storage unit such as a hard disk or a semiconductor memory, which associates features with features of a plurality of basic data, teacher data, determination results, classification results, data necessary for processing, etc. Remember. The storage unit 230 implements the storage module 231 in cooperation with the control unit 210. In addition, the storage unit 230 may include a database for storing a work associated with feature quantities of a plurality of basic data.

Suppose that the input / output unit 240 has functions necessary for using the work data classification system. As an example for realizing the input, a liquid crystal display that realizes a touch panel function, a keyboard, a mouse, a pen tablet, a hardware button on the apparatus, a microphone for performing voice recognition, and the like can be provided. Further, as an example for realizing the output, forms such as a liquid crystal display, a PC display, a projection on a projector, and an audio output can be considered. The function of the present invention is not particularly limited by the input / output method.

[Timeline generation processing]
FIG. 3 is a flowchart of the timeline generation process. Processing executed by each module described above will be described in accordance with this processing.

First, the storage module 231 of the computer 200 stores in the storage unit 230 the work associated with the feature quantities of a plurality of basic data (step S301). The basic data is a generic term that represents images, machine logs, and data of various sensors acquired from the apparatus 100. The basic data always includes information for specifying the time of work. The work associated with the feature quantities of the plurality of basic data may be acquired from another computer or a storage medium, or may be created by the computer 200. Further, a dedicated database may be provided in the storage unit 230. The process of step S301 is skipped when there is already stored an association between the work and the feature values of the plurality of basic data, or there is no association between the work to be newly stored and the feature data of the basic data. I shall do it.

Returning to FIG. 3, the acquisition module 211 of the computer 200 requests the apparatus 100 to transmit basic data (step 302). When there are a plurality of apparatuses 100, it is assumed that transmission of basic data is requested to all apparatuses.

In response to the basic data transmission request from the computer 200, the control unit 110 of the apparatus 100 stores the basic data in the storage unit 130 (step S303).

Then, the device 100 transmits unknown basic data to the computer 200 via the communication unit 120 (step S304).

The acquisition module 211 of the computer 200 acquires unknown basic data (step S305). When there are a plurality of devices 100, that is, when there are a plurality of unknown basic data, all are acquired as basic data relating to one operation. The acquisition module 211 may not only acquire basic data acquired in real time by the device 100 but also acquire basic data acquired by the device 100 in the past and stored in the storage unit 130.

3 again, the extraction module 212 of the computer 200 extracts the feature quantity of the unknown basic data acquired in step S305 (step S306). The feature amount here may be, for example, according to the type of data, or may be determined by analyzing the content of the data, and depending on the system. It may be used as For example, if the data type is a moving image, the image obtained by analyzing the moving image may be used as the feature amount. If the data type is the acceleration sensor, the operation amount may be used as the feature amount. For example, an analysis of machine operation and time may be used as the feature amount, and in the case of GPS data, the location, height, and weather may be used as the feature amount with reference to map data, weather data, and the like.

Next, the determination module 213 of the computer 200 determines which basic data feature quantity stored in the storage unit 230 is similar to the characteristic quantity of unknown basic data (step S307). Details of the determination method will be described later.

Next, the classification module 214 of the computer 200 classifies which work the acquired unknown basic data is related to (step S308). 9 and 11, the data F is similar to the data E of the work Y, the data G is similar to the data D of the work Y, and the data H is similar to the data A of the work X. I understand. As a classification method, for example, when classification is performed with priority on the number of determinations of basic data, based on the result that two of three sensors are similar to work Y and one is similar to work X, data F, data G, The work based on data H can be classified as work Y. Or, when classification is performed with priority given to the type of data, for example, if the classification result based on moving images is prioritized, the data H gives priority to the determination result that the work is similar to the data A of the work X, and the data The work by F, data G, and data H can be classified as work X. Here, the case where it is determined that there is stored basic data having a similar feature amount in the unknown basic data is described. However, when it is determined that there is no similar stored basic data, the classification module 214 performs the determination. , Unclassifiable or unclassified.

