CN111684376B - Sequence data analysis device, sequence data analysis method, and computer-readable recording medium - Google Patents
Sequence data analysis device, sequence data analysis method, and computer-readable recording medium Download PDFInfo
- Publication number
- CN111684376B CN111684376B CN201880088063.3A CN201880088063A CN111684376B CN 111684376 B CN111684376 B CN 111684376B CN 201880088063 A CN201880088063 A CN 201880088063A CN 111684376 B CN111684376 B CN 111684376B
- Authority
- CN
- China
- Prior art keywords
- sequence data
- data
- input sequence
- event
- progress
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000007405 data analysis Methods 0.000 title claims abstract description 42
- 238000000034 method Methods 0.000 title claims description 50
- 238000004364 calculation method Methods 0.000 claims abstract description 86
- 230000005856 abnormality Effects 0.000 claims abstract description 61
- 239000000284 extract Substances 0.000 claims abstract description 7
- 230000008569 process Effects 0.000 claims description 32
- 230000002159 abnormal effect Effects 0.000 claims description 13
- 238000012937 correction Methods 0.000 claims description 4
- 238000001514 detection method Methods 0.000 description 45
- 238000012545 processing Methods 0.000 description 44
- 238000003780 insertion Methods 0.000 description 33
- 230000037431 insertion Effects 0.000 description 33
- 238000013500 data storage Methods 0.000 description 20
- 238000003860 storage Methods 0.000 description 15
- 238000010586 diagram Methods 0.000 description 10
- 230000006870 function Effects 0.000 description 10
- 238000012217 deletion Methods 0.000 description 9
- 230000037430 deletion Effects 0.000 description 9
- 238000012986 modification Methods 0.000 description 6
- 230000004048 modification Effects 0.000 description 6
- 238000004891 communication Methods 0.000 description 4
- 238000000605 extraction Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 239000002131 composite material Substances 0.000 description 2
- 238000003745 diagnosis Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 239000004973 liquid crystal related substance Substances 0.000 description 2
- 230000008859 change Effects 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 238000003754 machining Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000003908 quality control method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B19/00—Programme-control systems
- G05B19/02—Programme-control systems electric
- G05B19/418—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM]
- G05B19/41875—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM] characterised by quality surveillance of production
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/20—Pc systems
- G05B2219/23—Pc programming
- G05B2219/23249—Using audio and or video playback
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/30—Nc systems
- G05B2219/31—From computer integrated manufacturing till monitoring
- G05B2219/31447—Process error event detection and continuous process image detection, storage
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B23/00—Testing or monitoring of control systems or parts thereof
- G05B23/02—Electric testing or monitoring
- G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
- G05B23/0218—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterised by the fault detection method dealing with either existing or incipient faults
- G05B23/0224—Process history based detection method, e.g. whereby history implies the availability of large amounts of data
- G05B23/0227—Qualitative history assessment, whereby the type of data acted upon, e.g. waveforms, images or patterns, is not relevant, e.g. rule based assessment; if-then decisions
- G05B23/0235—Qualitative history assessment, whereby the type of data acted upon, e.g. waveforms, images or patterns, is not relevant, e.g. rule based assessment; if-then decisions based on a comparison with predetermined threshold or range, e.g. "classical methods", carried out during normal operation; threshold adaptation or choice; when or how to compare with the threshold
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Landscapes
- Engineering & Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Manufacturing & Machinery (AREA)
- Quality & Reliability (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Testing And Monitoring For Control Systems (AREA)
- Debugging And Monitoring (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
In a sequence data analysis device (10), a calculation unit (21) compares input sequence data (32) which is time-series data obtained by observing an event with reference sequence data (31) which is standard time-series data concerning the event, and extracts a combination of data elements of the input sequence data (32) and data elements of the reference sequence data (31) which correspond to each other. A calculation unit (21) calculates a relative speed of progress, which is a relative speed of progress of an event in the input sequence data (32) with respect to the progress of an event in the reference sequence data (31), for each extracted combination. A determination unit (22) determines whether or not there is an abnormality in the progress of an event in the input sequence data (32) based on the relative progress speed calculated by the calculation unit (21).
Description
Technical Field
The present invention relates to a sequence data analysis device, a sequence data analysis method, and a computer-readable recording medium.
Background
Patent document 1 describes the following method: the distance between the standard time-series data and the abnormality detection object time-series data is calculated by DTW, and whether the abnormality detection object time-series data is abnormal or normal is investigated. "DTW" is an abbreviation for Dynamic Time Warping (dynamic time warping).
Patent document 2 describes the following method: based on an extremum generated in a waveform of a time series of sensor data and a generation timing of the extremum, a normal model of the waveform is learned, and the normal model is compared with sensor data of a diagnosis object, and whether the sensor data of the diagnosis object is abnormal or normal is determined.
Patent document 3 describes the following method: a reference score is obtained from the reference data and the feature coefficients extracted from the time series of the reference data, a target score is obtained from the target data and the feature coefficients extracted from the time series of the target data, and the reference score is compared with the target score to determine whether the target data is abnormal or normal.
Prior art literature
Patent literature
Patent document 1: japanese patent application laid-open No. 2010-271997
Patent document 2: japanese patent laid-open publication No. 2017-033470
Patent document 3: japanese patent laid-open publication 2016-038856
Disclosure of Invention
Problems to be solved by the invention
In a factory production site, there is a demand for finding abnormal operations of operators such as "use of dropped components", "forget to screw down" and "different work orders" in order to perform quality control. However, the abnormal operation is currently found by visually checking the monitor image, and the load on the monitor is high. Therefore, a technique for automatically detecting an abnormal work scene from a work image by an image analysis technique is demanded.
In the work image of the factory, a series of work operation modes are repeated. Therefore, by comparing the work image as a reference with the monitor image and finding out the difference, the abnormal operation of the worker can be automatically detected. Here, if the analysis of the job image is interpreted as the analysis of the series data, the "use of the dropped component", "forgetting to screw down", and "different job sequence" which are cited as the detection target scenes may be interpreted as "insertion of series data other than the standard", "deletion of series data", and "replacement of series data", respectively. Therefore, a sequence data analysis technique capable of detecting abnormality of these sequence data in units of frames is required. However, in the related art, abnormality of the sequence data cannot be detected in units of frames.
