Selection system, selection device, and selection method

Abstract

A selection system according to the present disclosure includes: an acquisition unit that acquires information regarding the operation status of an operation being performed by an operator; and a selection unit that selects an analysis unit for performing an analysis of data to be analyzed about the operation from among a first analysis unit and a second analysis unit according to the information regarding the operation status. The first analysis unit performs a first analysis of the data to be analyzed. The second analysis unit acquires the data to be analyzed via the first analysis unit, and performs a second analysis of the acquired data to be analyzed.

Description

Selection system, selection device and selection method

[0001] The present disclosure relates to a selection system, a selection device, and a selection method.

[0002] There are known techniques for monitoring the work of workers. For example, Patent Literature 1 (JP-A-2005-102666) describes a work support system that analyzes image data of a worker performing work to identify variations in the worker's movements during the work. Patent Literature 1 also describes that all or part of the functions of the work support system may be distributed across one or more computers, such as a cloud.

[0003] JP 2023-28120 A

[0004] However, simply distributing the functions of the system as in Patent Document 1 does not allow for efficient analysis of the work performed by the workers. For example, although the load of the analysis process may vary depending on the worker performing the work and the work content, simply distributing the load as in Patent Document 1 makes it difficult to appropriately select the execution destination of the analysis process.

[0005] In view of the above-described problems, an object of the present disclosure is to provide a selection system, a selection device, and a selection method that are capable of appropriately selecting the destination for executing analysis processing.

[0006] A selection system according to one aspect of the present disclosure includes an acquisition means for acquiring information regarding the work status of work performed by a worker, and a selection means for selecting, from first analysis processing means and second analysis processing means, an analysis processing means for performing analysis processing on data to be analyzed for the work, depending on the information regarding the work status, wherein the first analysis processing means performs a first analysis processing on the data to be analyzed, and the second analysis processing means acquires the data to be analyzed via the first analysis processing means and performs a second analysis processing on the acquired data to be analyzed.

[0007] A selection device according to one aspect of the present disclosure includes an acquisition means for acquiring information regarding the work status of work performed by a worker, and a selection means for selecting, from first analysis processing means and second analysis processing means, an analysis processing means for performing analysis processing on data to be analyzed for the work, depending on the information regarding the work status, wherein the first analysis processing means performs a first analysis processing on the data to be analyzed, and the second analysis processing means acquires the data to be analyzed via the first analysis processing means and performs a second analysis processing on the acquired data to be analyzed.

[0008] A selection method according to one aspect of the present disclosure includes acquiring information relating to the work status of work performed by a worker, and selecting, according to the information relating to the work status, an analysis processing means for performing analysis processing on data to be analyzed for the work from among first analysis processing means and second analysis processing means, wherein the first analysis processing means performs a first analysis processing on the data to be analyzed, and the second analysis processing means acquires the data to be analyzed via the first analysis processing means and performs a second analysis processing on the acquired data to be analyzed.

[0009] According to the present disclosure, it is possible to appropriately select the destination for executing the analysis process.

[0010] 1 is a configuration diagram showing an example of a configuration of a selection system according to an embodiment. FIG. 1 is a configuration diagram showing an example of a configuration of a selection device according to an embodiment. FIG. 2 is a flowchart showing an example of a selection method according to an embodiment. FIG. 2 is a configuration diagram showing an example of a configuration of a monitoring system according to an embodiment. FIG. 3 is a configuration diagram showing an example of a configuration of each device in the monitoring system according to an embodiment. FIG. 4 is a configuration diagram showing an example of a configuration of a hand movement analysis unit according to an embodiment. FIG. 5 is a configuration diagram showing an example of a configuration of a finger movement analysis unit according to an embodiment. FIG. 6 is a flowchart showing an example of operation of the monitoring system according to an embodiment. FIG. 7 is a diagram showing an example of a table used in the analysis destination selection process according to an embodiment. FIG. 8 is a diagram showing another example of a table used in the analysis destination selection process according to an embodiment. FIG. 9 is a diagram showing another example of a table used in the analysis destination selection process according to an embodiment. FIG. 10 is a configuration diagram showing an example of a configuration of each device in the monitoring system according to an embodiment. FIG. 11 is a flowchart showing an example of operation of the monitoring system according to an embodiment. FIG. 12 is a diagram showing an example of a table used in the analysis destination selection process according to an embodiment. FIG. 13 is a diagram showing another example of a table used in the analysis destination selection process according to an embodiment. FIG. 14 is a diagram showing another example of a table used in the analysis destination selection process according to an embodiment.

[0011] Hereinafter, embodiments will be described with reference to the drawings. In the drawings, the same elements are denoted by the same reference numerals, and redundant description will be omitted as necessary.

[0012] (Overview of the Embodiments) First, an overview of the embodiments will be described. Fig. 1 shows an example configuration of a selection system 10 according to some embodiments. The selection system 10 can be applied to a system that analyzes images when monitoring the work of workers on a production line, for example.

[0013] In the example of FIG. 1, the selection system 10 includes an acquisition unit 11 and a selection unit 12. The acquisition unit 11 acquires information about the work status of work performed by a worker. The information about the work status is information about the worker performing the work and the work status. For example, the information about the work status may include the status of the worker. The status of the worker indicates the worker's level of experience in the work. The status of the worker may be the worker's years of service, level of proficiency, frequency of errors, etc. Furthermore, the information about the work status may include the working time or work content of the work. The work content may be a work process associated with the work content.

[0014] The selection unit 12 selects an analysis processing unit that performs analysis processing on analysis target data of the work from the first analysis processing unit 13 and the second analysis processing unit 14, depending on the information about the work status acquired by the acquisition unit 11. For example, the analysis target data is an image captured by a camera installed on a production line. For example, when the status of a worker is acquired, the selection unit 12 may select the second analysis processing unit 14 if the acquired status of the worker is lower than a predetermined status, and may select the first analysis processing unit 13 if the acquired status of the worker is higher than the predetermined status. For example, when the work time is acquired, the selection unit 12 may select the second analysis processing unit 14 if the acquired work time is longer than a predetermined time, and may select the first analysis processing unit 13 if the acquired work time is shorter than the predetermined time.

[0015] The first analysis processing unit 13 and the second analysis processing unit 14 may be included in the selection system 10 or may be included in a system external to the selection system 10. The first analysis processing unit 13 performs a first analysis process on the analysis target data. The second analysis processing unit 14 performs a second analysis process on the analysis target data. The second analysis processing unit 14 acquires the analysis target data via the first analysis processing unit 13 and performs a second analysis process on the acquired analysis target data. The first analysis process is a process that simply analyzes tasks from the analysis target data and is a light-load process. The second analysis process is a process that analyzes tasks from the analysis target data in more detail than the first analysis process and is a heavier-load process than the first analysis process. For example, the first analysis process is a hand movement analysis process that analyzes the hand movements of a worker. The first analysis process may also be other processes such as person detection, skeletal estimation, object detection, and object anomaly detection. For example, the second analysis process is a finger movement analysis process that analyzes the finger movements of a worker. The second analysis process may also be other processes such as action recognition, skeletal structure estimation, segmentation, scene division, video summarization, etc. The first analysis process and the second analysis process are not limited to a combination of these processes. Either of the processes exemplified above as the first analysis process or the processes exemplified above as the second analysis process may be used as the first analysis process or the second analysis process. For example, the first analysis process may be scene division, and the second analysis process may be video summarization.

