Data analysis apparatus and method

Abstract

The present invention provides a data analysis device that can easily suppress interruptions in the data sequence obtained by detecting objects in the field over time. [Solution] The data analysis device (2) in this disclosure controls the generation of a data sequence showing the results of detecting an object in a time series at the site. The data analysis device includes a display unit (23) for displaying information, an input unit (22) for inputting user operations, and a control unit (20) for controlling a data management unit that manages the detection results of an object and generates a data sequence. The control unit causes the display unit to display display information including a range where the ends of multiple data sequences are adjacent to each other at the site (S1), receives a user operation at the input unit to adjust parameters indicating the conditions under which multiple data sequences are connected to each other in the adjacent range (S3), and controls the data management unit to apply the adjusted parameters in the adjacent range according to the user operation at the input unit (S4).

Description

Technical Field

[0001] This disclosure relates to a data analysis apparatus and method.

Background Art

[0002] Patent Document 1 discloses a video analysis apparatus that tracks a person from continuous image frames captured by a camera. This video analysis apparatus detects a person area for each image frame and outputs a score, outputs a score for the change in the person area for each pair of two image frames, recognizes a real person from each person area and outputs a score, and outputs a score for the change in person recognition of the person area. The video analysis apparatus assigns a movement line ID and associates a person ID to each person area in the image frame using all the above scores. Thereby, it is possible to continue person tracking robustly against person occlusion.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] This disclosure provides a data analysis apparatus that can easily suppress breaks in a data series obtained by detecting an object at a site in time series.

Means for Solving the Problems

[0005] The data analysis device in this disclosure controls the generation of data sequences showing the results of detecting objects in a field over time. The data analysis device comprises a display unit for displaying information, an input unit for receiving user operations, and a control unit for controlling a data management unit that manages the detection results of objects and generates data sequences. The control unit causes the display unit to display display information including the range in which the ends of multiple data sequences are adjacent to each other at the field, receives user operations at the input unit to adjust parameters indicating the conditions under which multiple data sequences are connected to each other in the adjacent range, and controls the data management unit to apply the adjusted parameters in the adjacent range according to the user operations at the input unit.

[0006] These general and specific embodiments may be implemented by systems, methods, and computer programs, or combinations thereof. [Effects of the Invention]

[0007] According to the data analysis device described in this disclosure, it is possible to easily suppress interruptions in the data sequence of objects detected in the field over time. [Brief explanation of the drawing]

[0008] [Figure 1] A diagram showing an overview of the movement flow analysis system according to Embodiment 1 of this disclosure. [Figure 2] Block diagram illustrating the configuration of a data analysis device in a movement flow analysis system. [Figure 3] Block diagram illustrating the configuration of a movement management server in a movement analysis system. [Figure 4] A diagram illustrating the breaks in movement patterns in a movement pattern analysis system. [Figure 5] A flowchart illustrating the operation of a data analysis device in a movement flow analysis system. [Figure 6] A flowchart illustrating the process of visualizing interrupted data in a data analysis device. [Figure 7] This figure shows an example of the display screen for movement analysis in a data analysis device. [Figure 8] This figure shows an example of how detailed information is displayed in a data analysis device. [Figure 9] A flowchart illustrating parameter adjustment processing in a data analysis device. [Figure 10] A diagram illustrating a discontinuation pattern table in a data analysis device. [Figure 11] This figure shows an example of the display of parameter adjustment processing in a data analysis device. [Figure 12] A diagram illustrating parameter adjustment information in a data analysis device. [Figure 13] A diagram illustrating a modified version of a data analysis device. [Modes for carrying out the invention]

[0009] The embodiments will be described in detail below, with reference to the drawings as appropriate. However, unnecessarily detailed explanations may be omitted. For example, detailed explanations of already well-known matters and redundant explanations of substantially identical configurations may be omitted. This is to avoid the following explanation becoming unnecessarily verbose and to facilitate understanding for those skilled in the art.

[0010] The applicant provides the accompanying drawings and the following description so that a person skilled in the art can fully understand the disclosure, and not intends to limit the subject matter described in the claims.

[0011] (Embodiment 1) Hereinafter, Embodiment 1 of this disclosure will be described with reference to the drawings.

[0012] 1. Structure The configuration of the system using the data analysis device according to Embodiment 1 will be explained with reference to Figure 1.

[0013] 1-1. About the movement flow analysis system Figure 1 shows an overview of the flow line analysis system 1 according to the present embodiment. As shown in FIG. 1, for example, the system 1 includes a plurality of cameras 11, a data analysis device 2, and a flow line management server 3. Each device 11, 2, 3 of the system 1 is connected to a communication network 13 such as a LAN, WAN, or the Internet, and can communicate data with each other.

[0014] In the system 1, for example, in a workplace 10 such as a factory, a logistics warehouse, or a store where one or more workers W work, information such as the trajectory of each worker W, that is, the flow line, is accumulated for analysis. The system 1 can be applied to data analysis in which a user 15 such as a manager or an analyst of the workplace 10 analyzes the time distribution or efficiency of individual workers W working in the workplace 10.

[0015] In the above data analysis, it is useful to obtain the flow line of each individual worker W continuously from the beginning to the end without interruption. The data analysis device 2 of the present embodiment can make it easier to obtain a continuous flow line with suppressed interruptions in the flow line analysis system 1 by reflecting the analysis results of the user 15.

[0016] In the system 1, each camera 11 is arranged, for example, in the workplace 10 so that different ranges are included in each captured image. The number of cameras 11 in the system 1 is not particularly limited to a plurality, and may be one camera. The camera 11 is communicatively connected to the flow line management server 3 via the communication network 13 so as to be able to transmit video data D0 of the imaging result of the workplace 10. The camera 11 may be, for example, an omnidirectional camera or a box camera.

[0017] The flow line management server 3 is a server device that accumulates and manages information such as video data D0 of the imaging result by each camera 11 and flow line data D1 indicating various flow lines based on the video data D0. The configuration of the flow line management server 3 will be described later.

[0018] 1-2. Configuration of the data analysis device In this embodiment, the data analysis device 2 presents information for user 15 to analyze from the accumulated information such as video data D0 and movement data D1 in the system 1. The data analysis device 2 is composed of an information processing device such as a PC (personal computer). The configuration of the data analysis device 2 will be explained with reference to Figure 2.

[0019] Figure 2 is a block diagram illustrating the configuration of the data analysis device 2. The data analysis device 2 illustrated in Figure 2 comprises a control unit 20, a storage unit 21, an operation unit 22, a display unit 23, a device interface 24, and a network interface 25. Hereinafter, interfaces will be abbreviated as "I / F".

[0020] The control unit 20 includes, for example, a CPU or MPU that works in cooperation with software to realize predetermined functions. The control unit 20 controls, for example, the overall operation of the data analysis device 2. The control unit 20 reads data and programs stored in the storage unit 21, performs various calculations, and realizes various functions.