The object of the present invention is to detect an abnormality of sequence data in units of data elements.
Means for solving the problems
The sequence data analysis device according to one embodiment of the present invention includes:
a calculation unit that compares input sequence data, which is time-series data obtained by observing an event, with reference sequence data, which is standard time-series data concerning the event, extracts combinations of data elements of the input sequence data and data elements of the reference sequence data that correspond to each other, and calculates a relative speed, which is a relative speed of progress of the event in the input sequence data with respect to the event in the reference sequence data, for each extracted combination; and
and a determination unit configured to determine whether or not there is an abnormality in progress of the event in the input sequence data, based on the relative progress speed calculated by the calculation unit.
Effects of the invention
In the present invention, the relative speed of progress of an event in input sequence data, that is, the relative speed of progress of an event in reference sequence data is calculated for each combination of data elements of input sequence data and data elements of reference sequence data, which correspond to each other. Then, it is determined whether or not there is an abnormality in the progress of the event in the input sequence data, based on the relative progress speed. Therefore, according to the present invention, an abnormality of input sequence data can be detected in units of data elements.
Drawings
Fig. 1 is a block diagram showing the configuration of a sequence data analysis device according to embodiment 1.
Fig. 2 is a flowchart showing the operation of the sequence data analysis device according to embodiment 1.
Fig. 3 is a flowchart showing the procedure of the regular path calculation process according to embodiment 1.
Fig. 4 is a diagram showing an example of the regular route calculated by the regular route calculation process according to embodiment 1.
Fig. 5 is a flowchart showing the procedure of the relative speed of progress calculation process according to embodiment 1.
Fig. 6 is a diagram showing an example of the slope of the regular route calculated by the relative speed of progress calculation process according to embodiment 1.
Fig. 7 is a diagram showing an example of the relative progress speed calculated by the relative progress speed calculation process according to embodiment 1.
Fig. 8 is a flowchart showing the sequence data missing section detection processing according to embodiment 1.
Fig. 9 is a diagram showing an example of a sequence data loss section detected by the sequence data loss section detection process according to embodiment 1.
Fig. 10 is a flowchart showing the procedure of the outside-standard sequence data insertion section detection process according to embodiment 1.
Fig. 11 is a diagram showing an example of the extra-standard sequence data insertion section detected by the extra-standard sequence data insertion section detection process according to embodiment 1.
Fig. 12 is a flowchart showing the sequence data exchange section detection processing procedure according to embodiment 1.
Fig. 13 is a diagram showing an example of a sequence data replacement section detected by the sequence data replacement section detection process according to embodiment 1.
Fig. 14 is a diagram showing an example of abnormality information recorded by the abnormality information recording process according to embodiment 1.
Fig. 15 is a diagram showing an example of abnormality information recorded by the abnormality information recording process according to embodiment 1.
Fig. 16 is a block diagram showing the configuration of a sequence data analysis device according to a modification of embodiment 1.
Detailed Description
Hereinafter, embodiments of the present invention will be described with reference to the drawings. In the drawings, the same or corresponding portions are denoted by the same reference numerals. In the description of the embodiments, the description is omitted or simplified as appropriate for the same or corresponding portions. The present invention is not limited to the embodiments described below, and various modifications may be made as necessary. For example, the embodiments described below may be partially implemented.
Embodiment 1
The present embodiment will be described with reference to fig. 1 to 15.
* Description of the structure
Referring to fig. 1, the configuration of a sequence data analysis device 10 according to the present embodiment will be described.
The sequence data analysis device 10 is a computer. The sequence data analysis device 10 includes a processor 11, and further includes other hardware such as a memory 12, a reference sequence data storage device 13, an input sequence data storage device 14, and an abnormality information storage device 15. The processor 11 is connected to other hardware via signal lines, and controls these other hardware.
The sequence data analysis device 10 includes a calculation unit 21 and a determination unit 22 as functional elements. The calculation unit 21 includes a regular path calculation unit 23 and a relative progress speed calculation unit 24. The determination unit 22 includes a data loss detection unit 25, an off-standard data insertion detection unit 26, a data replacement estimation unit 27, and an abnormality information recording unit 28. The functions of the calculation unit 21 and the determination unit 22 are realized by software.
The processor 11 is a device that executes a sequence data analysis program. The sequence data analysis program is a program that realizes the functions of the calculation unit 21 and the determination unit 22. The processor 11 is, for example, CPU, GPU, DSP or a combination of some or all of them. "CPU" is an abbreviation of Central Processing Unit (central processing unit). "GPU" is an abbreviation for Graphics Processing Unit (graphics processing unit). "DSP" is an abbreviation for Digital Signal Processor (digital signal processor).
The memory 12 is a device for storing a sequence data analysis program, regular path information 33 generated by the sequence data analysis program, relative progress speed information 34, sequence data deletion section information 35, sequence data insertion section information 36 outside the standard, sequence data replacement section information 37, and other information. The memory 12 is, for example, RAM, flash memory or a combination thereof. "RAM" is an abbreviation for Random Access Memory (random access memory).
The reference sequence data storage device 13 is a device that stores reference sequence data 31 input to a sequence data analysis program. The reference sequence data storage 13 is, for example, an HDD, a flash memory, or a combination thereof. "HDD" is an abbreviation for Hard Disk Drive.
The input sequence data storage device 14 is a device that stores input sequence data 32 input to a sequence data analysis program. The input sequence data storage 14 is, for example, a HDD, a flash memory, or a combination thereof.
The abnormality information storage device 15 is a device that stores abnormality information 38 output from the sequence data analysis program. The abnormality information storage device 15 is, for example, an HDD, a flash memory, or a combination thereof.
The reference sequence data storage device 13, the input sequence data storage device 14, and the abnormality information storage device 15 may be provided outside the sequence data analysis device 10. The reference sequence data storage device 13, the input sequence data storage device 14, and the abnormality information storage device 15 may be implemented as separate hardware, or may be implemented as integrated hardware.