[0016] For example, the first analysis processing unit 13 may be located in a first device, and the second analysis processing unit 14 may be located in a second device. The first device has processing capabilities suitable for performing lightweight analysis, and the second device has processing capabilities suitable for performing heavy analysis. For example, the first device may be an edge device located at an edge, and the second device may be a cloud server located in a cloud. In an example of a system that analyzes video captured by a camera, the edge is a location relatively closer to the camera than the cloud, and the cloud is a location relatively farther from the camera than the edge. In an example of a system that provides services to a terminal, the edge is a location relatively closer to the terminal than the cloud, and the cloud is a location relatively farther from the terminal than the edge.

[0017] The selection system 10 may be configured by one device or by multiple devices. Fig. 2 shows an example configuration of a selection device 20 according to some embodiments. In the example of Fig. 2, the selection device 20 includes the acquisition unit 11 and the selection unit 12 shown in Fig. 1. For example, part or all of the selection system 10 and the selection device 20 may be located in an edge device or a cloud server.

[0018] 3 illustrates an example of a selection method according to some embodiments, which may be performed by, for example, the selection system 10 of FIG. 1 or the selection device 20 of FIG.

[0019] 3, the acquisition unit 11 acquires information about the work status of the work performed by the worker (S11). For example, the acquisition unit 11 acquires the status of the worker, the work time, etc. as information about the work status.

[0020] Next, the selection unit 12 selects an analysis processing unit that will analyze the analysis target data of the work, depending on the acquired information on the work status (S12). For example, the selection unit 12 selects either the first analysis processing unit 13 or the second analysis processing unit 14, depending on the status of the worker, the work time, etc.

[0021] As described above, in the embodiment, when analyzing data to be analyzed such as images, an analysis processing unit to perform the analysis is selected from a first analysis processing unit such as an edge processing unit and a second analysis processing unit such as a cloud processing unit, depending on information about the work situation, such as the worker's status and work time. This allows the appropriate destination for the analysis processing to be selected depending on the work situation, and the analysis processing can be distributed efficiently. This allows for efficient analysis and improved analysis accuracy.

[0022] (Embodiment 1) Next, a description will be given of embodiment 1. This embodiment is an example in which a device for analyzing an image is selected based on the status of a worker, the working time, and the like.

[0023] FIG. 4 shows an example configuration of a monitoring system 1 according to some embodiments. The monitoring system 1 is, for example, a system that monitors the work of workers using video captured by cameras installed on a production line. The production line is a line that manufactures target products through multiple work processes. For example, the target products may be, but are not limited to, home appliances such as televisions, electronic devices such as personal computers, electronic components such as semiconductor devices, mobile devices such as automobiles, processed foods, etc. Note that video includes multiple images (frames) in a time series, so the terms video and image are interchangeable. In other words, the monitoring system 1 can be considered both as a video processing system that processes video captured by cameras and as an image processing system that processes images captured by cameras.

[0024] 4, the monitoring system 1 includes multiple cameras 101, an edge device 100, a control server 200, and a management server 300. For example, the edge device 100 is located at a manufacturing site (edge) close to the camera 101. The control server 200 and the management server 300 are located at a remote location (cloud) away from the manufacturing site. Note that either or both of the control server 200 and the management server 300 may be located at the same manufacturing site as the edge device 100.

[0025] The cameras 101 and the edge device 100 are communicably connected via a network NW1. The network NW1 may be a wired network such as a local area network (LAN) or a wireless network such as a wireless LAN.

[0026] The edge device 100 and the management server 300 are communicatively connected via a network NW2. Like the network NW1, the network NW2 may be a wired network or a wireless network. Examples of wired networks include the Internet, a wide area network (WAN), and a LAN. Examples of wireless networks include 4G, local 5G / 5G, long term evolution (LTE), and a wireless LAN. Furthermore, the networks NW1 and NW2 may be the same network.

[0027] The edge device 100 and the control server 200 are communicatively connected via any network. The edge device 100 and the control server 200 may be connected via the network NW2 or the network NW1, or via another network.

[0028] Multiple cameras 101 are installed on the production line. Each camera 101 is installed so as to be able to capture images of each work process on the production line. If there are multiple production lines, a camera 101 is installed at each work process on each production line. A camera 101 may be installed not only for each work process but also for each production line. In this example, the camera 101 captures the work area of ​​each work process. For example, the camera 101 can capture an image of a workbench (work table) from above, capturing images of a worker working, the worker's hands, the parts being worked on, the tools used in the work, etc. The camera 101 may be a fixed camera fixed at a predetermined position, a wearable camera worn by a worker, or a camera equipped on a robot having an autonomous mobile mechanism. The camera 101 may capture an image of a work area of ​​one work process or of work areas of multiple work processes. In other words, the image captured by the camera 101 may include one work area or multiple work areas. The image captured by the camera 101 may be an image of a production line including a plurality of work areas. The camera 101 transmits the captured image including the work areas to the edge device 100 via the network NW1.

[0029] The edge device 100 is an edge processing device located on the edge side. For example, the edge device 100 may be a multi-access edge computing (MEC). For example, if a base station is located between the network NW1 and the network NW2, the edge device 100 may be connected to the base station. The edge device 100 is also a first analysis device that performs a first analysis process on the video from the camera 101. By performing the first analysis process at an edge close to the camera 101, the edge device 100 can obtain processing results faster than, for example, a cloud management server 300. For example, the edge device 100 may be a processing device with lower processing power than the management server 300. In this case, the first analysis process may be, for example, a light and simple analysis process that can be executed with low processing power. As an example, the first analysis process may be a hand movement analysis process that simply analyzes a task based on the hand movements of a worker. Note that the edge device 100 and the management server 300 may have approximately the same processing power. In this case, the first analysis process may be, for example, a process that requires the processing result to be obtained quickly. The edge device 100 also functions as a video transmission device that transmits the video from the camera 101 to the management server 300.

[0030] The edge device 100 receives an analysis instruction indicating a device (also referred to as an analysis destination device) that will analyze the video captured by the camera 101, and switches the analysis destination device between the edge device 100 or the cloud management server 300 in accordance with the received analysis instruction. For example, when the edge device 100 analyzes the video, the edge device 100 executes a hand movement analysis process on the video received from the camera 101. When the management server 300 analyzes the video, the edge device 100 transmits the video received from the camera 101 to the management server 300.

[0031] The control server 200 is a selection device that selects an analysis destination device that analyzes the video captured by the camera 101. The control server 200 is also a control device that controls switching of the analysis destination device depending on the selection result. The control server 200 may be one or more physical servers, or may be a virtualized server built on a virtualization platform. The control server 200 may be a cloud server on the cloud, or an on-premise server installed at the work site. The control server 200 and the management server 300 may be the same server.

[0032] The control server 200 selects an analysis destination device based on information about the work being performed in the video and information about the worker performing the work, and transmits the selection result to the edge device 100. It can also be said that the control server 200 selects a device that will execute analysis processing according to the work information and worker information. The control server 200 selects an analysis destination device for each video captured by the camera 101, and notifies the edge device 100 of an analysis instruction indicating the analysis destination device that will analyze the video. It is also possible to select an analysis destination device for each region in the video captured by the camera 101, and notify the edge device 100 of an analysis instruction indicating the analysis destination device that will analyze the region in the video.