Finally, the timeline generation module 215 of the computer 200 determines the content of the work based on the result classified in step S308, and generates a timeline based on the work time associated with the acquired basic data ( Step S309). When generating the timeline, a timeline is generated for each person or device performing the work X. Alternatively, the ratio of the work time for each person or device performing the work X is generated. In the present invention, the timeline includes work contents and work time. Here, when generating a timeline or generating a work time ratio, information for specifying a work time included in the basic data is used. In order to generate a timeline of work time for each person or device, information on which device is the basic data acquired is used together. The generated timeline information may be output by display or voice.

[Classification process according to the number of judgments of unknown basic data]
FIG. 4 is a flowchart when the computer 200 classifies the unknown basic data according to the number of determinations. As a configuration, it is assumed that the configuration is the same as that of the apparatus 100 and the computer 200 of FIG. This process corresponds to step S307 and step S308 in the flowchart of FIG. Hereinafter, basic data classification processing according to the number of unknown basic data determinations will be described as processing after the flow up to step S306 in FIG. Here, for the sake of explanation, the data examples shown in FIGS. 9, 10, and 11 are used.

For the unknown basic data acquired in step S305, the feature amount of each basic data has been extracted in step S306. Here, the number of acquired unknown basic data is counted (step S401). In the example of FIG. 10 described above, the number of unknown basic data is three.

Next, one unknown basic data is selected (step S402). Here, it is assumed that data F in FIG. 10 is selected.

The determination module 213 of the computer 200 determines which feature value of the basic data stored in the storage unit 230 is similar to the feature value of the unknown basic data (step S403). Here, it is assumed that the feature amount of the data F is determined to be similar to the data E as shown in FIG.

Next, the determination module 213 confirms whether determination of all unknown basic data has been completed (step S404).

At this point, since the process has not been completed, the process returns to step S402 to select one unknown basic data. Here, it is assumed that data G in FIG. 10 is selected.

Next, in step S403, the determination module 213 determines that the feature amount of the data G is similar to the data D as shown in FIG.

The determination module 213 confirms again whether or not determination of all unknown basic data has been completed in step S404.

At this point in time, determination of all basic data has not been completed yet, so the process returns to step S402 and one unknown basic data is selected. Here, it is assumed that the data H in FIG. 10 is selected.

Next, in step S403, the determination module 213 determines that the feature amount of the data H is similar to the data A as shown in FIG.

The determination module 213 confirms whether or not determination of all unknown basic data has been completed in step S404.

Here, since the determination of all the basic data has been completed, the process proceeds to step S405, and the unknown basic data is classified into the work having a large number determined to be similar to the feature value of the related basic data. Here, since the data F is similar to the data E of the work Y, the data G is similar to the data D of the work Y, and the data H is similar to the data A of the work X, three unknowns Among the basic data, the number determined to be similar to the feature value of the work Y is two, and the number determined to be similar to the feature value of the work X is one. Therefore, the classification module 214 classifies the data F, data G, and data H as basic data related to the work Y.

As described above, according to the present invention, a plurality of basic data relating to various works is acquired, a certain regularity and correlation of the series of works is determined as a feature quantity, and the feature quantity and the work are associated with each other. By storing and performing basic data classification processing according to the number of determinations of unknown basic data, it is possible to appropriately classify which work the unknown basic data is.

[Feature determination processing]
FIG. 5 is an example of a flowchart for determining whether or not the feature quantity of the unknown basic data is similar to the feature quantity of the stored basic data in the computer 200. As a configuration, it is assumed that the configuration is the same as that of the apparatus 100 and the computer 200 of FIG. Also, an example of processing corresponding to step S307 in the flowchart of FIG. Hereinafter, the feature amount determination process will be described as the process after the flow up to step S306 in FIG. Here, for the sake of explanation, the data examples shown in FIGS. 9, 10, and 11 are used.

For the unknown basic data acquired in step S305, the feature amount of each basic data has been extracted in step S306. Here, one unknown basic data is selected (step S501). Here, it is assumed that data F in FIG. 10 is selected.