As hardware, the sequence data analysis apparatus 10 may include a communication device, an input device, and a display.
The communication device includes a receiver that receives the reference sequence data 31 input to the sequence data analysis program, the input sequence data 32, and other data, and a transmitter that transmits the abnormality information 38 and other information output from the sequence data analysis program. The communication device is, for example, a communication chip or NIC. "NIC" is an abbreviation for Network Interface Card (network interface card).
The input device is a device operated by a user to input parameters such as a threshold value to the sequence data analysis program. The input device is, for example, a mouse, a keyboard, a touch panel, or a combination of some or all of them.
The display is a device that displays the abnormality information 38 and other information output from the sequence data analysis program on a screen. The display is for example an LCD. "LCD" is an abbreviation for Liquid Crystal Display (liquid crystal display).
The sequence data analysis program is read from the memory 12 into the processor 11, and executed by the processor 11. The memory 12 stores not only the sequence data analysis program but also the OS. "OS" is an abbreviation for Operating System. The processor 11 executes the sequence data analysis program while executing the OS. In addition, a part or the whole of the sequence data analysis program may be incorporated into the OS.
The sequence data analysis program and the OS may be stored in the auxiliary storage device. The secondary storage device is, for example, a HDD, flash memory, or a combination thereof. The sequence data analysis program and the OS are loaded into the memory 12 and executed by the processor 11 when stored in the auxiliary storage device. The auxiliary storage device may also be used as at least any one of the reference sequence data storage device 13, the input sequence data storage device 14, and the abnormality information storage device 15.
The sequence data analysis device 10 may include a plurality of processors instead of the processor 11. The plurality of processors share execution of the sequence data analysis program. The individual processors are, for example, CPU, GPU, DSP or a combination of some or all of them.
The signal values and variable values utilized, processed or output by the sequence data analysis program are stored in a memory 12, a secondary storage device or a register or cache within the processor 11.
The sequence data analysis program is a program for causing a computer to execute the processing performed by the calculation section 21 and the determination section 22 as calculation processing and determination processing, respectively. The sequence data analysis program may be provided by being recorded in a computer-readable medium, may be provided by being stored in a recording medium, or may be provided as a program product.
The sequence data analysis device 10 may be constituted by one computer or by a plurality of computers. In the case where the sequence data analysis device 10 is configured by a plurality of computers, the functions of the calculation unit 21 and the determination unit 22 may be realized by being distributed among the computers.
* Description of the actions
The operation of the sequence data analysis device 10 according to the present embodiment will be described with reference to fig. 2 to 15. The operation of the sequence data analysis device 10 corresponds to the sequence data analysis method of the present embodiment.
Fig. 2 shows the sequence of the sequence data analysis processing performed by the sequence data analysis device 10.
The sequence data analysis processing is roughly divided into 4 steps, namely, a sequence data reading processing in step S100, a regular path calculation processing in step S200, an abnormality extraction processing in step S300, and an abnormality information writing processing in step S400.
In step S100, the calculation unit 21 acquires the input sequence data 32 and the reference sequence data 31. The input sequence data 32 is time-series data obtained by observing a certain event X. The event X may be any event, but in the present embodiment, it is an event that can be captured and recorded in a video, specifically, a job in a factory. That is, the input sequence data 32 is data of an image in the present embodiment, specifically, data of an image of an operation performed in a factory. The reference sequence data 31 is standard time-series data about the event X. The reference sequence data 31 may not be the same form of data as the input sequence data 32, but in the present embodiment, is the same form of data as the input sequence data 32, that is, video data, specifically, data in which a video of a work that is performed correctly is captured.
In the sequence data reading processing in step S100, specifically, the reference sequence data reading processing in step S110 and the input sequence data reading processing in step S120 are performed.
In the reference sequence data reading processing in step S110, the processing of writing the reference sequence data 31 read from the reference sequence data storage device 13 into the memory 12 is performed by the calculation unit 21. In the input sequence data reading processing in step S120, the regular path calculation unit 23 of the calculation unit 21 performs processing of writing the input sequence data 32 read from the input sequence data storage device 14 into the memory 12.
In step S200, the calculation unit 21 compares the input sequence data 32 with the reference sequence data 31, and extracts a combination of data elements of the input sequence data 32 and data elements of the reference sequence data 31 that correspond to each other. As described above, the input sequence data 32 is data of an image, and each frame of the image is handled as a data element of the input sequence data 32. The reference sequence data 31 is also data of an image, and each frame of the image is handled as a data element of the reference sequence data 31.
In the regular path calculation process of step S200, specifically, the similarity map calculation process of step S210, the score map calculation process of step S220, and the regular path calculation process of step S230 are performed.
The similarity map calculation process of step S210, the score map calculation process of step S220, and the regular path calculation process of step S230 will be described later.
In step S300, the calculation unit 21 calculates the relative progress speed V for each combination extracted in step S200. The relative speed of progress V is the relative speed of progress of the event X in the input sequence data 32 relative to the progress of the event X in the reference sequence data 31. The determination unit 22 determines whether or not there is an abnormality in the progress of the event X in the input sequence data 32 based on the relative progress speed V calculated by the calculation unit 21.
In the abnormality extraction processing of step S300, specifically, the relative progress speed calculation processing of step S310, the sequence data missing section detection processing of step S320, the sequence data insertion section detection processing other than the standard of step S330, and the sequence data replacement section estimation processing of step S340 are performed.
The relative progress speed calculation process of step S310, the sequence data missing section detection process of step S320, the sequence data insertion section detection process other than the standard of step S330, and the sequence data replacement section estimation process of step S340 will be described later.
In the abnormality information writing process in step S400, the abnormality information recording unit 28 of the determination unit 22 performs a process of writing the abnormality information 38 extracted in the abnormality extraction process in step S300 into the abnormality information storage device 15.
Fig. 3 shows the sequence of the regular path calculation processing of step S200.