[0021] The control unit 20 executes a program that includes a set of instructions for realizing each of the above functions. This program may be provided from the communication network 13 or stored on a portable recording medium. The control unit 20 may also be a dedicated electronic circuit or a hardware circuit such as a reconfigurable electronic circuit designed to realize each of the above functions. The control unit 20 may be composed of various semiconductor integrated circuits such as a CPU, MPU, GPU, GPGPU, TPU, microcontroller, DSP, FPGA, and ASIC.

[0022] The storage unit 21 is a storage medium that stores the programs and data necessary to realize the functions of the data analysis device 2. As shown in Figure 2, the storage unit 21 includes a storage unit 21a and a temporary storage unit 21b.

[0023] The storage unit 21a stores parameters, data, and control programs for realizing predetermined functions. The storage unit 21a is composed of, for example, an HDD or an SSD. For example, the storage unit 21a stores the above-mentioned programs and the interruption pattern table D10. The interruption pattern table D10 is an example of pattern information that patterns factors that are anticipated to cause interruptions in the movement path 40, as will be described later.

[0024] The temporary storage unit 21b is composed of RAM such as DRAM or SRAM, and temporarily stores (i.e., holds) data. For example, the temporary storage unit 21b holds movement data D1 and video data D0 received from the movement management server 3 (Figure 1). The temporary storage unit 21b may also function as a work area for the control unit 20, or it may be composed of a storage area in the internal memory of the control unit 20.

[0025] The operation unit 22 is a general term for the operating components that the user operates. The operation unit 22 may be, for example, a keyboard, mouse, touchpad, touch panel, buttons, and switches. The operation unit 22 may also include virtual operating components such as various buttons or pointers displayed on the display unit 23. The operation unit 22 is an example of an input unit that acquires various information input by the user's operation.

[0026] The display unit 23 is an example of an output unit, which may be composed of, for example, a liquid crystal display or an organic EL display. The display unit 23 may display various types of information, such as virtual operating members of the operation unit 22 and information input from the operation unit 22.

[0027] The device interface 24 is a circuit for connecting external devices to the data analysis device 2. The device interface 24 is an example of a communication unit that communicates according to a predetermined communication standard. The predetermined standard includes USB, HDMI (registered trademark), IEEE1394, WiFi, Bluetooth, etc. The device interface 24 may constitute an input unit that receives various information from external devices or an output unit that transmits various information to external devices in the data analysis device 2.

[0028] The network interface 25 is a circuit for connecting the data analysis device 2 to the communication network 13 via a wireless or wired communication line. The network interface 25 is an example of a communication unit that performs communication in accordance with a predetermined communication standard. The predetermined communication standard includes communication standards such as IEEE 802.3, IEEE 802.11a / 11b / 11g / 11ac, etc. The network interface 25 may constitute an input unit that receives various information or an output unit that transmits information via the communication network 13 in the data analysis device 2.

[0029] The configuration of the data analysis device 2 described above is just one example, and the configuration of the data analysis device 2 is not limited to this. For example, the data analysis device 2 may be composed of various computers, including a server device. The data analysis device 2 may also be configured integrally with the movement management server 3.

[0030] Furthermore, the input unit in the data analysis device 2 may be realized through cooperation with various software in the control unit 20, etc. The input unit in the data analysis device 2 may acquire various information by reading various information stored in various storage media (e.g., storage unit 21a) into the work area of ​​the control unit 20 (e.g., temporary storage unit 21b).

[0031] Furthermore, the display unit 23 of the data analysis device 2 may utilize various display devices such as a projector and a head-mounted display. Additionally, if an external display device is used, the display unit 23 of the data analysis device 2 may be an output interface circuit for video signals, such as one conforming to the HDMI® standard.

[0032] 1-3. Configuration of the traffic flow management server The configuration of the movement management server 3 in this embodiment will be explained with reference to Figure 3. The movement management server 3 is an example of a data management unit in this embodiment.

[0033] Figure 3 is a block diagram illustrating the configuration of the movement management server 3 in this system 1. The movement management server 3 comprises a control unit 30 and a storage unit 31, as shown in Figure 3, for example. The movement management server 3 further comprises interface circuits (not shown) that communicate with the camera 11 and the data analysis device 2, respectively, via a communication network 13 (Figure 1).

[0034] In the movement management server 3, the control unit 30 includes, for example, a CPU or MPU that works in cooperation with software to realize predetermined functions, and executes calculation processing such as a program that includes a set of instructions for realizing each of the above functions. For example, the control unit 30 has a functional configuration that includes a machine learning unit 32 and a movement extraction unit 33.

[0035] Furthermore, the control unit 30 may be a dedicated electronic circuit or a hardware circuit such as a reconfigurable electronic circuit designed to realize each of the above functions. The control unit 30 may be composed of various semiconductor integrated circuits such as a CPU, MPU, GPU, GPGPU, TPU, microcontroller, DSP, FPGA, and ASIC.

[0036] The storage unit 31 is a storage medium that stores programs and data necessary to realize the functions of the movement management server 3, and includes, for example, an HDD or SSD. The storage unit 31 may also include RAM such as DRAM or SRAM, and may function as a work area for the control unit 30. For example, the storage unit 31 stores video data D0, movement data D1, image recognition model M1, and map information D2. The map information D2 shows the arrangement of equipment, etc., in a predetermined coordinate system (hereinafter also referred to as "map coordinates"), for example, in the workplace 10.

[0037] The machine learning unit 32 performs machine learning on the image recognition model M1 based on training data that includes, for example, video data D0 of images of the workspace 10 taken in advance, and annotation information related to those images. The machine learning unit 32 stores the trained image recognition model M1 in the storage unit 31.

[0038] Annotation information is information that, for example, is manually entered by a user 15, etc., to determine whether or not an object to be detected by image recognition, such as a worker W, is visible in the image, and the position of the object in the image. Annotation information may be stored in the storage unit 31 as appropriate, such as when training the image recognition model M1. In this system 1, the data analysis device 2 may further have a function to set annotation information.

[0039] Image recognition model M1 is configured using machine learning to detect the position of objects such as worker W in the input image, for example, when an image such as a frame from video data D0 is input. Various machine learning models, such as convolutional neural networks, can be employed in image recognition model M1. Image recognition model M1 may also detect the type of object appearing in the input image. For example, worker W performing multiple pre-set tasks may be detected as different types of objects, or individual worker W may be identified.

[0040] The movement path extraction unit 33 extracts the movement path of an object from the results of sequential object detection by the image recognition model M1 in the video data D0, based on the video data D0, the image recognition model M1, and the map information D2, and generates movement path data D1 that can be managed using map coordinates. In this embodiment, the movement path extraction unit 33 may be an example of a data management unit. The movement path data D1 shows an example of a movement path in a data sequence that is constructed by sequentially connecting the detection positions for each frame by the image recognition model M1.