[0033] The management server 300 is a monitoring device that monitors the work of each work process on the production line. Like the control server 200, the management server 300 may be one or more physical servers, or may be a virtualized server built on a virtualization platform. In this example, the management server 300 is a cloud server on the cloud, but it may also be an on-premise server installed at the work site. For example, the management server 300 is a GPU (Graphics Processing Unit) server that processes images.

[0034] The management server 300 is a second analysis device that performs a second analysis process on the video captured by the camera 101. For example, the management server 300 may be a processing device with higher processing power than the edge device 100. In this case, the second analysis process may be, for example, a heavy and detailed analysis process that can be executed with high processing power. As an example, the second analysis process may be a finger movement analysis process that analyzes the work in detail based on the finger movements of the worker. As described above, the edge device 100 and the management server 300 may have approximately the same processing power. In this case, the second analysis process may be, for example, a process that does not require rapid acquisition of processing results. The management server 300 also functions as a video receiving device that receives the video captured by the camera 101 from the edge device 100. For example, the management server 300 receives the video captured by the camera 101 from the edge device 100 and performs finger movement analysis on the received video. The management server 300 may output the analysis results of the finger movement analysis process to a display device or the like.

[0035] 5 shows an example configuration of each device in the monitoring system 1 according to some embodiments. Note that the configuration of each device is an example, and other configurations may be used as long as they are capable of executing the operation examples described below. Some of the functions of the control server 200 may be arranged in the edge device 100 or the management server 300. Some of the functions of the edge device 100 may be arranged in the control server 200 or the management server 300. Some of the functions of the management server 300 may be arranged in the edge device 100 or the control server 200.

[0036] In the example of FIG. 5 , the edge device 100 includes an image acquisition unit 110 , a detection unit 120 , an image switching unit 130 , a hand movement analysis unit 140 , and an image transmission unit 150 .

[0037] The video acquisition unit 110 acquires video captured by each of the multiple cameras 101 via the network NW1. The video captured by the cameras 101 is also referred to as work video. For example, the work video includes video of a work area for one work process, or video of work areas for multiple work processes. The video acquisition unit 110 also functions as an image acquisition unit that acquires multiple images, i.e., frames, in time series.

[0038] The detection unit 120 detects information that the control server 200 uses to select an analysis destination device for analyzing video. The detection unit 120 transmits the detected detection information to the control server 200. The detection unit 120 may detect necessary information from work video acquired from the camera 101, or may acquire necessary information from information input to an input device or the like. The detection unit 120 includes a notification unit that notifies the control server 200 of the detection information. The detection unit 120 may notify the control server 200 of the detection information via the network NW2, or may notify the control server 200 of the detection information via another communication path.

[0039] For example, the detection unit 120 may detect, as the detection information, the identification information of a worker performing work. The detection unit 120 may detect a person from a work video, perform face recognition on an image of the detected person, and detect the identification information of the authenticated worker. The detection unit 120 may acquire work schedule information, identify a worker currently performing work from the work schedule, and detect the worker's identification information. The detection unit 120 may acquire card information input into a card reader or the like when the worker performs work, and detect the worker's identification information from the card information.

[0040] Furthermore, the detection unit 120 may detect, as the detected information, the work time of the work performed by the worker. The work time is the time required for the work process that the worker is responsible for. For example, the detection unit 120 may analyze a work video and detect the work time based on the worker's movements, etc. The detection unit 120 may detect the work time from time information input by the worker when the work starts and ends. Furthermore, the detection unit 120 may detect the work process of the work performed by the worker. The detection unit 120 may analyze a work video and detect the work process based on the worker's movements, etc. The detection unit 120 may detect the work process from information on the work process input by the worker when the work starts and ends.

[0041] The video switching unit 130 switches the output destination (analysis destination) of the work video acquired from the camera 101. The video switching unit 130 includes an acquisition unit that acquires an analysis instruction from the control server 200. The video switching unit 130 receives an analysis instruction from the control server 200 indicating an analysis destination device to analyze the work video, and switches the output destination of the work video to either the hand movement analysis unit 140 or the video sending unit 150 in accordance with the received analysis instruction. Note that the video switching unit 130 may switch the output destination of an area in the work video when an analysis destination device to analyze the area in the work video is instructed.

[0042] The hand movement analysis unit 140 analyzes the hand movements of the worker from the work video output from the video switching unit 130. The hand movement analysis unit 140 is a first analysis processing unit that performs a first analysis process. For example, the hand movement analysis unit 140 detects the worker's hands in the work video and analyzes an overview of the worker's work in the work video. The hand movement analysis unit 140 may output the analysis results of the hand movement analysis performed on the work video to a display device or the like, or may send them to the management server 300.

[0043] 6 shows an example of the configuration of the hand movement analysis unit 140 according to some embodiments. In the example of FIG. 6, the hand movement analysis unit 140 includes an object detection unit 141 and an analysis unit 142.

[0044] The object detection unit 141 detects objects in the input work video. For example, the object detection unit 141 detects objects in a work area in the work video. Detecting an object includes identifying the type of object and detecting the position of the object. The object detection unit 141 extracts an object region, such as a rectangle containing an object, from each image included in the work video and recognizes the object type of the object in the extracted object region. The object detection unit 141 calculates image features of the object included in the object region and recognizes the object based on the calculated features. For example, the object detection unit 141 recognizes objects in an image using an object recognition engine that uses machine learning such as deep learning. Objects can be recognized by machine learning the image features and object type of the object. The object detection result includes the object type, position information of the object region containing the object, an object type score, and the like. The object position information is, for example, the coordinates of each vertex of the object region, but may also be the center position of the object region or the position of any point on the object. The object type score is the likelihood of the detected object type, i.e., reliability or confidence.

[0045] The object detection unit 141 determines whether the object types of objects detected in the work area include a person's hand, and if a hand is included, outputs a detection result of a hand. The object detection unit 141 may output a detection result of not only a person's hand, but also parts, tools, etc. used in work. Furthermore, the object detection unit 141 may detect a person (worker) performing work.

[0046] The analysis unit 142 analyzes the worker's work summary based on the hand detection results of the object detection unit 141. For example, the analysis unit 142 may detect a cycle pattern of work based on the detection results of the worker's hands in the work area of ​​the work video and output a cycle waveform indicating the cycle pattern as the analysis result. A cycle is a unit of a series of work performed repeatedly in the work area. The cycle waveform is a waveform that shows the hand detection results in chronological order and represents the work status of each cycle. For example, the hand detection results may indicate whether or not a hand is detected in the work area, or may indicate the coordinates of the hand within the camera's angle of view. For example, the cycle waveform may be a waveform obtained by converting hand position information between the start and end of work, or the time when the hand is within the work area, to 1 and other times to 0, and then taking a time average. The analysis unit 142 may calculate the cycle time (work time) from the start to the end of work in the cycle based on the cycle waveform and output the calculated cycle time of the work as the analysis result. The analysis result may include information detected by the object detection unit 141, such as hand detection results, detection results of parts and tools, and detection results of the worker.