In step S200, the regular path calculation unit 23 of the calculation unit 21 calculates the regular path information 33 indicating the corresponding frame between the reference sequence data 31 and the input sequence data 32 read from the reference sequence data storage device 13 and the input sequence data storage device 14, using DP matching. "DP" is an abbreviation for Dynamic Programming (dynamic programming).
As described above, in the regular path calculation process of step S200, specifically, the similarity map calculation process of step S210, the score map calculation process of step S220, and the regular path calculation process of step S230 are performed.
In step S210, the regular path calculation unit 23 calculates the similarities SIM [ i, j ] between all frames with respect to the input sequence data 32 and the reference sequence data 31, and writes the calculated similarities SIM [ i, j ] in the memory 12. The similarity SIM [ i, j ] is calculated by the following formula.
SIM[i，j]＝A[i]·B[j]/(||A[i]||||B[j]||)
J= {0, (ii) the method comprises the steps of (1), j= {0, ··, P-1}, Q is the length of the input sequence data, and P is the length of the reference sequence data. A and B are input sequence data 32 and reference sequence data 31, respectively, and are expressed by the following formulas.
A＝{A[0]，A[1]，···，A[i]，···，A[Q-1]}
B＝{B[0]，B[1]，···，B[j]，···，B[P-1]}
Here, A [ i ] = { A [ i,0], A [ i,1],. Cndot.A [ i, D-1] }, B [ j ] = { B [ j,0], B [ j,1],. Cndot.B [ j, D-1] }. D is the data dimension.
In step S220, the regular route calculation unit 23 initializes the score map DTW by the following expression.
DTW[＊，0]＝SIM[＊，0]
DTW[0，＊]＝SIM[0，＊]
Then, the regular path calculation unit 23 calculates the score map DTW by the following expression based on the similarity map SIM.
DTW[k，h]＝SIM[k-1，h]+max{DTW[k-1，h]，DTW[k，h-1]，DTW[k-1，h-1]}
Here, k= {1, Q-1, h= {1, ··, P-1}.
In step S230, the regular route calculation unit 23 searches the score map DTW for a regular route. The regular path calculation unit 23 searches for a regular path by repeatedly searching for an adjacent cell selected by the max function of the equation used for calculating the cell value, starting from the cell of DTW [ Q-1, p-1 ].
As described above, in the present embodiment, the calculation unit 21 calculates the regular path between the input sequence data 32 and the reference sequence data 31. Then, the calculation unit 21 extracts the elements of the calculated regular path as a combination of the data elements of the input sequence data 32 and the data elements of the reference sequence data 31.
An example of a regular path is shown in fig. 4. In this example, as a combination of the start frame of the input sequence data 32 and the start frame of the reference sequence data 31, the start element of the regular path is extracted. As a combination of the 2 nd frame of the input sequence data 32 and the start frame of the reference sequence data 31, the 2 nd element of the regular path is extracted. As a combination of the 3 rd frame of the input sequence data 32 and the 2 nd frame of the reference sequence data 31, the 3 rd element of the regular path is extracted. As a combination of the 4 th frame of the input sequence data 32 and the 2 nd frame of the reference sequence data 31, the 4 th element of the regular path is extracted. As a combination of an arbitrary frame of the input sequence data 32 and an arbitrary frame of the reference sequence data 31, elements from element 5 of the regular path are extracted.
Fig. 5 shows the sequence of the relative-progress-speed calculation process of step S310.
In step S310, the relative progress speed calculation unit 24 of the calculation unit 21 receives the regular route information 33 from the regular route calculation unit 23, investigates the relative progress speed between the reference sequence data 31 and the input sequence data 32 in units of frames, and outputs the relative progress speed as the relative progress speed information 34.
In the relative progress speed calculation processing of step S310, specifically, processing of step S311 of calculating the slope θ of the regular path and processing of step S312 of calculating the relative progress speed V from the slope θ of the regular path are performed.
In step S311, the relative progress speed calculation unit 24 calculates the local slope θ for the regular path calculated in step S230 by the following equation.
d is less than or equal to M and less than M-1-d:
θ[m]＝tan -1 (q[m+d]-q[m-d]，p[m+d]-p[m-d])
m < d:
θ[m]＝tan -1 (q[m+d]-q[0]，p[m+d]-p[0])
when M is more than or equal to M-1-d:
θ[m]＝tan -1 (q[M-1]-q[m-d]，p[M-1]-p[m-d])
here, m= {0, and M-1, M is the length of the regular path. p [ m ] and q [ m ] denote the frame number of the reference sequence data 31 and the frame number of the input sequence data 32 corresponding to the mth cell on the regular path, respectively.
In step S312, the relative progress speed calculation unit 24 calculates the relative progress speed V from the slope θ of the regular path using the following equation.
V[m]＝(θ[m]-π/4)/(π/4)
In this way, in the present embodiment, the calculation unit 21 calculates the relative progress velocity V for each combination of the data elements of the extracted input sequence data 32 and the data elements of the reference sequence data 31.
Fig. 6 shows an example of the slope θ of the regular path. In this example, d=1, but d may be set to an integer greater than 1.
Fig. 7 shows an example of the relative progress velocity V. In this example, the relative progress speed V is calculated according to the examples of fig. 4 and 6. Thus, for the starting element of the regular path, θ [ 0] is calculated]＝tan -1 (q[1]-q[0]，p[1]-p[0])＝tan -1 (1，0)，V[0]＝(θ[0]-π/4)/(π/4)＝(tan -1 (1, 0) -pi/4)/(pi/4) = -1.0. Regarding the 2 nd element of the regular path, θ [ 1] is calculated]＝tan -1 (q[2]-q[0]，p[2]-p[0])＝tan -1 (1，2)，V[1]＝(θ[1]-π/4)/(π/4)＝(tan -1 (1, 2) -pi/4)/(pi/4) = -0.4. Regarding the 3 rd element of the regular path, the θ 2 is calculated]＝tan -1 (q[3]-q[1]，p[3]-p[1])＝tan -1 (1，2)，V[2]＝(θ[2]-π/4)/(π/4)＝(tan -1 (12) -pi/4)/(pi/4) = -0.4. For element 4 of the regular path, θ [3 ] is calculated]＝tan -1 (q[4]-q[2]，p[4]-p[2])＝tan -1 (1，1)，V[3]＝(θ[3]-π/4)/(π/4)＝(tan -1 (1, 1) -pi/4)/(pi/4) =0. For the elements from element 5 of the regular path, V [ m ] is also calculated]。
Fig. 8 shows the sequence of the sequence data missing section detection processing in step S320.