Next, the determination module 213 of the computer 200 selects stored basic data to be compared (step S502). Here, it is assumed that data A in FIG. 9 is selected.

The determination module 213 obtains the inner product of the feature quantity s of the data F that is unknown basic data and the feature quantity v of the data A that is stored basic data to be compared (step S503).

Further, the determination module 213 obtains the product of the absolute value of the feature value s of the data F, which is unknown basic data, and the absolute value of the feature value v of the data A, which is stored basic data to be compared (step S504).

The determination module 213 obtains a difference between the inner product obtained in step S503 and the product obtained in step S504 (step S505).

If the difference obtained in step S506 is smaller than the predetermined range, the determination module 213 determines that the data F is similar to the data A (step S506), and the difference is greater than or equal to the predetermined range. Is determined not to be similar to data A (step S507). Here, it is assumed that the data F is determined not to be similar to the data A.

Next, the determination module 213 confirms whether the determination of all stored basic data has been completed (step S508). If the determination has not ended, the process returns to step S502 to continue the process. If completed, the process proceeds to the next step S509. That is, when all the determinations as to whether the data F, which is unknown basic data, are similar to the data A, data B, data C, data D, data E, which are stored basic data, have been completed, Proceed to the next Step S509. Here, it is assumed that data F is determined to be similar to data E only.

Finally, the determination module 213 confirms whether the determination of all unknown basic data has been completed (step S509). If the determination has not ended, the process returns to step S501 to continue the process. If it has ended, the feature amount determination processing ends. That is, when it is determined that the stored basic data is similar to the data F, data G, and data H, which are unknown basic data, the process ends. Here, it is assumed that the data F is similar to only the data E, the data G is similar to only the data D, and the data H is similar to only the data A.

After the above feature amount determination processing, the process returns to step S308 in the flowchart of FIG. 3, and the classification module 214 of the computer 200 classifies which work the acquired unknown basic data is related to. In step S308, classification is performed based on the determination result that the data F is similar to the data E of the work Y, the data G is similar to the data D of the work Y, and the data H is similar to the data A of the work X. . As an example of the classification method, as described above with reference to FIG. 4, priority is given to the number of determinations of basic data, and based on the result that two of the three sensors are similar to work Y, data F, data G, data The work by H is classified as work Y. As another classification method, in an example in which classification is performed with priority on the type of data, for example, if the setting is to prioritize the classification result by moving images, for example, the determination result that the data H is similar to the data A of the work X It may be possible to classify the work using data F, data G, and data H as work X with emphasis on. Here, a case has been described in which it is determined that data F resembles only data E, data G resembles only data D, and data H resembles only data A, but unknown basic data If there are a plurality of stored basic data similar to the above, a plurality of centers may be held as similar sensors, and when the determination of all stored basic data is completed in step S508, the determination is made in step S505. Alternatively, only one sensor having the smallest difference between the inner product and the product may be held as a similar sensor. If it is determined that there is no stored basic data similar to the unknown basic data, the classification module 214 may determine that classification is not possible or that classification is not possible.

The feature quantity determination method of the flowchart of FIG. 5 is that when the feature quantity vector of unknown basic data is a and the feature quantity vector b of stored basic data is “absolute value a × absolute value b” which is the inner product of these. The “× cosine θ” uses the fact that the direction of the feature vector a and the feature vector b is considered to be more similar as θ is smaller. That is, when the angle θ approaches 0, the value of cosine θ approaches 1, so the value of the inner product of absolute value a × absolute value b × cosine θ is also absolute value a × absolute value b × 1 = absolute value a ×. It approaches the absolute value b. That is, when the difference between “absolute value a × absolute value b × cosine θ” and “absolute value a × absolute value b” is within a predetermined range, that is, a value closer to 0, the feature quantity vector a It can be said that the feature vector b is similar. However, this is merely an example of a feature amount determination method, and the determination method in step S308 is not limited to this method.