[0047] 5 , the video transmission unit 150 transmits the video output from the video switching unit 130 to the management server 300 via the network NW2. The video transmission unit 150 is a communication interface capable of communication via the network NW2, and may be, for example, a wired interface such as a LAN or a WAN, or a wireless interface such as 4G, local 5G / 5G, LTE, or a wireless LAN. The video transmission unit 150 may transmit the analysis results of the hand movement analysis unit 140 to the management server 300.

[0048] In the example of FIG. 5 , the control server 200 includes a detection information acquisition unit 210 , a worker information acquisition unit 220 , and an analysis control unit 230 .

[0049] The detection information acquisition unit 210 acquires the detection information detected by the detection unit 120 from the edge device 100. The detection information acquisition unit 210 may acquire the detection information from the edge device 100 via the network NW2, or may acquire the detection information from the edge device 100 via another communication path. For example, the detection information acquisition unit 210 acquires detection information such as worker identification information, work time, and work process.

[0050] The worker information acquisition unit 220 acquires information about the characteristics of the worker performing the work. The worker information acquisition unit 220 may acquire information about the characteristics of the worker corresponding to the worker's identification information included in the detection information from a database in the control server 200 or from an external system. For example, the information about the characteristics of the worker is the worker's status. The worker's status may be the worker's years of service, proficiency level, or error frequency. The status may be a value such as years of service, or a level corresponding to years of service, etc. The status level may be rookie, veteran, etc.

[0051] The analysis control unit 230 is a selection unit that selects an analysis destination device that analyzes the video based on the acquired detection information and information on the worker's characteristics. It can also be said that the analysis control unit 230 selects a device that performs analysis processing according to the acquired detection information and information on the worker's characteristics. Furthermore, the analysis control unit 230 is a control unit that controls the edge device 100 to analyze the video using the selected analysis destination device. For example, the analysis control unit 230 may select an analysis destination device that analyzes the video based on the worker's status acquired by the worker information acquisition unit 220. The analysis control unit 230 may select an analysis destination device that analyzes the video based on the work time or work process included in the detection information acquired by the detection information acquisition unit 210. The analysis control unit 230 may also select an analysis destination device that analyzes an area within the video (e.g., an area including a worker).

[0052] The analysis control unit 230 notifies the edge device 100 of an analysis instruction indicating the selected analysis destination device. The analysis control unit 230 may notify the edge device 100 of the analysis instruction via the network NW2, or may notify the edge device 100 of the analysis instruction via another communication path.

[0053] In the example of FIG. 5, the management server 300 includes a video receiving unit 310 and a finger motion analyzing unit 320 .

[0054] The video receiving unit 310 receives the video transmitted from the edge device 100 via the network NW2. The video receiving unit 310 receives the video that the management server 300 has determined to be analyzed. The video receiving unit 310 may also receive the analysis results of the hand gesture from the edge device 100. Like the video transmitting unit 150, the video receiving unit 310 is a communication interface capable of communicating via the network NW2, and may be a wired interface or a wireless interface.

[0055] The finger movement analysis unit 320 analyzes the worker's finger movements from the received video. The finger movement analysis unit 320 is a second analysis processing unit that performs a second analysis process. For example, the finger movement analysis unit 320 inputs the received video into a trained finger movement recognition model and predicts the work content according to the finger movements in the video. The finger movement recognition model is a learning model that recognizes the worker's finger movements from the video. The finger movement recognition model is a learning model that can learn and infer the work content according to the worker's finger movements included in the video. The finger movement recognition model may be a convolutional neural network (CNN) or a recurrent neural network (RNN), or may be any other neural network. Furthermore, the finger movement analysis unit 320 is not limited to a neural network and may be any other learning model. The finger movement analysis unit 320 outputs the analysis results obtained by performing finger movement analysis on the video to a display device or the like.

[0056] 7 shows an example of the configuration of the finger motion analysis unit 320 according to some embodiments. In the example of Fig. 7, the finger motion analysis unit 320 includes a skeleton estimation unit 321, an object detection unit 322, and an inference unit 323.

[0057] The skeleton estimation unit 321 estimates the finger skeletons of a person in the input video. For example, the skeleton estimation unit 321 estimates the finger skeletons from each image included in the video using a skeleton estimation engine that uses machine learning such as deep learning. The skeleton estimation unit 321 detects feature points such as the joints of each finger in the image and estimates the skeleton of each finger connecting the feature points. The skeleton estimation unit 321 calculates feature amounts of the estimated finger skeletons and outputs the calculated feature amounts of the finger skeletons. The feature amounts of the finger skeletons are feature amounts corresponding to the shape and movement of the finger skeletons.

[0058] The object detection unit 322 detects objects in the input video. For example, the object detection unit 322 recognizes objects using an object recognition engine that uses machine learning such as deep learning. The object detection unit 322 extracts an object area, such as a rectangle, that includes an object from each image of the video and recognizes the object type of the object in the extracted object area. The object detection unit 322 determines whether the object type of the object detected around the person's finger includes a part or tool used for work, and if a specific part or tool is included, calculates feature amounts of the image including the part or tool and outputs the calculated feature amounts of the image.

[0059] The inference unit 323 infers the work content of the worker based on the skeleton estimation result of the skeleton estimation unit 321 and the object detection result of the object detection unit 322. For example, the inference unit 323 inputs the finger skeleton feature amount estimated by the skeleton estimation unit 321 and the feature amount of the image of the part or tool detected by the object detection unit 322 into a trained finger motion recognition model, and infers the work content corresponding to the finger motion based on the co-occurrence relationship between the finger skeleton feature amount and the image feature amount.

[0060] Fig. 8 shows an example of the operation of the monitoring system 1 according to some embodiments. In the example of Fig. 8, the edge device 100 acquires video from the camera 101 (S101). The camera 101 captures an image of the work area of ​​each work process and transmits the captured video to the edge device 100. The video acquisition unit 110 acquires the video transmitted from the camera 101, i.e., the work video, via the network NW1. For example, the work video may include the hands of a worker performing work in the work area, parts and tools being worked on, etc.

[0061] Next, the control server 200 acquires detection information to be used for selecting an analysis destination device (S102). For example, the detection unit 120 of the edge device 100 detects the worker's identification information, work time, work process, etc. from the acquired work video and input information, and notifies the control server 200 of the detected detection information. The detection information acquisition unit 210 of the control server 200 acquires the detection information from the edge device 100.

[0062] Next, the control server 200 acquires information about the characteristics of the worker (S103). For example, the worker information acquisition unit 220 acquires, from a database or the like, the status of the worker identified from the worker's identification information included in the acquired detection information.

[0063] Next, the control server 200 selects an analysis destination device for analyzing the work video (S104). The analysis control unit 230 selects an analysis destination device for analyzing the work video based on the acquired detection information and information related to the characteristics of the worker. For example, the analysis control unit 230 may use a table or the like to set in advance an analysis destination device for each worker's status, work time, or work process. Note that the example is not limited to using a table, and the analysis destination device may also be selected using a learning model or the like that has been machine-learned to identify analysis destination devices according to the worker's status, work time, and work process.

[0064] Fig. 9 shows an example of an analysis destination table for setting analysis destination devices. In the example of Fig. 9, an analysis destination device is associated with each worker's status. For example, when analyzing work video, the analysis control unit 230 acquires the worker's status and then refers to the analysis destination table in Fig. 9 to select an analysis destination device according to the worker's status.