In step S320, the data loss detection unit 25 of the determination unit 22 checks whether or not there is no loss of sequence data as in the reference sequence data 31 with respect to the input sequence data 32, based on the relative progress speed information 34 received from the relative progress speed calculation unit 24. When there is a section in which sequence data is missing, the data missing detection unit 25 outputs sequence data missing section information 35. That is, the data loss detection unit 25 detects a sequence data loss section when compared with the reference sequence data 31 with respect to the input sequence data 32 based on the relative progress speed V calculated in step S310.
In step S321, the data loss detection unit 25 initializes a "start frame number" and a "counter" storing the start frame and the section length of the detection target section with-1 and 0, respectively.
Then, the data loss detection unit 25 sequentially performs the following processing from the start unit to the end unit using the index r indicating the unit constituting the regular path.
When the relative progress speed vr is 1.0, the data loss detection unit 25 increments the counter by 1 in step S322. At this time, if the initialization code-1 is set for the start frame number, the data loss detection unit 25 substitutes the index r for the start frame number in step S323. When the relative progress speed vr is not 1.0, the data loss detection unit 25 checks the counter and the threshold T1. If the value of the counter is greater than the threshold T1, the data loss detection unit 25 detects the section [ start frame number, r-1] as a sequence data loss section in step S324. Next, in step S325, the data loss detection unit 25 initializes a counter and a start frame number, respectively. In addition, the threshold T1 may be appropriately adjusted.
As described above, in the present embodiment, the determination unit 22 determines whether or not there is an abnormality in the progress of the event X in the input sequence data 32 based on the relative progress speed V and whether or not consecutive data elements of the reference sequence data 31 correspond to the same data elements of the input sequence data 32 among the 2 or more combinations extracted by the calculation unit 21.
The determination unit 22 determines whether or not there is an abnormality such that the data of the section R2 corresponding to the section R1 is missing from the input sequence data 32, based on the relative speed of the progress of the event X in the data of the section R1 included in the reference sequence data 31.
Fig. 9 shows an example of a sequence data deletion section. In the present embodiment, a section R1 having a length exceeding the length determined by the threshold T1 and continuing to be 1 with respect to the speed V of progress is detected as a sequence data missing section. This makes it possible to automatically detect an abnormal operation of the operator who forgets to tighten the screw.
Fig. 10 shows the procedure of the outside-standard sequence data insertion section detection processing in step S330.
In step S330, the data insertion detection unit 26 outside the standard of the determination unit 22 checks whether or not the sequence data that is not present in the reference sequence data 31 is not inserted into the input sequence data 32, based on the relative progress speed information 34 received from the relative progress speed calculation unit 24. The extra-standard data insertion detection unit 26 outputs extra-standard sequence data insertion section information 36 when there is a sequence data section not included in the reference sequence data 31. That is, the outside-standard data insertion detection unit 26 detects the outside-standard sequence data insertion section when compared with the reference sequence data 31 with respect to the input sequence data 32 based on the relative progress speed V calculated in step S310.
In step S331, the off-standard data insertion detection unit 26 initializes a "start frame number" and a "counter" storing the start frame and the section length of the detection target section with-1 and 0, respectively.
Then, the off-standard data insertion detection unit 26 sequentially performs the following processing from the start unit to the end unit using the index r indicating the unit constituting the regular path.
When the relative progress speed vr is-1.0, the off-standard data insertion detection unit 26 increments the counter by 1 in step S332. At this time, if the initialization code-1 is set for the start frame number, the off-standard data insertion detection unit 26 substitutes the index r for the start frame number in step S333. When the relative progress speed vr is not-1.0, the off-standard data insertion detection unit 26 checks the counter and the threshold T2. If the value of the counter is greater than the threshold T2, in step S334, the out-of-standard data insertion detection unit 26 detects the section [ start frame number, r-1] as the out-of-standard sequence data insertion section. Next, in step S335, the off-standard data insertion detection unit 26 initializes a counter and a start frame number, respectively. In addition, the threshold T2 may be appropriately adjusted.
As described above, in the present embodiment, the determination unit 22 determines whether or not there is an abnormality in the progress of the event X in the input sequence data 32 based on the relative progress speed V and whether or not consecutive data elements of the input sequence data 32 correspond to the same data elements of the reference sequence data 31 in the combinations of 2 or more extracted by the calculation unit 21.
The determination unit 22 determines whether or not there is an abnormality in which the data of a certain section R2 is data other than the standard, based on the relative speed of progress of the event X in the data of the section R2 included in the input sequence data 32.
Fig. 11 shows an example of an insertion section of sequence data outside the standard. In the present embodiment, the section R2, in which the relative progress speed V continues to be-1 and the length exceeds the length determined by the threshold T2, is detected as the non-standard sequence data insertion section. This makes it possible to automatically detect an abnormal operation of the operator such as "use the dropped component".
Fig. 12 shows the sequence of the sequential data exchange section estimation processing in step S340.
In step S340, the determination unit 22 data exchange estimation unit 27 checks whether or not the sequence data is in the order of the reference sequence data 31 with respect to the input sequence data 32, based on the sequence data deletion section information 35 and the sequence data insertion section information 36 received from the data deletion detection unit 25 and the data insertion detection unit 26. If the sequence data exchange is found, the data exchange estimation unit 27 outputs the sequence data exchange section information 37. That is, the data exchange estimating unit 27 checks whether or not the sequence data exchange with respect to the reference sequence data 31 has not occurred with respect to the input sequence data 32 based on the sequence data missing section information 35 and the sequence data insertion section information 36 other than the standard.