As described above, according to the present invention, a plurality of basic data relating to various works is acquired, a certain regularity and correlation of the series of works is determined as a feature quantity, and the feature quantity and the work are associated with each other. By storing it, it becomes possible to determine which of the stored sensor features is similar to the feature value of the unknown basic data.

[Feature amount judgment processing using machine learning]
FIG. 6 is a flowchart in the case of performing determination processing by machine learning of a combination of past work and feature values of basic data. As a configuration, it is assumed that the configuration is the same as that of the apparatus 100 and the computer 200 of FIG.

First, the storage module 231 of the computer 200 stores a combination of work and feature quantities of a plurality of basic data in the storage unit 230 as teacher data (step S601). A combination of work and feature quantities of a plurality of basic data may be acquired from another computer or a storage medium, or may be created by the computer 200. Alternatively, the feature quantity of the unknown basic data classified in the past and the classified work may be used as the teacher data. Further, a database dedicated to teacher data may be provided in the storage unit 230.

Next, the determination module 213 of the computer 200 performs machine learning of the determination method using the teacher data (step S602). It is assumed that supervised learning is used as a machine learning method here. Based on a large number of teacher data stored in the storage unit 230 by the storage module 231, what kind of basic data features are determined to be similar to the basic data features of the work learn. The processing in step S601 and step S602 may be skipped when machine learning of the determination method is unnecessary. Further, since the accuracy is expected to decrease when the number of teacher data is small, it is desirable to use the flow of FIG. The advantage of having the determination module 213 machine-learned by supervised learning is that it uses the features of unknown basic data that have been classified in the past and the classified work as teacher data. It is a point that can be improved. However, since it is assumed that it takes time for machine learning, the processing in step S602 may be performed in a time zone where the load of the work data classification system is low.

Since the next processing from step S603 to step S609 corresponds to the processing from step S302 to step S308 in FIG. 3, description thereof is omitted here.

As described above, according to the present invention, the combination of the feature value and the work is used as the teacher data, and the determination module 213 performs the supervised learning, so that the feature value of the unknown basic data is stored in any sensor. It is possible to improve the accuracy of determining whether or not the feature amount is similar to the feature amount, thereby improving the classification accuracy for classifying which work the unknown basic data is related to.

[Schedule comparison process]
FIG. 7 is a diagram illustrating the functional blocks of the apparatus 100 and the computer 200 and the relationship between the functions when performing schedule comparison processing. In addition to the configuration of FIG. 2, the control unit 210 implements the schedule storage module 216 in cooperation with the storage unit 230. Further, the control unit 210 realizes a schedule comparison module 217 in cooperation with the storage unit 230 and the input / output unit 240.

FIG. 8 is a flowchart for performing the schedule comparison process. Processing executed by each module described above will be described in accordance with this processing.

First, the storage module 231 of the computer 200 stores the work associated with the feature quantities of the plurality of basic data in the storage unit 230 (step S801). The basic data is a generic term that represents images, machine logs, and data of various sensors acquired from the apparatus 100. The basic data always includes information for specifying the time of work. The work associated with the feature quantities of the plurality of basic data may be acquired from another computer or a storage medium, or may be created by the computer 200. Further, a dedicated database may be provided in the storage unit 230. The process of step S801 is skipped when there is already stored an association between the work and the feature values of the plurality of basic data, or there is no association between the work to be newly stored and the feature data of the basic data. I shall do it.

Next, the schedule storage module 216 of the computer 200 stores the work schedule of the person or device in the storage unit 230 (step S802). Here, the user of the work data classification system may input a schedule via the input / output unit 240 or may receive schedule data via the communication unit 220.

The processing from step S803 to step S810 in FIG. 8 corresponds to the processing from step S302 to step S309 in FIG.

Finally, the schedule comparison module 217 of the computer 200 compares the schedule stored in step S802 with the timeline generated in step S810 (step S309). When comparing timelines, it shall be done for each person or device.