[0065] For example, when the length of service is acquired as the worker's status, the analysis control unit 230 determines whether the worker is a new employee or a veteran based on the length of service. The analysis control unit 230 may compare the worker's length of service with a predetermined number of years of service. In this example, if the analysis control unit 230 determines that the worker is a new employee, the analysis control unit 230 selects the management server 300, which performs finger movement analysis, as the analysis destination device, because it is necessary to analyze in detail the work of new employees, who are more likely to make errors. Furthermore, if the analysis control unit 230 determines that the worker is a veteran, the analysis control unit 230 selects the edge device 100, which performs hand movement analysis, as the analysis destination device, because it is not necessary to analyze in detail the work of veteran employees, who are less likely to make errors.

[0066] For example, when the proficiency level is acquired as the worker's status, the analysis control unit 230 compares the worker's proficiency level with a predetermined proficiency level. In this example, if the worker's proficiency level is lower than the predetermined proficiency level, the analysis control unit 230 selects the management server 300, which performs finger movement analysis, as the analysis destination device, because the work of a worker with low proficiency needs to be analyzed in detail, just as with a newcomer. Also, if the worker's proficiency level is higher than the predetermined proficiency level, the analysis control unit 230 selects the edge device 100, which performs hand movement analysis, as the analysis destination device, because the work of a worker with high proficiency does not need to be analyzed in detail, just as with a veteran.

[0067] For example, when an error frequency is acquired as the worker's status, the analysis control unit 230 compares the worker's error frequency with a predetermined error frequency. In this example, if the worker's error frequency is higher than the predetermined error frequency, the analysis control unit 230 selects the management server 300, which performs finger movement analysis, as the analysis destination device, because the work of workers with a high error frequency needs to be analyzed in detail, as with newcomers. Furthermore, if the worker's error frequency is lower than the predetermined error frequency, the analysis control unit 230 selects the edge device 100, which performs hand movement analysis, as the analysis destination device, because the work of workers with a low error frequency does not need to be analyzed in detail, as with veterans.

[0068] Fig. 10 shows another example of an analysis destination table for setting analysis destination devices. In the example of Fig. 10, an analysis destination device is associated with each work time. For example, when analyzing work video, the analysis control unit 230 acquires the work time, refers to the analysis destination table in Fig. 10, and selects an analysis destination device according to the work time.

[0069] For example, the analysis control unit 230 compares the acquired work time with the normal work time. The normal work time may be the time designed for each work process when the production line was designed, or it may be the average work time, or it may be the work time for each time period. In this example, if the acquired work time is longer than the normal work time, the analysis control unit 230 selects the management server 300, which performs finger movement analysis, as the analysis destination device because there is a possibility of an error and the work needs to be analyzed in detail. As an example, if a work task that usually takes an average of 50 seconds to complete between 1:00 PM and 3:00 PM has taken 70 seconds as of 2:00 PM, the analysis control unit 230 selects the management server 300 as the analysis destination device. If a certain work process is completed in an average of 60 seconds but is currently taking 80 seconds, the analysis control unit 230 selects the management server 300 as the analysis destination device. Furthermore, if the acquired work time is the same as or shorter than the normal work time, there is a low possibility of an error and there is no need to analyze the work in detail, so the analysis control unit 230 selects the edge device 100, which performs hand movement analysis, as the analysis destination device.

[0070] 9 and 10, an analysis destination device may be selected according to a single condition such as the status of the worker or the working time, or multiple conditions may be combined to select an analysis destination device. Figures 11 and 12 show examples of analysis destination tables when multiple conditions are combined.

[0071] 11 shows another example of an analysis destination table for setting analysis destination devices. In the example of Fig. 11, an analysis destination device is associated with each combination of worker status and work time. For example, when analyzing work video, the analysis control unit 230 acquires the worker's status and work time, and then refers to the analysis destination table in Fig. 11 to select an analysis destination device based on the worker's status and work time.

[0072] In this example, if the worker is a new employee and the working time is longer than the normal working time, or if the worker is a new employee and the working time is the same as the normal working time, the analysis control unit 230 selects the management server 300 that performs finger movement analysis as the analysis destination device in order to perform a detailed analysis of the new employee's work that is likely to be error-prone.Also, if the worker is a new employee and the working time is shorter than the normal working time, the analysis control unit 230 selects the edge device 100 that performs hand movement analysis as the analysis destination device because even if the worker is a new employee, the possibility of error is low and detailed analysis is not necessary.

[0073] Furthermore, if the worker is an experienced worker and the working time is longer than the normal working time, the analysis control unit 230 selects the management server 300 that performs finger movement analysis as the analysis destination device, since even the work of an experienced worker is likely to be error-prone and a detailed analysis is necessary. Furthermore, if the worker is an experienced worker and the working time is the same as the normal working time, or if the worker is an experienced worker and the working time is shorter than the normal working time, the analysis control unit 230 selects the edge device 100 that performs hand movement analysis as the analysis destination device, since there is no need to analyze in detail the work of an experienced worker that is less likely to be error-prone.

[0074] 12 shows another example of an analysis destination table for setting an analysis destination device. In the example of Fig. 12, an analysis destination device is associated with each combination of worker identification information and work process. For example, when analyzing work video, the analysis control unit 230 acquires the worker identification information and work process, and then refers to the analysis destination table in Fig. 12 to select an analysis destination device according to the worker and work process.

[0075] In this example, when the worker is A and the work process is W1, for example, if worker A is not good at work process W1, there is a possibility of an error and the work needs to be analyzed in detail, so the analysis control unit 230 selects the management server 300 that performs finger movement analysis as the analysis destination device. Also, when the worker is A and the work process is W2 or W3, for example, if worker A is good at work processes W2 and W3, there is a low possibility of an error and there is no need to analyze the work in detail, so the analysis control unit 230 selects the edge device 100 that performs hand movement analysis as the analysis destination device.

[0076] Furthermore, when the worker is B and the work process is W1, for example, if worker B is good at work process W1, the possibility of an error is low and there is no need to analyze the work in detail, so the analysis control unit 230 selects the edge device 100 that performs hand movement analysis as the analysis destination device. Furthermore, when the worker is B and the work process is W2 or W3, for example, if worker B is bad at work processes W2 and W3, the possibility of an error is high and there is no need to analyze the work in detail, so the analysis control unit 230 selects the management server 300 that performs finger movement analysis as the analysis destination device.

[0077] Next, the control server 200 controls to switch the analysis destination device that will analyze the work video (S105). When the analysis control unit 230 of the control server 200 selects the analysis destination device that will analyze the work video, it notifies the edge device 100 of an analysis instruction indicating the selected analysis destination device. The video switching unit 130 of the edge device 100 receives the analysis instruction indicating the analysis destination device from the control server 200, and switches the output destination of the work video to either the hand movement analysis unit 140 or the video sending unit 150 in accordance with the received analysis instruction.

[0078] In response to the switching in S105, the edge device 100 or the management server 300 analyzes the work video. For example, when the edge device 100 is designated as the analysis destination device, the video switching unit 130 switches the output destination of the work video acquired from the camera 101 to the hand movement analysis unit 140. Then, the hand movement analysis unit 140 analyzes the worker's hand movements based on the work video output from the video switching unit 130 (S106). For example, the object detection unit 141 of the hand movement analysis unit 140 uses an object recognition engine to detect object regions, such as rectangles, within the image included in the work video and recognize the object type of the object within the detected object region. The object detection unit 141 outputs detection results for objects detected in the work area of ​​the work video whose object type includes a person's hand. The analysis unit 142 of the hand movement analysis unit 140 generates cycle waveforms, cycle times, etc. based on the detection results of the worker's hands in the work area of ​​the work video and outputs the generated cycle waveforms, cycle times, etc. as analysis results.

[0079] Furthermore, when the management server 300 is designated as the analysis destination device, the video switching unit 130 switches the output destination of the work video acquired from the camera 101 to the video transmission unit 150. Then, the video transmission unit 150 transmits the work video output from the video switching unit 130 to the management server 300 (S107).

[0080] Next, the management server 300 receives the video from the edge device 100 (S108). The video receiving unit 310 receives the video from the edge device 100 via the network NW2.

[0081] Next, the management server 300 analyzes the worker's finger movements based on the received video (S109). For example, the skeleton estimation unit 321 of the finger movement analysis unit 320 uses a skeleton estimation engine to estimate the finger skeletons in the received video and output feature quantities of the estimated finger skeletons. The object detection unit 322 of the finger movement analysis unit 320 uses an object recognition engine to recognize objects in the received video and output feature quantities of images including parts, tools, etc. used in the work. The inference unit 323 of the finger movement analysis unit 320 uses a trained finger movement recognition model to infer the work content corresponding to the finger movements based on the feature quantities of the finger skeletons and the feature quantities of the image. The finger movement analysis unit 320 outputs the analyzed work content as the analysis result.

[0082] As described above, in this embodiment, in a system including an edge device and a cloud-side management server, when analyzing video of work, either the edge device or the management server is selected as the analysis destination device for analyzing the video depending on the worker's status, work time, etc. This allows the video analysis to be performed on an appropriate device depending on the worker's status, work time, etc., and allows the video to be analyzed efficiently.

[0083] (Embodiment 2) Next, embodiment 2 will be described. This embodiment is an example in which priority is determined based on the status of the worker, the working time, etc., and a device for analyzing images is selected according to the priority. Note that this embodiment can be implemented in combination with embodiment 1, and each configuration shown in embodiment 1 may be used as appropriate.

[0084] 13 shows an example of the configuration of each device in a monitoring system 1 according to some embodiments. In the example of Fig. 13, the control server 200 includes a priority determination unit 240 in addition to the configuration of Fig. 5.

[0085] The priority determination unit 240 determines the priority based on the detection information acquired by the detection information acquisition unit 210 and the information on the worker's characteristics acquired by the worker information acquisition unit 220. The priority is the priority for performing analysis processing on the video. Video with a higher priority is subjected to heavier and more detailed analysis processing preferentially than video with a lower priority. For example, when video is analyzed by either the edge device 100 or the management server 300, video with a higher priority is analyzed preferentially by the management server 300. Furthermore, when multiple videos are analyzed by the same device, video with a higher priority may be analyzed before video with a lower priority. For example, the priority determination unit 240 may determine the priority based on the status of the acquired worker. The priority determination unit 240 may determine the priority based on the work time or work process included in the acquired detection information.

[0086] Furthermore, the analysis control unit 230 selects an analysis destination device to analyze the video according to the priority determined by the priority determination unit 240. For example, the analysis control unit 230 may select the management server 300 when the priority is high, and may select the edge device 100 when the priority is low. The other configurations are the same as those in the example described in the first embodiment.

[0087] Fig. 14 shows an example of the operation of the monitoring system 1 according to some embodiments. In the example of Fig. 14, similar to Fig. 8, the edge device 100 acquires video from the camera 101 (S101). The control server 200 acquires detection information used to select an analysis destination device (S102), and acquires information on the characteristics of the worker (S103).

[0088] Next, the control server 200 determines the priority of the video to be analyzed (S110). The priority determination unit 240 determines the priority of the video based on the acquired detection information and information related to the worker's characteristics. For example, the priority determination unit 240 may use a table or the like to pre-set the worker's status, work time, and priority for each work process, similar to the analysis control unit 230 described in embodiment 1. Note that the example using a table is not limited to this, and priorities may also be determined using a learning model or the like that machine-learns priorities according to the worker's status, work time, and work process. Figures 15 to 18 show examples in which the analysis destination tables shown in Figures 9 to 12 are applied to priority tables. The determination conditions for the priority tables in Figures 15 to 18 are the same as those for the analysis destination tables in Figures 9 to 12. Note that in this example, priority is divided into two levels, high priority and low priority, but multiple levels may also be used.

[0089] Fig. 15 shows an example of a priority table for setting priorities. In the example of Fig. 15, a priority is associated with each worker's status. For example, when the priority determination unit 240 acquires the worker's status, it refers to the priority table of Fig. 15 and determines the priority according to the worker's status.

[0090] For example, when the length of service is acquired as the worker's status and it is determined that the worker is a new employee based on the length of service, the priority determination unit 240 determines the priority to be high because a detailed analysis is required as a priority. On the other hand, when it is determined that the worker is a veteran based on the length of service, it is not necessary to perform a detailed analysis as a priority, and the priority determination unit 240 determines the priority to be low.

[0091] For example, when the proficiency level of the worker is acquired as the worker's status and the proficiency level of the worker is lower than a predetermined proficiency level, the priority determination unit 240 determines the priority to be high because a detailed analysis is required as a priority. On the other hand, when the proficiency level of the worker is higher than the predetermined proficiency level, the priority determination unit 240 determines the priority to be low because a detailed analysis is not required as a priority.

[0092] For example, if the error frequency of the worker is acquired as the worker's status and the error frequency of the worker is higher than a predetermined error frequency, the priority determination unit 240 determines the priority to be high because a detailed analysis is required as a priority.Furthermore, if the error frequency of the worker is lower than the predetermined error frequency, the priority determination unit 240 determines the priority to be low because a detailed analysis is not required as a priority.

[0093] Fig. 16 shows another example of a priority table for setting priorities. In the example of Fig. 16, a priority is associated with each work time. For example, when the priority determination unit 240 acquires the work time, it refers to the priority table of Fig. 16 and determines the priority according to the work time.

[0094] For example, if the acquired work time is longer than the normal work time, the priority determination unit 240 determines the priority to be high because a detailed analysis is required as a priority. On the other hand, if the acquired work time is the same as or shorter than the normal work time, the priority determination unit 240 determines the priority to be low because a detailed analysis is not required as a priority.

[0095] 17 shows another example of a priority table for setting priorities. In the example of Fig. 17, a priority is associated with each combination of worker status and working time. For example, upon acquiring the worker status and working time, the priority determination unit 240 refers to the priority table in Fig. 17 and determines a priority based on the worker status and working time.

[0096] For example, if the worker is a new employee and the working time is longer than the normal working time, or if the worker is a new employee and the working time is the same as the normal working time, the priority determination unit 240 determines the priority to be high because a detailed analysis needs to be prioritized.Also, if the worker is a new employee and the working time is shorter than the normal working time, the priority determination unit 240 determines the priority to be low because a detailed analysis does not need to be prioritized.

[0097] Furthermore, if the worker is an experienced worker and the working time is longer than the normal working time, the priority determination unit 240 determines the priority to be high because a detailed analysis is required as a priority. Furthermore, if the worker is an experienced worker and the working time is the same as the normal working time, or if the worker is an experienced worker and the working time is shorter than the normal working time, the priority determination unit 240 determines the priority to be low because a detailed analysis is not required as a priority.

[0098] Fig. 18 shows another example of a priority table for setting priorities. In the example of Fig. 18, a priority is associated with each combination of worker identification information and work process. For example, when the priority determination unit 240 acquires the worker identification information and work process, it references the priority table in Fig. 18 and determines a priority according to the worker and work process.

[0099] For example, when the worker is A and the work process is W1, the priority determination unit 240 determines the priority to be high because a detailed analysis is required as a priority. On the other hand, when the worker is A and the work process is W2 or W3, the priority determination unit 240 determines the priority to be low because a detailed analysis is not required as a priority.

[0100] Furthermore, when the worker is B and the work process is W1, the priority determination unit 240 determines the priority to be low because there is no need to give priority to detailed analysis. Furthermore, when the worker is B and the work process is W2 or W3, the priority determination unit 240 determines the priority to be high because there is a need to give priority to detailed analysis.

[0101] Next, the control server 200 selects an analysis destination device for analyzing the work video (S104). The analysis control unit 230 selects an analysis destination device for analyzing the work video according to the determined priority. For example, the analysis control unit 230 selects the management server 300 if the priority is high, and selects the edge device 100 if the priority is low. The analysis control unit 230 may select the management server 300 if the priority is higher than a predetermined priority, and select the edge device 100 if the priority is lower than the predetermined priority.

[0102] 8 , the control server 200 controls to switch the analysis destination device that analyzes the work video (S105), and when the edge device 100 is controlled to be the analysis destination device, the edge device 100 performs hand movement analysis (S106). When performing hand movement analysis on multiple videos, the hand movement analysis unit 140 may perform the hand movement analysis in order of priority.

[0103] Furthermore, when the management server 300 is controlled to be the analysis destination device, the edge device 100 transmits video (S107), and the management server 300 receives the video (S108) and performs finger movement analysis from the received video (S109). The video transmission unit 150 may control the image quality of the video to be transmitted according to priority. For example, video with a low priority, such as that of an expert, may require lower analysis accuracy, so the edge device 100 may transmit video with a lower image quality by lowering the frame rate or bit rate to the management server 300. Furthermore, when performing finger movement analysis on multiple videos, the finger movement analysis unit 320 may perform the finger movement analysis in order of priority.

[0104] As described above, in this embodiment, priorities are determined according to the status and working time of the worker, and an analysis destination device for analyzing the video is selected according to the priority. This allows the video analysis to be performed on an appropriate device according to the priority, making it possible to analyze the video efficiently. Furthermore, when analyzing multiple videos, videos with higher priorities can be analyzed preferentially.

[0105] The present disclosure is not limited to the above-described embodiment and may be modified as appropriate without departing from the spirit of the present disclosure. For example, in the above-described embodiment, the analysis destination device is selected based on priority, but the transmission of video from the edge device to the management server may also be controlled based on priority. For example, when multiple videos are transmitted from the edge device to the management server, videos with low priority, such as those of veterans, do not need to be transmitted and analyzed immediately, and therefore may be transmitted from the edge device to the management server at a different time as a batch process rather than a real-time process.

[0106] Each component in the above-described embodiments may be configured with hardware or software, or both, and may be configured with a single piece of hardware or software, or may be configured with multiple pieces of hardware or software. Each device and each function (processing) such as the edge device, control server, and management server may be realized by a computer 30 having a processor 31 such as a CPU (Central Processing Unit) and a memory 32 serving as a storage device, as shown in FIG. 19 . For example, a program for performing the method (selection method) in the embodiment may be stored in the memory 32, and each function may be realized by the processor 31 executing the program stored in the memory 32.

[0107] These programs include instructions (or software code) that, when loaded into a computer, cause the computer to perform one or more functions described in the embodiments. The programs may be stored on a non-transitory computer-readable medium or a tangible storage medium. By way of example and not limitation, computer-readable media or tangible storage media include random-access memory (RAM), read-only memory (ROM), flash memory, solid-state drive (SSD) or other memory technology, CD-ROM, digital versatile disc (DVD), Blu-ray disc or other optical disk storage, magnetic cassette, magnetic tape, magnetic disk storage or other magnetic storage device. The programs may also be transmitted on a transitory computer-readable medium or communication medium. By way of example and not limitation, transitory computer-readable media or communication media include electrical, optical, acoustic, or other forms of propagated signals.

[0108] Although the present disclosure has been described above with reference to the embodiments, the present disclosure is not limited to the above-described embodiments. Various modifications that can be understood by those skilled in the art can be made to the configuration and details of the present disclosure within the scope of the present disclosure. Furthermore, each embodiment can be combined with other embodiments as appropriate.

[0109] Each drawing is merely an example for describing one or more embodiments. Each drawing may not relate to only one particular embodiment, but may also relate to one or more other embodiments. As will be understood by those skilled in the art, various features or steps described with reference to any one drawing can be combined with features or steps shown in one or more other drawings to create, for example, an embodiment not explicitly shown or described. Not all features or steps shown in any one drawing are necessary to describe an exemplary embodiment, and some features or steps may be omitted. The order of steps described in any drawing may be changed as appropriate.

[0110] A part or all of the above-described embodiments can be described as, but not limited to, the following supplementary notes.

[0111] (Supplementary Note 1) A selection system comprising: an acquisition means for acquiring information regarding the work status of work performed by a worker; and a selection means for selecting, from a first analysis processing means and a second analysis processing means, an analysis processing means for performing analysis processing on analysis target data of the work according to the information regarding the work status, wherein the first analysis processing means performs a first analysis processing on the analysis target data, and the second analysis processing means acquires the analysis target data via the first analysis processing means and performs a second analysis processing on the acquired analysis target data. (Supplementary Note 2) The selection system according to Supplementary Note 1, wherein the information regarding the work status includes the status of the worker. (Supplementary Note 3) The selection system according to Supplementary Note 2, wherein the selection means selects the second analysis processing means if the status of the worker is lower than a predetermined status, and selects the first analysis processing means if the status of the worker is higher than the predetermined status. (Supplementary Note 4) The selection system according to any one of Supplements 1 to 3, wherein the information regarding the work status includes a working time or a work content of the work. (Supplementary Note 5) The selection system according to Supplementary Note 4, wherein the selection means selects the second analysis processing means when the work time is longer than a predetermined time, and selects the first analysis processing means when the work time is shorter than the predetermined time. (Supplementary Note 6) The selection system according to any one of Supplements 1 to 4, wherein the selection means selects either the first analysis processing means or the second analysis processing means according to a priority based on information relating to the work status. (Supplementary Note 7) The selection system according to Supplementary Note 6, wherein the analysis target data is an image of the work, and the first analysis processing means transmits the image, the image quality of which has been controlled according to the priority, to the second analysis processing means. (Supplementary Note 8) The selection system according to any one of Supplements 1 to 7, wherein the first analysis processing means is included in an edge device located at an edge, and the second analysis processing means is included in a cloud server located in a cloud.(Supplementary Note 9) A selection device comprising: an acquisition means for acquiring information regarding the work status of work performed by a worker; and a selection means for selecting, according to the information regarding the work status, an analysis processing means for performing analysis processing on analysis target data of the work from a first analysis processing means and a second analysis processing means, wherein the first analysis processing means performs a first analysis processing on the analysis target data, and the second analysis processing means acquires the analysis target data via the first analysis processing means and performs a second analysis processing on the acquired analysis target data. (Supplementary Note 10) The selection device according to Supplementary Note 9, wherein the information regarding the work status includes the status of the worker. (Supplementary Note 11) The selection device according to Supplementary Note 10, wherein the selection means selects the second analysis processing means when the status of the worker is lower than a predetermined status, and selects the first analysis processing means when the status of the worker is higher than the predetermined status. (Supplementary Note 12) The selection device according to any one of Supplements 9 to 11, wherein the information relating to the work status includes a work time or a work content of the work. (Supplementary Note 13) The selection device according to Supplementary Note 12, wherein the selection means selects the second analysis processing means when the work time is longer than a predetermined time, and selects the first analysis processing means when the work time is shorter than the predetermined time. (Supplementary Note 14) The selection device according to any one of Supplements 9 to 13, wherein the selection means selects either the first analysis processing means or the second analysis processing means according to a priority based on the information relating to the work status. (Supplementary Note 15) A selection method comprising: acquiring information relating to the work status of work performed by a worker; and selecting, according to the information about the work status, an analysis processing means that performs analysis processing on analysis target data of the work from first analysis processing means and second analysis processing means, wherein the first analysis processing means performs first analysis processing on the analysis target data, and the second analysis processing means acquires the analysis target data via the first analysis processing means and performs second analysis processing on the acquired analysis target data. (Supplementary Note 16) The selection method according to Supplementary Note 15, wherein the information about the work status includes the status of the worker.(Supplementary Note 17) The selection method according to Supplementary Note 16, wherein the selecting comprises selecting the second analysis processing means if the status of the worker is lower than a predetermined status, and selecting the first analysis processing means if the status of the worker is higher than the predetermined status. (Supplementary Note 18) The selection method according to any one of Supplements 15 to 17, wherein the information related to the work status includes a work time or work content of the work. (Supplementary Note 19) The selection method according to Supplementary Note 18, wherein the selecting comprises selecting the second analysis processing means if the work time is longer than a predetermined time, and selecting the first analysis processing means if the work time is shorter than the predetermined time. (Supplementary Note 20) The selection method according to any one of Supplements 15 to 19, wherein the selecting comprises selecting either the first analysis processing means or the second analysis processing means according to a priority based on the information related to the work status.

[0112] REFERENCE SIGNS LIST 1 Monitoring system 10 Selection system 11 Acquisition unit 12 Selection unit 13 First analysis processing unit 14 Second analysis processing unit 20 Selection device 30 Computer 31 Processor 32 Memory 100 Edge device 101 Camera 110 Video acquisition unit 120 Detection unit 130 Video switching unit 140 Hand movement analysis unit 141 Object detection unit 142 Analysis unit 150 Video transmission unit 200 Control server 210 Detection information acquisition unit 220 Worker information acquisition unit 230 Analysis control unit 240 Priority determination unit 300 Management server 310 Video reception unit 320 Finger movement analysis unit 321 Skeleton estimation unit 322 Object detection unit 323 Inference unit

Claims

1. A selection system comprising: an acquisition means for acquiring information regarding the status of work performed by a worker; and a selection means for selecting, in accordance with the information regarding the work status, an analysis processing means for performing analysis processing on data to be analyzed for the work from a first analysis processing means and a second analysis processing means, wherein the first analysis processing means performs a first analysis processing on the data to be analyzed; and the second analysis processing means acquires the data to be analyzed via the first analysis processing means and performs a second analysis processing on the acquired data to be analyzed.

2. The selection system according to claim 1, wherein the information relating to the work situation includes the status of the worker.

3. The selection system according to claim 2, wherein the selection means selects the second analysis processing means when the worker's status is lower than a predetermined status, and selects the first analysis processing means when the worker's status is higher than the predetermined status.

4. The selection system according to any one of claims 1 to 3, wherein the information relating to the work status includes a work time or a work content of the work.

5. The selection system according to claim 4, wherein the selection means selects the second analysis processing means when the work time is longer than a predetermined time, and selects the first analysis processing means when the work time is shorter than the predetermined time.

6. The selection system according to any one of claims 1 to 4, wherein the selection means selects either the first analysis processing means or the second analysis processing means according to a priority based on information relating to the work status.

7. The selection system according to claim 6, wherein the data to be analyzed is an image of the work, and the first analysis processing means transmits the image, the image quality of which has been controlled according to the priority, to the second analysis processing means.

8. A selection system according to any one of claims 1 to 7, wherein the first analysis processing means is included in an edge device located at an edge, and the second analysis processing means is included in a cloud server located in a cloud.

9. A selection device comprising: an acquisition means for acquiring information regarding the status of work performed by a worker; and a selection means for selecting, in accordance with the information regarding the work status, an analysis processing means for performing analysis processing on data to be analyzed of the work from a first analysis processing means and a second analysis processing means, wherein the first analysis processing means performs a first analysis processing on the data to be analyzed, and the second analysis processing means acquires the data to be analyzed via the first analysis processing means and performs a second analysis processing on the acquired data to be analyzed.

10. The selection device according to claim 9, wherein the information relating to the work situation includes the status of the worker.

11. The selection device according to claim 10, wherein the selection means selects the second analysis processing means when the worker's status is lower than a predetermined status, and selects the first analysis processing means when the worker's status is higher than the predetermined status.

12. The selection device according to any one of claims 9 to 11, wherein the information relating to the work status includes a work time or a work content of the work.

13. The selection device according to claim 12, wherein the selection means selects the second analysis processing means when the operation time is longer than a predetermined time, and selects the first analysis processing means when the operation time is shorter than the predetermined time.

14. The selection device according to any one of claims 9 to 13, wherein the selection means selects either the first analysis processing means or the second analysis processing means according to a priority based on information relating to the work status.

15. A selection method comprising: acquiring information regarding the status of work performed by a worker; and selecting, according to the information regarding the work status, an analysis processing means for performing analysis processing on data to be analyzed of the work from among first analysis processing means and second analysis processing means, wherein the first analysis processing means performs a first analysis processing on the data to be analyzed; and the second analysis processing means acquires the data to be analyzed via the first analysis processing means and performs a second analysis processing on the acquired data to be analyzed.

16. The selection method according to claim 15, wherein the information relating to the work situation includes the status of the worker.

17. The selection method according to claim 16, wherein the selection includes selecting the second analysis processing means when the worker's status is lower than a predetermined status, and selecting the first analysis processing means when the worker's status is higher than the predetermined status.

18. The selection method according to any one of claims 15 to 17, wherein the information relating to the work status includes a work time or a work content of the work.

19. The selection method according to claim 18, wherein the selection includes selecting the second analysis processing means if the work time is longer than a predetermined time, and selecting the first analysis processing means if the work time is shorter than the predetermined time.

20. A selection method according to any one of claims 15 to 19, wherein the selection includes selecting either the first analysis processing means or the second analysis processing means according to a priority based on information relating to the work status.

Images