First, the data exchange estimating unit 27 checks whether or not the detected sequence data missing section and the sequence data insertion section other than the standard are one each. When the number of at least one section is not 1, the data exchange estimating unit 27 determines that there is no sequence data exchange. When the number of 2 sections is 1, the data exchange estimating unit 27 checks whether or not the difference between the length L1 of the section of the reference sequence data 31 corresponding to the sequence data missing section and the length L2 of the section of the input sequence data 32 corresponding to the sequence data insertion section other than the standard is smaller than the threshold T3. When the difference between the 2 sections is smaller than the threshold T3, the data exchange estimating unit 27 cuts out the sequence data of the sequence data section other than the standard and inserts the sequence data into the sequence data missing section generating section in step S341, thereby processing the input sequence data 32. The similarity map calculation processing in step S210, the score map calculation processing in step S220, and the regular path calculation processing in step S230 are performed on the processed input sequence data in the same manner as the original input sequence data 32. In step S342, the data exchange estimating unit 27 examines the score value of the regular route, and compares the score value with the score value of the regular route of the input sequence data 32 before processing. The score value refers to the value of the score map at the regular path termination unit location. When the score value is greater than the value before machining, the data exchange estimating unit 27 detects the 2 sections as the sequence data exchange sections in step S343. In addition, the threshold T3 may be appropriately adjusted.
As described above, in the present embodiment, the determination unit 22 may determine that there is an abnormality such that the data of the section R2a corresponding to the 1 st section R1a included in the reference sequence data 31 is missing from the input sequence data 32, or an abnormality such that the data of the 2 nd section R2b included in the input sequence data 32 is data other than the standard. In this case, the determination unit 22 deletes the data of the 2 nd section R2b from the input sequence data 32, and inserts the deleted data into the input sequence data 32 as the data of the section R2a corresponding to the 1 st section R1a, thereby correcting the input sequence data 32.
The calculation unit 21 calculates 1 st similarity, which is the similarity between the input sequence data 32 before correction by the determination unit 22 and the reference sequence data 31, and 2 nd similarity, which is the similarity between the input sequence data after correction by the determination unit 22 and the reference sequence data 31. The determination unit 22 compares the 1 st similarity with the 2 nd similarity. When the 2 nd similarity is higher than the 1 st similarity, the determination unit 22 determines that the event X in the input sequence data 32 has an abnormality such as sequential replacement.
Fig. 13 shows an example of a sequence data exchange section. As shown in this example, by detecting the sequence data exchange section, it is possible to automatically detect an abnormal operation of the operator such as "the work order is different".
In step 400 shown in fig. 2, the abnormality information recording unit 28 of the determination unit 22 generates the abnormality information 38 based on the sequence data deletion section information 35, the sequence data insertion section information outside the standard 36, and the sequence data replacement section information 37 received from the data deletion detection unit 25, the data insertion detection unit outside the standard 26, and the data replacement estimation unit 27, respectively. The abnormality information recording unit 28 writes the abnormality information 38 to the abnormality information recording device 109. Examples of output to the abnormality information storage device 109 are shown in fig. 14 and 15.
* Description of effects of the embodiments
In the present embodiment, the relative speed of progress of the event X in the input sequence data 32 with respect to the relative speed of progress of the event X in the reference sequence data 31, that is, the relative speed of progress is calculated for each combination of the data elements of the input sequence data 32 and the data elements of the reference sequence data 31 that correspond to each other. Then, it is determined whether or not there is an abnormality in the progress of the event X in the input sequence data 32, based on the relative progress speed. Therefore, according to the present embodiment, an abnormality of the input sequence data 32 can be detected in units of data elements.
According to the present embodiment, it is possible to detect "insertion of sequence data other than standard", "deletion of sequence data", and "replacement of sequence data" on a frame-by-frame basis from the input sequence data 32.
In the present embodiment, the relative frame progress rate between the reference sequence data 31 and the input sequence data 32 is calculated from the frame correspondence information between the input sequence data 32 and the reference sequence data 31 obtained by DTW. Specifically, the frame correspondence information is a regular path. The "missing sequence data" section and the "inserting out of standard sequence data" section are detected as abnormal sections in units of frames. Then, the order of the sequence data is changed for 2 abnormal sections. The change in similarity with the reference sequence data 31 was examined for the input sequence data 32 before replacement and the input sequence data after replacement. When the similarity increases due to replacement, it is determined that the input sequence data 32 has been replaced. By such a processing procedure, an abnormality of the input sequence data 32 can be detected with high detail.
* Other structures
In the present embodiment, the functions of the calculation unit 21 and the determination unit 22 are realized by software, but as a modification, the functions of the calculation unit 21 and the determination unit 22 may be realized by hardware. The differences from the present embodiment will be mainly described with respect to this modification.
The configuration of the sequence data analysis device 10 according to the modification of the present embodiment will be described with reference to fig. 16.
The sequence data analysis device 10 includes hardware such as an electronic circuit 16, a reference sequence data storage device 13, an input sequence data storage device 14, and an abnormality information storage device 15.
The electronic circuit 16 is dedicated hardware for realizing the functions of the calculation unit 21 and the determination unit 22. The electronic circuit 16 is, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, logic IC, GA, FPGA, ASIC, or a combination of some or all of them. "IC" is an abbreviation for Integrated Circuit (integrated circuit). "GA" is an abbreviation for Gate Array. "FPGA" is an abbreviation for Field-Programmable Gate Array (Field programmable gate array). "ASIC" is an abbreviation for Application Specific Integrated Circuit (application specific integrated circuit).
The sequence data analysis device 10 may include a plurality of electronic circuits instead of the electronic circuit 16. These plural electronic circuits as a whole realize the functions of the calculation section 21 and the determination section 22. The individual electronic circuits may be, for example, single circuits, composite circuits, programmed processors, parallel programmed processors, logic IC, GA, FPGA, ASIC, or some or all combinations thereof.
As another modification, the functions of the calculation unit 21 and the determination unit 22 may be realized by a combination of software and hardware. That is, part of the functions of the calculation unit 21 and the determination unit 22 may be realized by dedicated hardware, and the rest may be realized by software.