FIG. 15 is an example of the schedule comparison display of the present invention. Here, an example is shown in which a schedule stored in advance and a timeline generated based on the actual work situation shown in FIG. 12 are compared and displayed. For example, the portion from 11:30 to 12:00 of the hydraulic excavator A is highlighted, and a balloon 1501 indicates that “schedule exceeded 30 minutes”. Further, the portion from 11:00 to 12:00 of the bulldozer A is displayed with a dotted line, and a balloon 1502 indicates that “the schedule has been shortened by 60 minutes”. When the generated timeline is delayed or has a large difference with respect to the stored schedule, it is desirable to emphasize and output it so that it can be easily understood by the user. As an example for emphasizing and outputting, for example, displaying in a color that calls attention, such as red, blinking, displaying in bold, displaying large, sounding a warning sound, and the like can be given as examples.

As described above, according to the present invention, the generated timeline is compared with the schedule stored in advance, and if the generated timeline is delayed or has a large difference, it is emphasized so that the user can easily understand. Output, it becomes possible to easily manage the progress of the work.

The means and functions described above are realized by a computer (including a CPU, an information processing apparatus, and various terminals) reading and executing a predetermined program. The program may be, for example, in the form (SaaS: Software as a Service) provided from a computer via a network, or a flexible disk, CD (CD-ROM, etc.), DVD (DVD-ROM, DVD). -RAM, etc.) and a computer-readable recording medium such as a compact memory. In this case, the computer reads the program from the recording medium, transfers it to the internal storage device or the external storage device, stores it, and executes it. In addition, the program may be recorded in advance in a storage device (recording medium) such as a magnetic disk, an optical disk, or a magneto-optical disk, and provided from the storage device to a computer via a communication line.

As mentioned above, although embodiment of this invention was described, this invention is not limited to these embodiment mentioned above. The effects described in the embodiments of the present invention are only the most preferable effects resulting from the present invention, and the effects of the present invention are limited to those described in the embodiments of the present invention. is not.

100 devices, 200 computers, 300 communication networks

Claims

An acquisition means for acquiring a plurality of basic data relating to various operations performed by a person or a device;
Extraction means for extracting feature quantities of the plurality of basic data;
Storage means for storing the feature quantity and the work in association with each other;
Determining means for determining which feature quantity of the feature data of the unknown basic data is similar to the feature quantity in the feature data of the stored basic data;
Based on the result of the determination, classification means for classifying what the unknown basic data is data about what work by the person or device;
Timeline generating means for determining the content of work from the classified data and generating a timeline or a percentage of work time for each person or device based on the work time associated with the basic data;
A work data classification system comprising:
Schedule storage means for storing a schedule of work performed by the person or device;
A schedule comparison means for comparing the timeline generated by the timeline generation means with the stored schedule;
The work data classification system according to claim 1, further comprising:
The determination unit is configured to machine-learn past feature amounts and determine which feature amount in the feature amount of the basic data is similar to the feature amount of the unknown basic data. The work data classification system according to claim 1 or claim 2.
The said schedule comparison means emphasizes and outputs when the produced | generated timeline is late with respect to the said schedule, or when a difference is large. The work data classification system described in the section.
Obtaining a plurality of basic data on various tasks performed by a person or device;
Extracting features of the plurality of basic data;
Storing the feature quantity and the work in association with each other;
Determining which feature quantity of the unknown basic data feature quantity is similar to the feature quantity of the stored basic data; and
Classifying the unknown basic data based on the result of the determination as to what work by the person or device;
Determining the content of work from the classified data, and generating a timeline or percentage of work time for each person or device based on the work time associated with the basic data;
A work data classification method comprising:
Work data classification system
Obtaining a plurality of basic data relating to various operations performed by a person or device;
Extracting features of the plurality of basic data;
Storing the feature quantity and the work in association with each other;
Determining which feature quantity of the unknown basic data feature quantity is similar to the feature quantity of the stored basic data;
Classifying the unknown basic data based on the result of the determination as to what work by the person or device,
Determining the content of work from the classified data, and generating a timeline or a percentage of work time for each person or device based on the work time associated with the basic data;
A program for running