[0041] The movement path extraction unit 33 has, for example, parameters for connecting frame-by-frame detection positions by the image recognition model M1 when extracting movement paths, i.e., movement path connection parameters P1. The movement path connection parameters P1 indicate connection conditions, such as the range in which other detection positions are allowed as connection destinations from the source detection position, and include, for example, the distance range from the source to the destination, the connection direction, and the waiting time. By connecting the detection positions, movement paths having each detection position at one end can be connected to each other.

[0042] In this system 1, the movement path connection parameter P1 in the movement path management server 3 is configured to be adjustable by user operation in the data analysis device 2. The movement path connection parameter P1 can be set to different values ​​for each region on the map coordinates based on the map information D2, or on the camera coordinates corresponding to the image captured by the camera 11. Setting information indicating these various settings for the movement path connection parameter P1 is stored, for example, in the storage unit 31.

[0043] The configuration of the movement management server 3 in this system 1 described above is just one example, and the configuration of this system 1 is not limited to this. The various functions of the movement management server 3 in this embodiment may each be implemented using distributed computing. Furthermore, the movement management server 3, and by extension this system 1, does not need to include a machine learning unit 32.

[0044] Furthermore, the movement path extraction unit 33 may extract movement paths using machine learning. The movement path extraction unit 33 may extract movement paths using machine learning, and various parameters related to such machine learning may be adjustable from the data analysis device 2. In addition, the image recognition model M1 may be configured by implementing various image recognition algorithms, not limited to machine learning.

[0045] 2.Operation The operation of the movement analysis system 1 and data analysis device 2, which are configured as described above, will be explained below.

[0046] 2-1. System Operation In this embodiment, an example of the operation of automatically extracting the movement of worker W using a trained image recognition model M1 in the movement analysis system 1 (Figure 1) will be described. In this system 1, for example, multiple cameras 11 sequentially capture images of their respective imaging ranges in the workplace 10 frame by frame, and transmit the resulting video data D0 to the movement management server 3.

[0047] When the movement management server 3 receives video data D0 from each camera 11, it stores it in the storage unit 31 and inputs it to the movement extraction unit 33. The movement extraction unit 33 uses a trained image recognition model M1 to detect the position of the worker W in the captured images for each frame indicated by the input video data D0. The movement extraction unit 33 identifies the destination detection position relative to the source detection position between multiple detection positions for each frame, for example, according to a pre-set movement connection parameter P1, and extracts a movement path in which each identified detection position is sequentially connected. The movement extraction unit 33 generates movement data D1 indicating the extracted movement path and stores it in the storage unit 31.

[0048] The data analysis device 2 communicates with the movement management server 3 and receives various information from the movement management server 3, for example, in response to the operations of the user 15. Based on the information obtained in this manner, the data analysis device 2 generates various information that the user 15 uses for analysis of the work area 10, and displays it on the display unit 23.

[0049] 2-1-1. Regarding interruptions in the flow of movement In the automatic extraction of movement paths as described above, a problem may arise where a continuous movement path is extracted in a fragmented manner due to various factors in the workspace 10. To resolve this fragmentation of movement paths, the inventors of this invention have diligently researched how to appropriately adjust the movement path connection parameter P1, and have devised the data analysis device 2 of this embodiment. These findings will be explained with reference to Figure 4.

[0050] Figure 4 is a diagram illustrating the interruption of the movement path 40 in this system 1. Figure 4(A) shows examples of areas where the movement path 40 in the movement path data D1 is interrupted, i.e., interruption areas R11 and R12, on the work area map 50 shown by the map information D2 of the work area 10. Figure 4(B) shows an example of an image 51 of the first interruption area R11 in the video data D0. Figure 4(C) shows an example of an image 52 of the second interruption area R12.

[0051] The first interrupted area R11, illustrated in Figures 4(A) and (B), is a situation in the workplace 10 where an obstacle 12, such as a ceiling beam, is superimposed on the passageway through which worker W passes, and is captured in the image 51 of the camera 11. In such an interrupted area R11, worker W moving along the passageway passes behind the obstacle 12 from the perspective of the camera 11. In this case, it is conceivable that worker W may not be detected by the image recognition model M1, resulting in an interruption in the extracted movement path 40. In this case, it is thought that the interruption in the movement path 40 can be resolved by using the direction of movement path connection as the direction of movement path connection, based on the predicted destination of worker W when moving along the passageway after the movement path 40 is interrupted.

[0052] Furthermore, in the second interrupted area R12 illustrated in Figures 4(A) and (C), the obstacle 12 is superimposed on the area where worker W is working in the workspace 10 and appears in the captured image 52. In such an interrupted area R12, it is conceivable that when worker W is working behind the obstacle 12 as seen from the camera 11, worker W may not be detected and the movement path 40 may be interrupted. In this case, it is predicted that worker W behind the obstacle 12 will reappear in the captured image 52 near the obstacle 12 after finishing work, and it is thought that the interruption of the movement path 40 can be eliminated by reflecting this prediction in the adjustment of the movement path connection parameter P1.

[0053] As described above, analyzing the various situations in which the movement path 40 is interrupted would be difficult to implement using AI technology, etc. However, the inventors of this invention focused on the fact that the user 15 of this system 1 can specifically identify the causes of the interruption by visually observing the current state of the workspace 10. Therefore, the data analysis device 2 of this embodiment provides a user interface that visualizes these situations of interruption in the movement path 40 to the user 15 and allows the movement path connection parameter P1 to be adjusted in accordance with the user 15's analysis. The operation of the data analysis device 2 of this embodiment will be described in detail below.

[0054] 2-2. Operation of the Data Analysis Device The overall operation of the data analysis device 2 in this embodiment will be explained with reference to Figure 5.

[0055] Figure 5 is a flowchart illustrating the operation of the data analysis device 2 in this system 1. Each process in the flowchart illustrated in Figure 5 is executed, for example, by the control unit 20 of the data analysis device 2. The processing of this flowchart starts, for example, when the movement management server 3 has obtained movement data D1 from the video data D0 of the camera 11's image capture results using a trained image recognition model M1 and pre-set movement connection parameters P1.

[0056] First, the control unit 20 of the data analysis device 2 performs a process (S1) on the display unit 23 to visualize the status of interruptions in the movement path 40 in the work area 10 for the user 15, based on the movement path data D1 generated in the current movement path management server 3. In the interruption visualization process (S1), various display information regarding the interruption areas R11 and R12 of the movement path 40 in the work area 10 is presented to the user 15, for example, as shown in Figure 4. Details of the interruption visualization process (S1) will be described later. Hereinafter, the collective term for the interruption areas R11 and R12 will be referred to as interruption area R1.