Both the processor 11 and the electronic circuit 16 are processing circuits. That is, regardless of the configuration of the sequence data analysis device 10, the operations of the calculation unit 21 and the determination unit 22 are performed by the processing circuit, whether the configuration is the configuration shown in fig. 1 or the configuration shown in fig. 16.
Description of the reference numerals
10: a sequence data analysis device; 11: a processor; 12: a memory; 13: a reference sequence data storage device; 14: an input sequence data storage device; 15: an abnormality information storage device; 16: an electronic circuit; 21: a calculation unit; 22: a determination unit; 23: a regular path calculation unit; 24: a relative progress speed calculation unit; 25: a data loss detection unit; 26: a data insertion detection unit for inserting data outside the standard; 27: a data exchange estimation unit; 28: an abnormality information recording unit; 31: reference sequence data; 32: inputting sequence data; 33: regular path information; 34: relative progress speed information; 35: sequence data deletion interval information; 36: inserting section information into sequence data outside the standard; 37: sequence data exchange interval information; 38: abnormality information.
Claims (11)
1. A sequence data analysis device is provided with:
a calculation unit that compares input sequence data, which is time-series data obtained by observing an event, with reference sequence data, which is standard time-series data concerning the event, extracts combinations of data elements of the input sequence data and data elements of the reference sequence data that correspond to each other, and calculates a relative speed, which is a relative speed of progress of the event in the input sequence data with respect to the event in the reference sequence data, for each extracted combination; and
and a determination unit configured to determine whether or not there is an abnormality in progress of the event in the input sequence data, based on the relative progress speed calculated by the calculation unit.
2. The sequence data analysis device according to claim 1, wherein,
the calculation unit calculates a regular path between the input sequence data and the reference sequence data, and extracts an element of the calculated regular path as a combination of a data element of the input sequence data and a data element of the reference sequence data.
3. The sequence data analysis device according to claim 1 or 2, wherein,
the input sequence data is data of an image, and each frame of the image is treated as a data element of the input sequence data.
4. The sequence data analysis device according to claim 1 or 2, wherein,
the determination unit determines whether or not the abnormality exists based on the relative progress speed and whether or not consecutive data elements of the reference sequence data correspond to the same data element of the input sequence data among the combinations of 2 or more extracted by the calculation unit.
5. The sequence data analysis device according to claim 1 or 2, wherein,
the determination unit determines whether or not there is an abnormality such as a loss of data of a section corresponding to a certain section included in the reference sequence data from the input sequence data, based on a relative speed of progress of the event in the data.
6. The sequence data analysis device according to claim 1 or 2, wherein,
the determination unit determines whether or not the abnormality exists based on the relative progress speed and whether or not consecutive data elements of the input sequence data correspond to the same data elements of the reference sequence data among the 2 or more combinations extracted by the calculation unit.
7. The sequence data analysis device according to claim 1 or 2, wherein,
the determination unit determines whether or not there is an abnormality in which the data of a certain section included in the input sequence data is data other than the standard, based on the relative speed of progress of the event in the data.
8. The sequence data analysis device according to claim 1 or 2, wherein,
the determination unit corrects the input sequence data by deleting data of a certain 2 nd section from the input sequence data and inserting the deleted data into the input sequence data as data of a section corresponding to the 1 st section when it is determined that there is an abnormality such that the data of the section corresponding to the 1 st section included in the reference sequence data is missing from the input sequence data and an abnormality such that the data of the 2 nd section included in the input sequence data is not standard.
9. The sequence data analysis device according to claim 8, wherein,
the calculation unit calculates 1 st similarity, which is a similarity between the input sequence data before correction by the determination unit and the reference sequence data, and 2 nd similarity, which is a similarity between the input sequence data after correction by the determination unit and the reference sequence data,
the determination unit compares the 1 st similarity with the 2 nd similarity, and determines that the event in the input sequence data is abnormal such that the event is sequentially replaced when the 2 nd similarity is higher than the 1 st similarity.
10. A sequence data analysis method, wherein,
the calculation unit compares input sequence data, which is time-series data obtained by observing an event, with reference sequence data, which is standard time-series data concerning the event, extracts combinations of data elements of the input sequence data and data elements of the reference sequence data that correspond to each other, calculates a relative speed, which is a relative speed of progress of the event in the input sequence data with respect to the event in the reference sequence data, for each extracted combination,
a determination unit determines whether or not there is an abnormality in the progress of the event in the input sequence data, based on the relative progress speed calculated by the calculation unit.
11. A computer-readable recording medium having recorded thereon a sequence data analysis program that causes a computer to execute:
a calculation process of comparing input sequence data, which is time-series data obtained by observing an event, with reference sequence data, which is standard time-series data concerning the event, extracting combinations of data elements of the input sequence data and data elements of the reference sequence data corresponding to each other, and calculating a relative speed, which is a relative speed of progress of the event in the input sequence data with respect to the event in the reference sequence data, for each extracted combination; and
and a determination process of determining whether or not there is an abnormality in progress of the event in the input sequence data, based on the relative progress speed calculated by the calculation process.