[0057] User 15 can check the various display information visualized in step S1 and, if there is a break area R1 where the movement path connection parameter P1 should be adjusted, input an operation to specify the break area R1 in the operation unit 22. The control unit 20 determines whether or not the break area R1 has been specified based on the user operation input in the operation unit 22 (S2).

[0058] If a break area R1 is specified by the input user operation, the control unit 20 performs a process to accept the user operation to adjust the movement path connection parameter P1 in the specified break area R1 (S3). In the parameter adjustment process (S3), various information is presented to guide the user 15 in order to adjust the movement path connection parameter P1 in a manner that appropriately reflects the situation identified by the user 15 in the break area R1 specified by the user (S3). Details of the parameter adjustment process (S3) will be described later.

[0059] Next, the control unit 20 transmits parameter adjustment information, including the movement path connection parameter P1, which is the result of the parameter adjustment process (S3) for the interrupted area R1 specified by the user 15, to the movement path management server 3 (S4). After that, the control unit 20 repeats the process from step S1 onwards, for example. The interruption visualization process (S1) at this time is performed, for example, using movement path data D1, which is extracted again by the movement path management server 3, reflecting the adjustment results of step S3. For example, the user 15 can confirm the adjustment results of step S3 in the subsequent step S1 and repeat the parameter adjustment process (S3) until the interrupted area R is eliminated.

[0060] If the interruption area R1 is not specified by the input user operation (NO in S2), the control unit 20 terminates the process shown in this flowchart.

[0061] Through the above process, the data analysis device 2 visualizes the current state of interruptions in the movement paths 40 extracted by the movement path management server 3 to the user 15 (S1), and adjusts the movement path connection parameter P1 according to the user's operation (S3). This makes it easier to suppress interruptions in the movement paths 40 extracted in the work area 10 using the trained image recognition model M1 by reflecting the user 15's analysis.

[0062] 2-2-1. Visualization of interruptions The details of the discontinuity visualization process in step S1 of Figure 5 will be explained using Figures 6 to 8.

[0063] Figure 6 is a flowchart illustrating the interruption visualization process (S1 in Figure 5) in the data analysis device 2. Figure 7 shows an example of the movement analysis screen display in the data analysis device 2. Figure 8 shows an example of the detailed information display in the data analysis device 2.

[0064] First, the control unit 20 acquires movement data D1 for a predetermined period (e.g., one day) to be analyzed from the movement management server 3, for example via the network I / F 25, and displays a movement analysis screen including the movement 40 shown by the acquired movement data D1 on the display unit 23 (S11). An example of the display in step S11 is shown in Figure 7(A).

[0065] For example, in step S11, the movement management server 3 transmits movement data D1, which is the result of the movement extraction unit 33 extracting movement 40 from the detection results of the image recognition model M1 based on the current movement connection parameter P1, to the data analysis device 2, along with, for example, map information D2. For example, as shown in Figure 7(A), the display unit 23 of the data analysis device 2 displays a movement analysis screen in which the movement 40 indicated by the acquired movement data D1 is superimposed on the work area map 50 of the map information D2. The map information D2 may be stored in advance in the storage unit 21 of the data analysis device 2.

[0066] Next, the control unit 20 displays a notification to the user 15 on the movement analysis screen shown in Figure 7(A) regarding candidates that are assumed to be interrupted areas R1 (S12). An example of the display in step S12 is shown in Figure 7(B).

[0067] For example, in step S12, the control unit 20 first detects an area as a candidate discontinuation area R10 based on the acquired movement data D1, where the start and end points of the movement 40 are adjacent within a predetermined distance in the map information D2, and the number of adjacent ends is greater than or equal to a predetermined number. The predetermined distance and predetermined number are set in advance as criteria for determining, for example, that the ends of the movement 40 are densely packed.

[0068] Furthermore, in step S12, the control unit 20, for example, by referring to map information D2, excludes the area corresponding to the entrance and exit of the work area 10 from the notification target in the interruption candidate area R10 detected as described above. The targets for exclusion are not limited to the above, and can be set to various areas where both ends of the movement path 40 are assumed to be located, regardless of the interruption of the movement path 40, and may include, for example, the field of view of the camera 11 or the end area of ​​the detection range of this system 1.

[0069] The control unit 20 detects the remaining candidate interruption area R10 that was not excluded as described above as a target for notification, and, for example as shown in Figure 7(B), attaches a notification marker 45 to the detected candidate interruption area R10 and displays it on the display unit 23 (S12). The notification marker 45 is a marker for notifying the user 15 of the candidate interruption area R1.

[0070] The notification marker 45 may be in the form of various icons, or it may be a highlighting of the break candidate area R10, such as highlighting the color, hatching, or blinking. Also, the notification in step S12 does not necessarily have to display a marker, and only the break candidate area R10 that is the target of the notification may be displayed.

[0071] The control unit 20, for example, when the movement analysis screen shown in Figure 7(B) is displayed, receives various user operations from the operation unit 22 and determines whether or not a candidate interruption area R1 has been selected by the user operation (S13). For example, user 15 can input an operation to the operation unit 22 to select a desired interruption candidate area R10 from among the interruption candidate areas R10 notified on the movement analysis screen shown in Figure 7(B) using the pointer 4.

[0072] When a candidate interruption area R10 is selected by user operation (YES in S13), the control unit 20 displays detailed information about the selected candidate interruption area R10 on the display unit 23 (S14). The detailed information includes various information to allow the user 15 to confirm the situation in which the movement path 40 is interrupted in the work area 10. An example of the display in step S14 is shown in Figure 8.

[0073] In the example shown in Figure 8, the display unit 23 displays a movement path list 55 and a movement path video 56, which are examples of detailed information for each of the candidate interruption areas R10 selected by user operation in step S13. The movement path list 55 lists various information about each movement path 40 whose end is located within the range of the candidate interruption area R10, relating them to each other. The movement path video 56 is a video in which the candidate interruption area R10 was captured during a time period related to one of the movement paths 40 in the movement path list 55.

[0074] The information in the movement path list 55 includes, for example, the identification number of the movement path 40, "Movement Path No.", the time corresponding to the start of the movement path 40, the time corresponding to the end of the movement path 40, and "Video Link," which is link information associated with the movement path video 56 of the movement path 40. The movement path list 55 may also include information indicating whether the end of each movement path 40 located in the interruption candidate area R10 is the start or end. For example, either the "Start Time" or the "End Time" may be highlighted, or only one of them may be displayed.

[0075] In the example shown in Figure 8, a user operation is input to the operation unit 22 by using pointer 4 to specify the video link for movement path No. 2 from the movement path list 55. When a video link is specified, the control unit 20 obtains the corresponding video data D0 from, for example, the movement path management server 3 and plays and displays the movement path video 56. The movement path video 56 is played and displayed from, for example, the vicinity of the time when the corresponding movement path 40 was located in the selected interruption candidate area R10.

[0076] The various detailed information described above is displayed on the display unit 23 in response to user operation, for example, through a pop-up on the movement analysis screen (S14). By checking the information on the movement analysis screen and the various detailed information, the user 15 can identify whether a break in the movement path 40 occurred in the break candidate area R10 selected in step S13, or various other situations such as the cause of the break.

[0077] The control unit 20 determines, for example, whether the selected interruption candidate area R10 has been designated as interruption area R1 by user operation, based on user operation in the operation unit 22 (S15). The user operation targeted in step S15 may be, for example, an operation in which user 15 uses a pointer 4 or the like to specify a range on the work area map 50 or movement video 56 in which they believe an interruption has occurred in the interruption candidate area R10. Alternatively, it may be an operation to specify the entire selected interruption candidate area R10, or it may be an operation such as clicking the notification marker 45.

[0078] If the control unit 20 is designated as a discontinued area R1 by user operation (YES in S15), it obtains information indicating the discontinued area R1 designated by user operation (S16) and terminates the discontinuity visualization process (S1 in Figure 5). In this case, the control unit 20 proceeds to YES in step S2 in Figure 5.

[0079] On the other hand, if the control unit 20 does not designate an interruption area R1 by user operation (NO in S15), it repeats the processing from step S13 onwards, for example. For example, if user 15 confirms that no interruption has occurred in the interruption candidate area R10 that was selected once, user 15 can select a new interruption candidate area R10 (S13).

[0080] If, for example, the user does not select a candidate interruption area R10 (NO in S13), the control unit 20 does not specifically acquire information indicating the interruption area R1 (S16) and terminates the interruption visualization process (S1 in Figure 5). In this case, the control unit 20 proceeds to NO in step S2 in Figure 5.

[0081] According to the above interruption visualization process (S1 in Figure 5), the data analysis device 2 can visualize the current state of interruptions in the movement path 40 to the user 15 by generating various display information, for example, as illustrated in Figures 7 and 8.

[0082] 2-2-2. Parameter Adjustment Process The parameter adjustment process in step S3 of Figure 5 will be explained using Figures 9 to 12.

[0083] Figure 9 is a flowchart illustrating the parameter adjustment process (S3 in Figure 5) in the data analysis device 2. Figure 10 illustrates the interruption pattern table D10 in the data analysis device 2. Figures 11(A) to (C) show examples of the parameter adjustment process display.

[0084] The process illustrated in Figure 9 is performed, for example, using a discontinuation pattern table D10 pre-stored in the storage unit 21 of the data analysis device 2. The discontinuation pattern table D10 illustrated in Figure 10 manages multiple patterns by associating each pattern's identifying "No.", corresponding "use case", "designated area", and "adjustment parameter" with each other. The designated area is an example of the requirements to which a change in the movement connection parameter P1 is applied in that pattern, and indicates, for example, a location where a discontinuation is expected in the corresponding use case. The adjustment parameter indicates the type of movement connection parameter P1 that is expected to be adjusted in that pattern.

[0085] For example, in the interruption pattern table D10 in Figure 10, pattern No. 1 corresponds to the use case of "passing behind an obstacle" (see Figures 4(B) and 8), and includes the specified area "location where the connection is interrupted by an obstacle" and the adjustment parameter "connection direction". Pattern No. 2 corresponds to the use case of "working behind an obstacle" (see Figure 4(C)), and includes the specified area "work location" and the adjustment parameters "waiting time" and "distance range". Pattern No. 3 corresponds to the use case of "entering and exiting a room, etc.", and includes the specified area "room entrance / exit" and the adjustment parameter "waiting time". Pattern No. 4 shows a custom setting where, for example, user 15 can configure each item.

[0086] In the example shown in Figure 9, the data analysis device 2 uses, for example, the interruption pattern table D10 to sequentially display the screens shown in Figures 11(A) to (C) in dialog boxes, etc., in addition to the movement analysis screens and detailed information shown in Figures 7 and 8, etc., in order to guide the user 15.

[0087] Specifically, the control unit 20 of the data analysis device 2 first controls the display unit 23 to display a screen that presents the user 15 with use case options for the interrupted area R1 being adjusted, as specified in step S16 of Figure 6 (S21). An example of the display in step S21 is shown in Figure 11(A).

[0088] The use case selection screen illustrated in Figure 11(A) includes a message prompting the user 15 to select a use case, and multiple use cases as options. The control unit 20, for example, refers to the use cases for each pattern in the interruption pattern table D10, displays the use case selection screen on the display unit 23 (S21), and accepts user input at the operation unit 22.

[0089] User 15 can input an operation to the operation unit 22 to select a use case that is suitable for the interrupted area R1 being adjusted, from among several use case options, while checking information such as a movement video 56. For example, in the example in Figure 8 (and Figure 4(B)), use case No. 1 is selected. The control unit 20 acquires information on the selected use case based on the user operation in the operation unit 22 (S22).

[0090] Next, the control unit 20 displays a screen on the display unit 23 for specifying the requirements under which the change in the movement path connection parameter P1 is applied, according to the acquired use case information, etc. (S23). An example of the display in step S23 is shown in Figure 11(B).

[0091] Figure 11(B) illustrates the case where use case No. 1 in Figure 11(A) is selected. The requirements specification screen in the example of Figure 11(B) includes a message prompting the user 15 to specify requirements, a section for setting the specified area, and a section for setting other requirements. The section for setting the specified area includes the identification information of camera 11 in this system 1, an input field 61 for the specified area in the camera coordinates of said camera 11, and an input field 62 for the specified area in the corresponding map coordinates. In this example, the specified area is specified in each input field 61, 62 by the horizontal position x, vertical position y, width w, and height h in their respective coordinate systems.

[0092] For example, in step S23, the control unit 20 sets the designated area corresponding to the use case acquired in step S22 in the interruption pattern table D10 on the requirements specification screen. The control unit 20 also identifies the camera 11 that will image the interruption area R1 specified in step S16 in Figure 6, and displays the position of the interruption area R1 specified in step S16 as the initial value for each input field 61, 62 in the camera coordinates of the identified camera 11.

[0093] User 15 can, for example, check the movement video 56 and input an operation in the camera coordinate input field 61 to specify in detail the range to which the adjustment of the movement connection parameter P1 should be applied. For example, User 15 can input the position and size of the specified area so that the location where the camera coordinates are interrupted by the X obstacle 12 is enclosed by a rectangle or the like. The control unit 20 acquires information on the specified requirements, such as the specified area, based on the user operation on the operation unit 22 while the requirements specification screen is displayed (S24).

[0094] The above user operations may also be entered into the map coordinate input field 62. The control unit 20 calculates coordinate transformations between camera coordinates and map coordinates as appropriate so that each input field 61, 62 indicates the same position. The control unit 20 may also highlight the entered specified area on the work area map 50 or work area video 56 of the work area analysis screen on the display unit 23. The control unit 20 may also accept user operations to input a specified area on the work area map 50 or work area video 56 using a pointer 4 or the like via the operation unit 22.

[0095] Furthermore, the control unit 20 displays a screen for setting the movement path connection parameter P1 on the display unit 23 according to information such as acquired use cases (S25). An example of the display in step S25 is shown in Figure 11(C).

[0096] Figure 11(C) illustrates the parameter adjustment screen that follows Figure 11(B). The parameter adjustment screen in the example of Figure 11(C) includes a message prompting the user to set the movement path connection parameter P1, a section for setting the adjustment parameter for the selected use case among the various movement path connection parameters P1, and a section for setting others. The section for setting the various parameters P1 also includes an input field 63 for entering the setting value of the adjustment result.

[0097] For example, in step S25, the control unit 20 controls the display unit 23 to display such parameter adjustment screens by referring to, for example, the interruption pattern table D10. The control unit 20 also obtains the current settings of various parameters P1 by data communication with, for example, the movement management server 3 and displays them as initial values ​​in each input field 63.

[0098] User 15 can input user operations to change the setting value from the initial value displayed in the corresponding input field 63 in order to adjust the desired movement path connection parameter P1. Based on the user operations in the operation unit 22 while the parameter adjustment screen is displayed, the control unit 20 obtains the setting value of the movement path connection parameter P1 as a result of the adjustment (S26).

[0099] Based on the information obtained from the various user operations described above, the control unit 20 stores parameter adjustment information showing the adjustment result of the movement line connection parameter P1 (S27). The parameter adjustment information from step S27 is illustrated in Figure 12.

[0100] Figure 12 illustrates parameter adjustment information D15 corresponding to the adjustment results in Figures 11(A) to (C). For example, the control unit 20 generates parameter adjustment information D15 by relating the requirements such as the designated area acquired in step S24 and the setting value of the movement connection parameter P1 acquired in step S26, and stores it in the storage unit 21 (S27). The control unit 20 may also include use case information acquired in step S22 in the parameter adjustment information D15.

[0101] The control unit 20 terminates the parameter adjustment process (S3 in Figure 5) by saving the parameter adjustment information D15 (S27).

[0102] According to the parameter adjustment process described above (S3 in Figure 5), the data analysis device 2 presents various information to the user 15 prompting adjustment of the movement path connection parameter P1, and obtains parameter adjustment information D15 that reflects the user 15's intentions (S27). In the subsequent step S4, the control unit 20 transmits the saved parameter adjustment information D15 to the movement path management server 3. This controls the subsequent generation of movement path data D1 by the movement path management server 3.

[0103] When the movement management server 3 receives the parameter adjustment information D15, it refers to the received parameter adjustment information D15 during the subsequent operation of the movement extraction unit 33. Specifically, the movement extraction unit 33 determines whether or not the requirements in the received parameter adjustment information D15 are met, and if the requirements are met, it extracts the movement 40 using the setting value of the movement connection parameter P1 adjusted in the parameter adjustment information D15. On the other hand, if the requirements are not met, the movement extraction unit 33 does not use the setting value of the adjustment result, but uses the setting value before adjustment.

[0104] Thus, the system 1 can generate movement path data D1 that resolves the interruption in the movement path 40 for the interruption area R1 specified by the user 15. The movement path extraction unit 33 may use the movement path connection parameter P1 when connecting the movement path again after initially extracting or connecting the movement path 40. In this case as well, the system can switch whether or not to use the set value of the adjustment result depending on whether or not it meets the corresponding requirements.

[0105] Furthermore, the parameter adjustment process described above facilitates parameter adjustment by the user 15 by guiding them to adjust the movement path connection parameter P1 using use cases related to interruptions in the movement path 40.

[0106] For example, in the examples of Figures 4(B) and 8, user 15 selects a use case in step S22 where the user passes behind the obstacle 12 (Figure 11(A)). This leads to subsequent steps S23 and S25, where the user is prompted to input the location where the movement path 40 is interrupted by the obstacle 12 as a specified area, and to set the connection direction of the corresponding adjustment parameters, as shown in Figures 11(B) and (C).

[0107] Furthermore, in the example shown in Figure 4(C), by selecting a use case where work is performed behind obstacle 12, the system is guided to set a distance range corresponding to the area where the work area is hidden, and to set a waiting time corresponding to the work time at the work area. In this way, it is possible to easily adjust the appropriate movement path connection parameter P1 according to the use case selected by the user 15.

[0108] Furthermore, in the guidance described above, the user 15 can adjust not only the adjustment parameters corresponding to the selected use case, but also other movement path connection parameters P1, such as the waiting time when passing behind the obstacle 12, as shown in Figure 11(C) (S25, S26).

[0109] Furthermore, the requirements set in steps S23 and S24 are not limited to the specified area; other requirements may also be set. For example, the data analysis device 2 may set time information for applying changes to the movement path connection parameter P1, so that the adjustment results are applied only during specific time periods. Alternatively, the personal identification or various attributes of worker W may be set as requirements. The control unit 20 accepts user operations to specify various requirements as described above, for example, on the requirements specification screen in Figure 11(B) (S24), and includes the information of the specified requirements in the parameter adjustment information D15 (S27). By referring to the requirements in this parameter adjustment information D15, the movement path management server 3 can selectively apply the movement path connection parameter P1 when the various requirements apply.

[0110] 3. Summary As described above, the data analysis device 2 in this embodiment controls the generation of movement paths 40 as an example of a data sequence showing the results detected over time by a worker W, an example of an object at a work site such as a workplace 10. The data analysis device 2 comprises a display unit 23, an operation unit 22 (an example of an input unit), and a control unit 20. The display unit 23 displays information. The operation unit 22 receives user input. The control unit 20 controls a movement path management server 3, an example of a data management unit, which manages the detection results of objects and generates data sequences. The control unit 20 causes the display unit 23 to display various display information, including an example of a break area R1 or a candidate break area R10, where the ends of multiple movement paths 40 in the workplace 10 are adjacent to each other at intervals corresponding to breaks in the movement paths 40 (see S1, Figure 7, etc.). In this range, the ends of the movement paths 40 are adjacent at intervals that are expected to be judged as breaks in the movement paths 40. The control unit 20 receives a user operation via the operation unit 22 to adjust the traffic flow connection parameter P1, which is an example of a parameter indicating the conditions under which multiple traffic flows 40 are connected to each other in the interrupted area R1 (S3). The control unit 20 controls the traffic flow management server 3 to apply the adjusted traffic flow connection parameter P1 in the interrupted area R1 according to the user operation via the operation unit 22 (S4).

[0111] According to the data analysis device 2 described above, by visualizing the interrupted areas R1 etc. for the user 15 and adjusting the movement path connection parameter P1 through user operation, it becomes easier to suppress interruptions in the movement path 40, which detects workers W in the workplace 10 over time.

[0112] In the data analysis device 2 of this embodiment, the control unit 20 receives a user operation via the operation unit 22 that specifies the requirements for which the adjusted traffic flow connection parameter P1 should be applied in the interrupted area R1 (S24). The control unit 20 controls the traffic flow management server 3, for example, by parameter adjustment information D15, so that the adjusted traffic flow connection parameter P1 is not applied if the specified requirements are not met, and the adjusted traffic flow connection parameter P1 is applied if the specified requirements are met (S4). This allows the adjustment of the traffic flow connection parameter P1 to be applied within the range intended by the user 15, and enables accurate suppression of interruptions in the traffic flow 40.

[0113] In the data analysis device 2 of this embodiment, the control unit 20 detects the area adjacent to the ends of the movement path 40, based on the movement path 40 generated by the movement path management server 3, as an example of a potential interruption area R10 for the movement path 40, and notifies the user of the detected interruption area R10 using a notification marker 45 or the like in the display information (S12). As a result, the interruption area R10 detected by the data analysis device 2 is notified to the user 15, making it easier to suppress interruptions in the movement path 40.

[0114] In the data analysis device 2 of this embodiment, the control unit 20 receives a user operation via the operation unit 22 to specify a candidate interruption area R10 in the display information such as the movement path analysis screen (S13). The control unit 20 displays detailed information about the movement path 40 whose end is located in the specified candidate interruption area R10 on the display unit 23 (S14, see Figure 8). This makes it easier for the user 15 to analyze the interruption situation in detail.

[0115] In this embodiment, the data analysis device 2 further includes a storage unit 31 that stores a discontinuation pattern table D10, which is an example of pattern information including multiple patterns in which the worker W is not detected in the workplace 10. The control unit 20 receives a user operation via the operation unit 22 to select one pattern from the multiple patterns shown in the discontinuation pattern table D10 (S22), and displays information on the display unit 23 prompting the user to adjust the movement path connection parameter P1 according to the selected pattern (S23, S25). This makes it easier to suppress discontinuations in the movement path 40 by inducing parameter adjustments by the user 15.

[0116] In this embodiment, the movement management server 3 manages the detection results of the image recognition model M1, which detects workers W based on captured images of the work area 10, in the movement extraction unit 33, and generates movement lines 40 by sequentially connecting the detection results of the image recognition model M1 based on the movement line connection parameter P1. By adjusting the movement line connection parameter P1 applied in this way, the data analysis device 2 can control the generation of movement lines 40 and suppress interruptions in movement lines 40.

[0117] In this embodiment, the data analysis device 2 further includes a network I / F 25, which is an example of a communication unit that communicates data with the traffic flow management server 3. The control unit 20 receives traffic flow data D1 of the traffic flow 40 from the traffic flow management server 3 via the communication unit (S11), and transmits parameter adjustment information D15 of the adjusted traffic flow connection parameter P1 to the traffic flow management server 3 via the communication unit (S4). Through this data communication, the data analysis device 2 can control an externally configured data management unit to suppress interruptions in the traffic flow 40.

[0118] In the data analysis device 2 of this embodiment, the data sequence is a movement path 40 that shows the detected movement of worker W on a work map 50 representing the work area 10. The data analysis device 2 of this embodiment can easily suppress interruptions in such data sequences.

[0119] In this embodiment, a data analysis method is provided that controls the generation of a data sequence showing the results of detecting objects in a work site such as a workplace 10 over time. The data sequence is generated by a movement management server 3 that manages the detection results of objects. The method includes: step S1, displaying information on a display unit 23 that includes a range in the workplace 10 where the ends of multiple data sequences are adjacent to each other; step S3, receiving a user operation on an operation unit 22 to adjust parameters indicating the conditions under which multiple data sequences are connected to each other in adjacent ranges; and step S4, controlling the movement management server 3 to apply the adjusted parameters in adjacent ranges according to the user operation on the operation unit 22.

[0120] In this embodiment, a program is provided for causing the computer's control unit 20 to execute the above data analysis method. This data analysis method of this embodiment makes it easier to suppress interruptions in the data sequence in which objects at the site are detected in chronological order.

[0121] (Other embodiments) As described above, Embodiment 1 has been explained as an example of the technology disclosed in this application. However, the technology in this disclosure is not limited to this and can be applied to embodiments that have been modified, substituted, added, or omitted as appropriate. Furthermore, it is possible to create new embodiments by combining the components described in each of the above embodiments. Therefore, other embodiments are described below as examples.

[0122] In the above-described Embodiment 1, the movement path 40 of worker W was explained as an example of a data sequence in the data analysis device 2. The data analysis device 2 in this embodiment is not limited to the movement path 40, but may use various other data sequences. Such modifications will be explained with reference to Figure 13.

[0123] Figure 13 is a diagram illustrating a modified version of the data analysis device 2. This modified version of the data analysis device 2 is applied to a system that analyzes, for example, the efficiency of work performed by worker W. Figure 13(A) illustrates an image 53 captured by camera 11 in this system. Figure 13(B) illustrates a timeline 43 of an example of a data sequence in this system.

[0124] In the system example shown in Figure 13, the camera 11 is positioned so that, for example, the location where an individual worker W is performing work is captured in the image 53. The image recognition model M1 in this modified example detects when worker W has performed a predetermined task in the image 53. The predetermined task can be set to various tasks, such as holding or wrapping an object, or writing letters. This system includes a data management unit that generates a timeline 43 based on the detection results of the image recognition model M1, instead of, for example, the movement path extraction unit 33 or the movement path management server 3 in Embodiment 1. The timeline 43 is an example of a data sequence composed of the presence or absence of detection results indicating that worker W has performed a predetermined task, arranged in chronological order.

[0125] In the example shown in Figure 13(A), in the image area R2 near the obstacle 12 located between the camera 11 and the worker W, the image recognition model M1 may fail to detect when the worker W performs a predetermined task. Such detection failures can lead to discontinuities in the timeline 43, as shown in Figure 13(B).

[0126] In this case, since the detection results corresponding to each end of the interrupted timeline 43 are obtained in the image area R2 near the obstacle 12, it is thought that the interruption can be resolved by having the data management unit perform interpolation to connect the interrupted timeline 43 between these ends. For example, with respect to the detection results in the image area R2, it is thought that the interruption of the timeline 43 can be resolved by adjusting parameters such as extending the waiting time for connection even if the timeline 43 is interrupted, or widening the acceptable range of connection directions in the image area R2.

[0127] Therefore, the data analysis device 2 of this modified example visualizes the on-site situation to the user 15, similar to Embodiment 1, and adjusts parameters to reflect the user 15's analysis. For example, the control unit 20 of this modified example, in the same operation as Embodiment 1 (Figure 5), detects a break in the timeline 43 during the break visualization process (S1) and visualizes the break in the timeline 43 to the user 15 by displaying a notification marker 45, etc.

[0128] Furthermore, the control unit 20 of this modified example displays the captured image 53 for the corresponding time period on the display unit 23 in response to a user operation specifying a notification marker 45 for a break in the timeline 43, for example, in steps S13 and S14 of Figure 6. The control unit 20 receives, for example, a user operation in the parameter adjustment process (S3 in Figure 5) to specify the image area R2 as the designated area to which parameter adjustment should be applied, and a user operation to adjust the various parameters, via the operation unit 22. As a result, the data analysis device 2 of this modified example can reflect the user's intentions regarding the suppression of breaks in data sequences such as the timeline 43, similar to Embodiment 1.

[0129] Furthermore, the data analysis device 2 in this modified example may be applied to a system that analyzes both the timeline 43 and the movement path 40 as described above. For example, the timeline 43 for the work of worker W and the movement path 40 for the movement of worker W may be analyzed.

[0130] As described above, in the data analysis device 2 of this embodiment, the data sequence may include at least one of the movement path 40 of the object at the site and the timeline 43 showing the detection results of the object in chronological order. According to the data analysis device 2 of this embodiment, it is possible to easily suppress interruptions in various data sequences.

[0131] Furthermore, in each of the embodiments described above, an example was given in which the data management unit is an external component of the data analysis device 2, such as a movement management server 3. In this embodiment, the data management unit may be an internal component of the data analysis device 2. For example, the data analysis device 2 may include a movement extraction unit 33. In this case, in step S4 of Figure 5, the data analysis device 2 controls the generation of movement data D1 to reflect the adjustment results by setting parameter adjustment information D15 in the internal movement extraction unit 33.

[0132] In each of the embodiments described above, parameters for connecting data sequences, such as the movement path connection parameter P1, were exemplified. In the data analysis device 2 of this embodiment, the parameters for connecting data sequences are not limited to those described above, and various parameters can be used. For example, in this system 1, when determining the connection destination of movement paths using various scores, as disclosed in Patent Document 1, threshold values ​​for such scores may be adopted as the parameters described above. Similar to the embodiments described above, these parameters can also be appropriately guided to reflect the analysis of the user 15, making it easier to suppress interruptions in movement paths, etc.

[0133] In the embodiments described above, worker W was explained as an example of an object in the data analysis device 2, but the object is not particularly limited to this. In the data analysis device 2 of this embodiment, the object may be a person other than worker W, or it may be a living organism, various vehicles or robots or other moving objects. Furthermore, the object is not limited to a moving object, but may be various objects or the state of an object.

[0134] In each of the embodiments described above, a workplace 10 was used as an example of a site where the data analysis device 2 is applied. The data analysis device 2 of this embodiment is not limited to the workplace 10 and can be applied to various sites, and can be used to analyze data sequences of detection results for various objects at the site.

[0135] As described above, embodiments have been explained as examples of the technology in this disclosure. For this purpose, accompanying drawings and a detailed description have been provided.

[0136] Therefore, the components described in the attached drawings and detailed descriptions may include not only components essential for solving the problem, but also components that are not essential for solving the problem, provided that they illustrate the technology described above. For this reason, the mere presence of these non-essential components in the attached drawings and detailed descriptions should not be immediately assumed to mean that they are essential.

[0137] Furthermore, since the embodiments described above are for illustrative purposes of the technology described herein, various modifications, substitutions, additions, omissions, etc., can be made within the claims or their equivalents. [Industrial applicability]

[0138] This disclosure is applicable to analyzing data sequences showing the results of detecting various objects over time at various sites, and is applicable, for example, to analyzing the movement paths of moving objects.

Claims

1. A data analysis device that controls the generation of a data sequence representing a timeline showing the chronological order of the results of detecting objects at a site over time, A display unit that displays information, An input section for user input, The system includes a control unit that controls a data management unit that manages the detection results of the object and generates the data sequence, The control unit, Display information including the range where the ends of multiple timelines, each represented by multiple data columns, are adjacent to each other at the aforementioned site, is displayed on the display unit. The input unit receives a user operation to adjust parameters that indicate the conditions under which the multiple timelines are connected to each other in the adjacent range, The data management unit is controlled to apply the adjusted parameters in the adjacent range in accordance with the user operation in the input unit. Data analysis device.

2. The control unit, The input unit accepts user input specifying the requirements to which the adjusted parameters apply in the adjacent range. The data management unit is controlled to not apply the adjusted parameters if the specified requirements are not met, and to apply the adjusted parameters if the specified requirements are met. The data analysis apparatus according to claim 1.

3. The control unit detects the adjacent range as a candidate for a break in the timeline based on the data sequence generated by the data management unit, and notifies the detected range in the display information. The data analysis apparatus according to claim 1 or 2.

4. The control unit receives a user operation to specify the adjacent range in the display information via the input unit, Detailed information regarding the timeline whose end is located within the specified range is displayed on the display unit. The data analysis apparatus according to claim 1 or 2.

5. The system further includes a storage unit that stores pattern information including multiple patterns in which the object is not detected at the site, The control unit, The input unit receives a user operation to select one pattern from the multiple patterns indicated by the aforementioned pattern information. Depending on the selected pattern, information prompting the user to adjust the parameters will be displayed on the display unit. The data analysis apparatus according to claim 1 or 2.

6. The aforementioned data management unit, Based on the captured images of the aforementioned site, the detection results of the image recognition model that detects the object are managed. Based on the parameters, the detection results of the image recognition model are sequentially connected to generate the data sequence. The data analysis apparatus according to claim 1 or 2.

7. The system further comprises a communication unit that communicates data with the aforementioned data management unit, The control unit, The data sequence is received from the data management unit via the communication unit. The adjusted parameters are transmitted to the data management unit via the communication unit. The data analysis apparatus according to claim 1 or 2.

8. The aforementioned timeline shows the detection results of the on-site worker performing the prescribed task. The data analysis apparatus according to claim 1 or 2.

9. A data analysis method that controls the generation of a data column showing a timeline representing the chronological order of the results of detecting objects at a site over time, The aforementioned data sequence is generated by the data management unit that manages the detection results of the object, The steps include displaying information on a display unit that includes the range in which the ends of multiple timelines, each represented by multiple data columns, are adjacent to each other at the aforementioned site, and displaying this information on a display unit. The input unit receives a user operation to adjust parameters indicating the conditions under which the multiple timelines are connected to each other in the adjacent range. The steps include controlling the data management unit to apply the adjusted parameters in the adjacent range in accordance with the user operation in the input unit, and Data analysis methods including those mentioned above.

10. A program for causing a computer control unit to execute the data analysis method described in claim 9.

Images