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| PCT/JP2018/004097 WO2019155533A1 (en) | 2018-02-06 | 2018-02-06 | Sequential data analysis apparatus, sequential data analysis method, and sequential data analysis program |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111684376A CN111684376A (en) | 2020-09-18 |
| CN111684376B true CN111684376B (en) | 2023-08-18 |
Family
ID=67548828
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201880088063.3A Active CN111684376B (en) | 2018-02-06 | 2018-02-06 | Sequence data analysis device, sequence data analysis method, and computer-readable recording medium |
Country Status (4)
| Country | Link |
|---|---|
| JP (1) | JP6704543B2 (en) |
| CN (1) | CN111684376B (en) |
| DE (1) | DE112018006760T5 (en) |
| WO (1) | WO2019155533A1 (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH11225326A (en) * | 1998-02-05 | 1999-08-17 | Toshiba Corp | Image supervisory device |
| JP2010271997A (en) * | 2009-05-22 | 2010-12-02 | Yamatake Corp | Standard time series data calculation method, abnormality detection method, standard time series data calculation device, abnormality detection device, standard time series data calculation program, and abnormality detection program |
| JP2011203787A (en) * | 2010-03-24 | 2011-10-13 | Yamatake Corp | Data-monitoring method and data-monitoring apparatus |
| JP2012003608A (en) * | 2010-06-18 | 2012-01-05 | Yazaki Corp | Drive recorder for vehicle, recorded information analysis method, recorded information analysis program, and recorded information analyzer |
| CN103473791A (en) * | 2013-09-10 | 2013-12-25 | 惠州学院 | Method for automatically recognizing abnormal velocity event in surveillance video |
| CN107133343A (en) * | 2017-05-19 | 2017-09-05 | 哈工大大数据产业有限公司 | Big data abnormal state detection method and device based on time series approximate match |
| JP2017157072A (en) * | 2016-03-03 | 2017-09-07 | 株式会社日立製作所 | Abnormality detection apparatus, system stability monitoring device, and system thereof |
| CN107449958A (en) * | 2016-04-18 | 2017-12-08 | 丰田自动车株式会社 | Abnormity determining device and abnormality determination method |
| CN107615773A (en) * | 2015-05-27 | 2018-01-19 | 三菱电机株式会社 | Reception device, image display, instantaneous speech power and method of reseptance |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100208063A1 (en) * | 2009-02-19 | 2010-08-19 | Panasonic Corporation | System and methods for improving accuracy and robustness of abnormal behavior detection |
| US20170103672A1 (en) * | 2015-10-09 | 2017-04-13 | The Regents Of The University Of California | System and method for gesture capture and real-time cloud based avatar training |
| JP6451662B2 (en) * | 2016-02-23 | 2019-01-16 | 株式会社安川電機 | Abnormality determination device, abnormality determination program, abnormality determination system, and motor control device |
-
2018
- 2018-02-06 CN CN201880088063.3A patent/CN111684376B/en active Active
- 2018-02-06 JP JP2019570183A patent/JP6704543B2/en active Active
- 2018-02-06 DE DE112018006760.2T patent/DE112018006760T5/en active Pending
- 2018-02-06 WO PCT/JP2018/004097 patent/WO2019155533A1/en active Application Filing
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH11225326A (en) * | 1998-02-05 | 1999-08-17 | Toshiba Corp | Image supervisory device |
| JP2010271997A (en) * | 2009-05-22 | 2010-12-02 | Yamatake Corp | Standard time series data calculation method, abnormality detection method, standard time series data calculation device, abnormality detection device, standard time series data calculation program, and abnormality detection program |
| JP2011203787A (en) * | 2010-03-24 | 2011-10-13 | Yamatake Corp | Data-monitoring method and data-monitoring apparatus |
| JP2012003608A (en) * | 2010-06-18 | 2012-01-05 | Yazaki Corp | Drive recorder for vehicle, recorded information analysis method, recorded information analysis program, and recorded information analyzer |
| CN103473791A (en) * | 2013-09-10 | 2013-12-25 | 惠州学院 | Method for automatically recognizing abnormal velocity event in surveillance video |
| CN107615773A (en) * | 2015-05-27 | 2018-01-19 | 三菱电机株式会社 | Reception device, image display, instantaneous speech power and method of reseptance |
| JP2017157072A (en) * | 2016-03-03 | 2017-09-07 | 株式会社日立製作所 | Abnormality detection apparatus, system stability monitoring device, and system thereof |
| CN107449958A (en) * | 2016-04-18 | 2017-12-08 | 丰田自动车株式会社 | Abnormity determining device and abnormality determination method |
| CN107133343A (en) * | 2017-05-19 | 2017-09-05 | 哈工大大数据产业有限公司 | Big data abnormal state detection method and device based on time series approximate match |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2019155533A1 (en) | 2019-08-15 |
| JP6704543B2 (en) | 2020-06-03 |
| DE112018006760T5 (en) | 2020-10-01 |
| CN111684376A (en) | 2020-09-18 |
| JPWO2019155533A1 (en) | 2020-04-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8079001B2 (en) | Verification of requirements specification, design specification, and computer-readable storage medium apparatus, and method thereof | |
| US20150103145A1 (en) | Image processing apparatus, imaging apparatus, and image processing method | |
| US8582810B2 (en) | Detecting potential changed objects in images | |
| JPWO2019087803A1 (en) | Image processing equipment, image processing methods and programs | |
| JP2020030692A (en) | Discrimination device and machine learning method | |
| JP4858913B2 (en) | Data storage device, data storage method, and data storage program | |
| JP2019149154A (en) | Information processor, information processing method, and program | |
| CN116402739A (en) | Quality evaluation method and device for electronic endoscope detection flow | |
| JP6874864B2 (en) | Image processing equipment, image processing methods and programs | |
| JP7537753B2 (en) | Management device and management method | |
| JP2008146157A (en) | Network abnormality decision device | |
| CN111684376B (en) | Sequence data analysis device, sequence data analysis method, and computer-readable recording medium | |
| US20180286140A1 (en) | Information processing apparatus and information processing method | |
| JP6405851B2 (en) | Predictive detection support program, method, apparatus, and predictive detection program, | |
| US20230351729A1 (en) | Learning system, authentication system, learning method, computer program, learning model generation apparatus, and estimation apparatus | |
| US11216667B2 (en) | Information processing apparatus, method for information processing, and storage medium | |
| JP2019053527A (en) | Assembly work analysis device, assembly work analysis method, computer program, and storage medium | |
| US12080057B2 (en) | Image analysis apparatus, image analysis method, and storage medium | |
| JP7327548B2 (en) | Inspection support device, inspection support method and program | |
| JP7511797B2 (en) | Maintenance support system, maintenance support method, and maintenance support program | |
| US9100512B2 (en) | Reading apparatus and method of controlling the same | |
| WO2023089745A1 (en) | Computation processing device | |
| US11449703B2 (en) | Electronic apparatus, and non-transitory computer readable recording medium that stores image determination program | |
| JP5923394B2 (en) | Recognition device, recognition method, and recognition system | |
| WO2023188239A1 (en) | Business operation assistance system, business operation assistance device, business operation assistance method, and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |