Control system, control device, and control method

Abstract

This control system comprises an acquisition unit, an identification unit, and a control unit. The acquisition unit acquires a processing result of first application processing performed on a first video in which an object is imaged. On the basis of the processing result of the first application processing acquired by the acquisition unit, the identification unit identifies second application processing to be performed on a second video in which the object is imaged. The control unit controls acquisition of the second video according to a condition of the second video for executing the second application processing identified by the identification unit.

Description

Control system, control device and control method

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

[0002] There is known a technology for controlling a camera by performing predetermined processing on an image captured by the camera. For example, Patent Document 1 describes an automatic tracking and photography system that recognizes a subject from an image captured by a first camera and controls the photographing optical axis and photographing angle of view of a second camera based on the predicted position of the subject.

[0003] Japanese Patent Application Laid-Open No. 2017-204795

[0004] For example, in a monitoring system that monitors work using video footage of workers performing their tasks, various processes are performed on the video footage, and the video footage required varies depending on the process being performed. However, as in Patent Document 1, simply controlling the camera based on the subject recognized from the image makes it difficult to properly acquire the video footage required for the next process.

[0005] In view of the above problems, one object of the present disclosure is to provide a control system, a control device, and a control method that are capable of acquiring appropriate images.

[0006] A control system according to one aspect of the present disclosure includes an acquisition means for acquiring a processing result of a first process performed on a first image of an object, an identification means for identifying a second process to be performed on a second image of the object based on the processing result, and a control means for controlling the acquisition of the second image according to the conditions of the second image for performing the second process.

[0007] A control device according to one aspect of the present disclosure includes an acquisition means for acquiring a processing result of a first processing performed on a first image of an object, an identification means for identifying a second processing to be performed on a second image of the object based on the processing result, and a control means for controlling the acquisition of the second image according to the conditions of the second image for performing the second processing.

[0008] A control method according to one aspect of the present disclosure acquires a processing result of a first processing performed on a first image capturing an object, identifies a second processing to be performed on a second image capturing the object based on the processing result, and controls acquisition of the second image according to the conditions of the second image for performing the second processing.

[0009] According to the present disclosure, it is possible to acquire appropriate images.

[0010] 1 is a configuration diagram showing an example of a configuration of a control system according to an embodiment. FIG. 1 is a configuration diagram showing an example of a configuration of a control device according to an embodiment. FIG. 2 is a flowchart showing an example of a control 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 diagram showing an example of an image captured by a camera according to an embodiment. FIG. 4 is a diagram showing an example of an image captured by a camera according to an embodiment. FIG. 4 is a configuration diagram showing an example of a monitoring server according to an embodiment. FIG. 5 is a flowchart showing an example of an operation of the monitoring server according to an embodiment. FIG. 6 is a diagram showing an example of a table used in the attention item identification process according to an embodiment. FIG. 7 is a diagram showing another example of a table used in the attention item identification process according to an embodiment. FIG. 8 is a diagram showing an example of a table used in the video condition identification process according to an embodiment. FIG. 9 is a configuration diagram showing an example of a configuration of a monitoring server according to an embodiment. FIG. 10 is a diagram showing an example of a table used in the attention item identification process according to an embodiment. FIG. 11 is a diagram showing an example of a table used in the video condition identification process according to an embodiment. FIG. 12 is a configuration diagram showing an example of a configuration of a monitoring system according to an embodiment. FIG. 13 is a flowchart showing an example of an operation of the monitoring system according to an embodiment. FIG. 14 is a configuration diagram showing an example of a configuration of computer hardware 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) In manufacturing sites, a technology is known that automates the monitoring of work by analyzing video footage of workers' work captured by a camera. Such automated monitoring requires manual configuration of appropriate settings tailored to the monitoring content, etc. When monitoring is performed manually, it is possible to select check items according to the content to be monitored and switch cameras to view those items. However, such flexibility is difficult to achieve when automating monitoring. For example, when a work process shifts from a first component assembly process to a second component assembly process, it is necessary to switch to a different camera to match the check items for each process. Furthermore, if the work to be monitored is in a camera's blind spot, it is necessary to change the camera's angle or field of view. In the embodiments, even in such cases, it is possible to appropriately control the video captured by the camera without manual configuration.

[0013] 1 shows an example of the configuration of a control system 10 according to some embodiments. The control system 10 can be used, for example, as a system for capturing images of work by workers on a production line using a camera and processing the images.

[0014] In the example of FIG. 1 , the control system 10 includes an acquisition unit 11, an identification unit 12, and a control unit 13. The acquisition unit 11 acquires a processing result of a first process performed on a first video of an object. The object is an object captured by a camera, or an object monitored using the captured video. For example, the object is a worker performing work. For example, the first process is a process of analyzing work performed by a worker. The first process may be a process of simply analyzing work from video, or may be a process with a low load. For example, the first process may be a hand movement detection process that detects the hand movement of a worker. The first process may also be other processes such as person detection, skeleton estimation, object detection, and object abnormality detection.

[0015] The identification unit 12 identifies a second process to be performed on a second video of the target based on the acquired processing result of the first process. The second video may be captured by the same camera as the camera that captured the first video, or by a different camera. The second process may be a process that analyzes the work from the video in more detail than the first process and may be a process with a heavier load than the first process. For example, the second process may be a finger movement detection process that detects the movement of the worker's fingers, a posture detection process, or a gaze detection process. The second process may also be other processes such as behavior recognition, skeletal structure estimation, segmentation, scene division, and video summarization.

[0016] The control system 10 may include a first processing unit that performs a first process and a second processing unit that performs a second process. The first processing unit may acquire an image and transmit the acquired image to the second processing unit. For example, the first processing unit may be located in a first device, and the second processing unit may be located in a second device. The first device may have processing capabilities suitable for performing lightweight analysis, and the second device may have 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.

[0017] The control unit 13 controls the acquisition of the second image according to the second image conditions for performing the second process. The second image conditions are image conditions necessary for performing the second process. The control unit 13 may control the camera according to conditions for specifying the camera that will capture the second image as the second image conditions. For example, the control unit 13 may specify an image of a camera that captures the worker's hands, upper body, etc. In this case, the control unit 13 may instruct the specified camera to capture an image and acquire the image captured by the camera in accordance with the instruction. The control unit 13 may also control the camera according to the shooting conditions of the camera that will capture the second image as the second image conditions. For example, the shooting conditions may be the camera angle, brightness, zoom, angle of view, etc. In this case, the control unit 13 may instruct the camera to set shooting conditions such as the camera angle, brightness, zoom, angle of view, etc., and acquire the image captured by the camera for which the shooting conditions have been set in accordance with the instruction.

[0018] The control system 10 may be configured with one device or multiple devices. Fig. 2 shows an example configuration of a control device 20 according to some embodiments. In the example of Fig. 2, the control device 20 includes the acquisition unit 11, the identification unit 12, and the control unit 13 shown in Fig. 1. For example, part or all of the control system 10 or the control device 20 may be arranged in an edge device arranged at an edge or in a cloud server arranged in a cloud.

[0019] 3 shows an example of a control method according to some embodiments. For example, the control method according to some embodiments may be performed by the control system 10 of FIG. 1 or the control device 20 of FIG.

[0020] 3, the acquisition unit 11 acquires a processing result of a first process performed on a first video of an object (S11). For example, the acquisition unit 11 acquires a processing result of a hand gesture detection process or the like performed on a first video of an object captured by a camera.

[0021] Next, the identifying unit 12 identifies a second process to be performed on the second video of the target based on the processing result of the acquired first process (S12). For example, the identifying unit 12 identifies a finger motion detection process or the like that analyzes in detail the second video of the worker or the like captured by the camera based on the processing result of the hand motion detection process or the like as the second process.

[0022] Next, the control unit 13 controls the acquisition of the second image according to the conditions of the second image for performing the second process (S13). For example, the control unit 13 may select an image from a camera that captures an image of the worker's hands according to the conditions of the image required for the finger motion detection process or the like, or may set the imaging conditions for the camera to capture an image of the worker's hands.

[0023] In this manner, in the embodiment, the second processing to be performed next is identified based on the processing result of the first processing performed on the camera image, and the acquisition of the image from the camera is controlled according to the image conditions for performing the second processing. This makes it possible to appropriately acquire the image necessary for the second processing according to the processing result of the first processing. By acquiring appropriate image, it becomes possible to improve the accuracy of the second processing.

[0024] (Embodiment 1) Next, a description will be given of embodiment 1. This embodiment is an example in which a second application process or the like is identified according to a processing result of a first application process, and video conditions for the second application process or the like are identified.

[0025] 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.

[0026] 4, the monitoring system 1 includes a plurality of cameras 101 and a monitoring server 100. The plurality of cameras 101 are arranged at each work process of the production line. The monitoring server 100 may be arranged in the same production site as the production line including each work process, or may be arranged in a location away from the production site.

[0027] The cameras 101 and the surveillance server 100 are communicatively connected via any network. The network may be a wired network, a wireless network, or a combination thereof. 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.

[0028] The multiple cameras 101 include cameras 101 of different types. A set of cameras 101 of multiple types is installed so as to be able to capture the work area of ​​each work process. It can also be said that the multiple cameras 101 are capable of capturing images of workers in the work area. A camera 101 captures the work area of ​​one work process, but may capture the work areas of multiple work processes. For example, the set of cameras 101 of multiple types includes an overhead camera 101a, a handheld camera 101b, and a PTZ camera 101c. The set of cameras 101 of multiple types is not limited to this example and may include other types of cameras, such as a fisheye camera. A set of cameras 101 of different types may be installed for each work process. Note that each work process is not limited to a set of cameras 101 of multiple types, and only one type of camera 101 (e.g., a PTZ camera) may be installed.

[0029] The overhead camera 101a is a camera that can capture an overhead view of the entire work area, including the worker. FIG. 5 shows an example of an image captured by the overhead camera 101a. For example, the overhead camera 101a can capture an image of a workbench (work table) from above, capturing images of the worker working, the worker's hands, the parts being worked on, the tools being used for the work, etc. The overhead camera 101a is fixed in a position that allows it to capture an overhead view of the entire work area. Under control of the monitoring server 100, the overhead camera 101a captures video including the entire work area and transmits the captured video to the monitoring server 100 via a network.

[0030] The handheld camera 101b is a camera capable of capturing images of the worker's hands. FIG. 6 shows an example of an image captured by the handheld camera 101b. For example, the handheld camera 101b can capture images of the worker's hands from any angle. The handheld camera 101b is fixed in a position where it can capture images of the worker's hands. For example, the angle and field of view of the handheld camera 101b can be controlled to capture images of the worker's hands. The handheld camera 101b captures images including the worker's hands under control of the monitoring server 100 and transmits the captured images to the monitoring server 100 via a network.

[0031] The PTZ camera 101c is a camera that can capture images of the work area, including the worker, at any angle and zoom. The PTZ camera 101c may be fixed in a position where it can capture a portion of the work area, or its capture position may be movable. The PTZ camera 101c is controllable for panning (horizontal angle), tilting (vertical angle), and zooming. The PTZ camera 101c captures images of any part of the worker under control of the monitoring server 100 and transmits the captured images to the monitoring server 100 via a network.

[0032] The monitoring server 100 is a monitoring device that monitors each work process on a production line. The monitoring server 100 may be one or more physical servers, or may be a virtualized server built on a virtualization platform. The monitoring server 100 may be a cloud server on the cloud, or an on-premise server installed at a work site. For example, the monitoring server 100 is a GPU (Graphics Processing Unit) server that processes images.

[0033] The monitoring server 100 receives video captured by the camera 101 in each work process and analyzes the received video to monitor the work of workers in each work process. The monitoring server 100 performs application processing that analyzes the video received from the camera 101. The monitoring server 100 is capable of performing multiple types of application processing. The monitoring server 100 performs a first application processing, identifies a target item such as a second application processing based on the processing result of the first application processing, and performs the identified second application processing. The monitoring server 100 also identifies the video conditions required to perform the second application processing, and controls the camera 101 based on the video conditions. The monitoring server 100 is also a specifying device that specifies the second application processing, or a specifying device that specifies the video conditions required for the second application processing. The monitoring server 100 is also a control device that controls the video acquired from the camera 101.

[0034] 7 shows an example configuration of the monitoring server 100 according to some embodiments. Note that the configuration in FIG. 7 is merely 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 monitoring server 100 may be located in another device. For example, some or all of the application processing unit 160 may be located in an external device, and the monitoring server 100 may obtain the processing results of the application processing from the external device.

[0035] In the example of FIG. 7, the monitoring server 100 includes a first video acquisition unit 110 , a target item identification unit 120 , a condition identification unit 130 , a video acquisition control unit 140 , a second video acquisition unit 150 , and an application processing unit 160 .

[0036] The application processing unit 160 performs application processing on the video acquired from the camera 101. The application processing unit 160 may output the processing results of the application processing to a display device or the like. The application processing is processing that analyzes the video captured by the camera 101 and detects information related to the worker's work. The application processing unit 160 includes multiple processing units (detection units) that perform multiple application processing. For example, a first application processing among the multiple application processing is performed on the video of the camera 101 acquired by the first video acquisition unit 110, and a second application processing among the multiple application processing is performed on the video of the camera 101 acquired by the second video acquisition unit 150. Any of the multiple application processing may be the first application processing, or any of the multiple application processing may be the second application processing. For example, the first application processing is analysis processing that simply analyzes the video, and the second application processing is analysis processing that analyzes the video in detail.

[0037] In this example, the application processing unit 160 includes a hand movement detection unit 161, a finger movement detection unit 162, a posture detection unit 163, and a gaze detection unit 164. The application processing unit 160 may also include processing units that perform other processes. For example, the application processing unit 160 may also include an object detection unit that detects a predetermined object from a video. For example, the hand movement detection unit 161 is a first application processing unit that performs first application processing, and the finger movement detection unit 162, the posture detection unit 163, and the gaze detection unit 164 are a second application processing unit that performs second application processing, but this example is not limiting.

[0038] The hand movement detection unit 161 performs hand movement detection processing to detect the hand movements (hand movements) of a worker from the video. For example, the hand movement detection unit 161 detects the hand movements of a worker in the work area from the overhead video captured by the overhead camera 101a shown in FIG. 5. The hand movement detection processing can easily analyze the worker's work based on the worker's hand movements and detect work abnormalities, work time, etc. The hand movement detection unit 161 may perform the hand movement detection processing using an object recognition engine that uses machine learning such as deep learning. For example, the object recognition engine recognizes objects in the image and analyzes the worker's work from the position of the recognized person's (worker's) hand. For example, the hand movement detection processing can analyze the work status, including when the work is in progress or stopped, and the cycle time (work time) from the start to finish of the work. For example, the hand movement detection unit 161 may detect work abnormalities based on the difference between the detected cycle time and a predetermined work time. Note that a cycle is a unit of a series of work repeatedly performed in the work area. The hand movement detection unit 161 may detect a cycle pattern of a task based on the detection result of the worker's hand in the task area and output a cycle waveform indicating the cycle pattern. The cycle waveform is a waveform that shows the hand detection result in time series and represents the task status of each cycle. For example, the hand detection result may indicate whether a hand is detected in the task area or 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 a task or the time when the hand position is within the task area to 1 and other times to 0, and then averaging the time. The hand movement detection unit 161 may calculate a cycle time from the start to the end of the task in a cycle based on the cycle waveform and output the calculated cycle time of the task.

[0039] The finger movement detection unit 162 performs a finger movement detection process to detect the movement of the worker's fingers (finger movements) from the video. For example, the finger movement detection unit 162 detects the worker's finger movements from the video of the worker's hands captured by the hand-held camera 101b shown in FIG. 6 . The finger movement detection process analyzes the worker's work in detail based on the worker's finger movements, and can detect the content of the work, etc. The finger movement detection unit 162 may perform the finger movement detection process using a skeleton estimation engine, an object recognition engine, and a finger movement recognition engine that use machine learning such as deep learning. For example, the skeleton estimation engine estimates the skeleton of the fingers in the image, and the object recognition engine recognizes parts, tools, etc. in the image. Furthermore, the finger movement recognition engine recognizes the content of the work, etc., from the co-occurrence relationship between the feature amounts of the finger skeleton and the feature amounts of the parts, tools, etc.

[0040] The posture detection unit 163 performs posture detection processing to detect the posture of the worker from the video. For example, the posture detection unit 163 detects the posture of the worker from the video of the worker's upper body captured by the PTZ camera 101c. The posture detection processing can analyze the worker's behavior in detail based on the worker's posture. The posture detection unit 163 may perform the posture detection processing using a posture estimation engine that uses machine learning such as deep learning. For example, the posture detection unit 163 may estimate the posture of a person (worker) in the image using the posture estimation engine, and further detect the person's behavior from the movement of the estimated posture of the person.

[0041] The gaze detection unit 164 performs gaze detection processing to detect the gaze of a worker from the video. For example, the gaze detection unit 164 detects the gaze of the worker from the video of the worker's eyes captured by the PTZ camera 101c. The gaze detection processing can analyze in detail where the worker is looking based on the worker's gaze. The gaze detection unit 164 may perform the gaze detection processing using a gaze estimation engine that uses machine learning such as deep learning. For example, the gaze detection unit 164 may estimate the gaze of a person (worker) in the image using the gaze estimation engine, and further detect where the person is looking from the estimated gaze of the person.

[0042] The first image acquisition unit 110 acquires an image (first image) captured by the camera 101 from the camera 101 via a network. For example, the acquired image may include an image of a work area of ​​one work process, but may also include images of work areas of multiple work processes. The first image acquisition unit 110 also functions as an image acquisition unit that acquires multiple images in time series, i.e., frames. The first image acquisition unit 110 outputs the image acquired from the camera 101 to a processing unit (detection unit) in the application processing unit 160 that performs first application processing. For example, the first image acquisition unit 110 may output the acquired image to a hand movement detection unit 161 that performs hand movement detection processing.

[0043] The attention item identifying unit 120 identifies an attention item to be focused on from the video acquired from the camera 101. The attention item is an item that the monitor needs to pay attention to in order to monitor the work. For example, the attention item may be a second application process to be performed on the video, or may be the video of the camera to be focused on (or the camera capturing the video to be focused on). The attention item identifying unit 120 identifies the attention item based on the processing result of a processing unit that performs first application processing in the application processing unit 160. For example, when hand gesture detection processing is performed as the first application processing, the attention item identifying unit 120 may identify finger gesture detection processing as the second application processing to be performed next based on the processing result of the hand gesture detection processing.

[0044] The condition specifying unit 130 specifies video conditions required for the specified item of interest according to the specified item of interest. For example, if a second application process for video is specified as the item of interest, the condition specifying unit 130 may specify video conditions required for performing the second application process. If a finger motion detection process is specified as the second application process, the condition specifying unit 130 may specify video conditions including hands required for the finger motion detection process. If a posture detection process is specified as the second application process, the condition specifying unit 130 may specify video conditions including the entire upper body of a person required for the posture detection process. Furthermore, if video from a camera to be focused on is specified as the item of interest, the condition specifying unit 130 may specify video conditions for the corresponding camera.

[0045] The image acquisition control unit 140 controls the image to be acquired from the camera 101 based on the specified image conditions. When the image conditions include the camera's shooting conditions, the image acquisition control unit 140 may control the camera's shooting conditions according to the image conditions. When the image conditions include a condition for specifying the image of the camera 101, the image acquisition control unit 140 may control the specified camera 101 to shoot the image. The image acquisition control unit 140 may select the image shot by the specified camera 101.

[0046] The second image acquisition unit 150 acquires, via a network, an image (second image) captured by the camera 101 controlled by the image acquisition control unit 140. The second image acquisition unit 150 may acquire an image captured by the camera 101 under set shooting conditions. The second image acquisition unit 150 may also acquire an image captured by a specified camera 101. The second image acquisition unit 150 outputs the image acquired from the camera 101 to a processing unit (detection unit) in the application processing unit 160 that performs second application processing. For example, when finger motion detection processing is performed as the second application processing, the second image acquisition unit 150 may acquire an image captured by the handheld camera 101b and output the acquired image to the finger motion detection unit 162.

[0047] Fig. 8 shows an example of the operation of the monitoring system 1 according to some embodiments. In the example of Fig. 8, the monitoring server 100 acquires video from the camera 101 (S101). The camera 101 captures an image of a work area in a work process and transmits the captured video to the monitoring server 100. The first video acquisition unit 110 acquires the video transmitted from the camera 101 via the network. For example, when a hand gesture detection process is performed as the first application process, the first video acquisition unit 110 acquires video captured by the overhead camera 101a and outputs the acquired video to the hand gesture detection unit 161.

[0048] Next, the monitoring server 100 executes a first application process based on the video acquired from the camera 101 (S102). For example, when performing hand movement detection processing as the first application process, the hand movement detection unit 161 executes the hand movement detection process on the video acquired from the overhead camera 101a. For example, the hand movement detection unit 161 detects hand movements from the acquired video and analyzes the cycle time (work time). The hand movement detection unit 161 may detect an abnormality in the work if the difference between the analyzed cycle time and a predetermined work time is greater than a predetermined value, if the difference between the average cycle time and the predetermined work time is greater than a predetermined value, or if the variance of the cycle time is greater than a predetermined value. The predetermined work time may be, for example, a time designed when the production line was designed. The hand movement detection unit 161 outputs the cycle time and the detected abnormality as a processing result to the attention item identification unit 120.

[0049] Next, the monitoring server 100 identifies an item of interest to be focused on from the acquired video (S103). The item of interest identification unit 120 identifies an item of interest for the second application process, etc., based on the processing result of the first application process. For example, an item of interest to be identified for each processing result of the first application process may be set in advance using a table, etc. Note that the example is not limited to using a table, and an item of interest may also be identified using a learning model, etc., in which an item of interest corresponding to the processing result of the first application process is learned by machine learning.

[0050] 9 shows an example of an application table in which a second application process is set as a focus item. In the example of FIG. 9, a second application process is associated with each processing result of a first application. For example, upon acquiring the processing result of the first application, the focus item identification unit 120 refers to the application table of FIG. 9 and identifies a second application process according to the processing result of the first application process.

[0051] In this example, the attention item identification unit 120 performs hand movement detection processing as the first application processing, and if an abnormality is detected in the hand movement detection processing, a more detailed analysis is required by finger movement detection processing, and therefore identifies finger movement detection processing as the second application processing. Also, the attention item identification unit 120 performs hand movement detection processing as the first application processing, and if the working time detected in the hand movement detection processing is longer than a predetermined time, the cause is required to be analyzed by gaze detection processing, and therefore identifies gaze detection processing as the second application processing.

[0052] In addition, when object detection processing is performed as the first application processing and a specific part is detected in the object detection processing, the focus item identification unit 120 identifies finger movement detection processing as the second application processing because the work using the specific part needs to be analyzed in detail by finger movement detection processing.

[0053] 10 shows an example of a camera image table in which camera images are set as focus items. In the example of Fig. 10, camera images are associated with each processing result of a first application. For example, when the focus item identification unit 120 acquires the processing result of the first application, it refers to the camera image table of Fig. 10 and identifies the camera image according to the processing result of the first application.

[0054] In this example, the item of interest identification unit 120 performs hand movement detection processing as the first application processing, and if the work time of the work process detected by the hand movement detection processing is longer than a predetermined time, it is necessary to check whether there is a wait for work in the subsequent process, and therefore identifies the image from the camera of the subsequent process as the item of interest.

[0055] In addition, when an object detection process is performed as the first application process and a worker (person) is not detected in the work process detected by the object detection process, the item of interest identification unit 120 identifies the image from the camera of the previous process as the item of interest, since it is necessary to check whether a problem has occurred in the previous process.

[0056] Next, the monitoring server 100 identifies the required video conditions according to the identified item of interest (S104). The condition identification unit 130 identifies the video conditions required for the item of interest according to the item of interest, such as the second application process. For example, the video conditions to be identified for each item of interest may be set in advance using a table or the like. Note that the example using a table is not limited to this, and the video conditions may also be identified using a learning model or the like that has machine-learned the video conditions according to the item of interest.

[0057] 11 shows an example of a video condition table that sets video conditions when a second application process is identified as the item of interest. In the example of Fig. 11, video conditions are associated with each second application process. For example, when a second application process is identified as the item of interest, the condition identification unit 130 refers to the video condition table of Fig. 11 and identifies video conditions according to the second application process.

[0058] In this example, when finger motion detection processing is identified as the second application processing, the condition specifying unit 130 sets the video condition to be video that shows the worker's hands at a predetermined angle to make it easier to analyze the finger movements of the worker. When posture detection processing is identified as the second application processing, the condition specifying unit 130 sets the video condition to be video that shows the worker's entire upper body to make it easier to analyze the worker's posture. When gaze detection processing is identified as the second application processing, the condition specifying unit 130 sets the video condition to be video that shows a close-up of the worker's eyes to make it easier to analyze the worker's gaze. When hand motion detection processing is identified as the second application processing, the condition specifying unit 130 sets the video condition to be video that shows a bird's-eye view of the worker's hands to make it easier to analyze the worker's hand movements. When object detection processing is identified as the second application processing, the condition specifying unit 130 sets the video condition to be video that shows a wide range of the worker's surroundings to make it easier to analyze the worker's surroundings.

[0059] Next, the monitoring server 100 controls the camera 101 to capture video (S105a) or instructs the camera 101 to capture video (S105b) based on the conditions of the identified video. The video acquisition control unit 140 identifies the camera 101 based on the conditions of the video and controls the identified camera 101 to transmit the video. For example, the video acquisition control unit 140 sets shooting conditions such as angle, brightness, resolution, zoom, angle of view, and frame rate for the camera 101 according to the conditions of the video and instructs the camera 101 to capture video. The frame rate may be 5 FPS (frames per second) when performing finger motion detection processing and 10 FPS when performing hand motion detection processing. For example, if video from the handheld camera 101b capturing a hand or the PTZ camera 101c capturing an upper body is identified, the video acquisition control unit 140 instructs the handheld camera 101b or the PTZ camera 101c to capture video.

[0060] In the example of FIG. 11 , when an image capturing the hand at a predetermined angle is required for the finger movement detection process, the image acquisition control unit 140 instructs the hand-held camera 101b to set the image capturing angle to a predetermined angle and capture the hand. When an image capturing the entire upper body is required for the posture detection process, the image acquisition control unit 140 instructs the PTZ camera 101c to set the image capturing angle to capture the entire upper body of the worker and capture the image. When an image capturing a close-up image of the eyes is required for the gaze detection process, the image acquisition control unit 140 instructs the PTZ camera 101c to set the image capturing angle and zoom to capture a close-up image of the worker's eyes and capture the image. When an image capturing a bird's-eye view of the hand is required for the hand movement detection process, the image acquisition control unit 140 instructs the bird's-eye camera 101a to capture the image. When an image capturing a wide range of the surroundings is required for the object detection process, the image acquisition control unit 140 instructs the fisheye camera to capture the image.

[0061] Next, the monitoring server 100 acquires video according to the control (S106). The second video acquisition unit 150 acquires video controlled by the video acquisition control unit 140. For example, the second video acquisition unit 150 acquires, via a network, video captured by the camera 101 whose shooting conditions have been controlled. The second video acquisition unit 150 also acquires, via a network, video captured by the handheld camera 101b or the PTZ camera 101c that has been instructed to capture video. The second video acquisition unit 150 outputs the acquired video to a processing unit that performs the second application processing function of the application processing unit 160.

[0062] Furthermore, the second image acquisition unit 150 may acquire images for a certain period of time according to the conditions of the identified images. The second image acquisition unit 150 may collectively acquire images for a certain period of time according to the work content in the camera 101. The second image acquisition unit 150 may calculate the time based on the work content from manual input, a distribution of past work times, or the like.

[0063] 11 , when handheld camera 101b is controlled to capture a picture of the hand at a predetermined angle for finger movement detection processing, second image acquisition unit 150 acquires an image from controlled handheld camera 101b and outputs the acquired image to finger movement detection unit 162. When PTZ camera 101c is controlled to capture the entire upper body for posture detection processing, second image acquisition unit 150 acquires an image from controlled PTZ camera 101c and outputs the acquired image to posture detection unit 163. When PTZ camera 101c is controlled to capture an enlarged image of the eyes for gaze detection processing, second image acquisition unit 150 acquires an image from controlled PTZ camera 101c and outputs the acquired image to gaze detection unit 164. When overhead camera 101a is instructed to capture video of a hand from above for hand motion detection processing, second image acquisition unit 150 acquires video from instructed overhead camera 101a and outputs the acquired video to hand motion detection unit 161. When fish-eye camera is instructed to capture a wide range of the surroundings for object detection processing, second image acquisition unit 150 acquires video from the instructed fish-eye camera and outputs the acquired video to the object detection unit.

[0064] Next, the monitoring server 100 performs a second application process based on the acquired video (S107). Any processing unit (detection unit) of the application processing unit 160 performs the second application process and outputs the processing result to a display device or the like.

[0065] For example, when finger movement detection processing is performed as the second application processing, the finger movement detection unit 162 performs the finger movement detection processing on the image of the worker's hands acquired from the handheld camera 101b. The finger movement detection unit 162 detects the worker's finger movements from the acquired image and outputs the analyzed work content of the worker, etc. When posture detection processing is performed as the second application processing, the posture detection unit 163 performs the posture detection processing on the image of the upper body acquired from the PTZ camera 101c. The posture detection unit 163 detects the worker's posture from the acquired image and outputs the analyzed behavior of the worker, etc. When gaze detection processing is performed as the second application processing, the gaze detection unit 164 performs the gaze detection processing on the image of the worker's eyes acquired from the PTZ camera 101c. The gaze detection unit 164 detects the worker's gaze from the acquired image and outputs the analyzed part where the worker is looking, etc.

[0066] When performing hand movement detection processing as the second application processing, the hand movement detection unit 161 performs the hand movement detection processing on the image acquired from the overhead camera 101a. The hand movement detection unit 161 detects the hand movement of the worker from the acquired image and outputs the analyzed cycle time, etc. When performing object detection processing as the second application processing, the object detection unit performs the object detection processing on the image acquired from the fisheye camera. The object detection unit detects a predetermined object from the acquired image and outputs the presence or absence of the predetermined object, etc.

[0067] As described above, in this embodiment, a second application process or the like is identified according to the processing result of a first application process, image conditions for the second application process or the like are identified, and the camera is controlled according to the identified image conditions. For example, according to the processing result of a hand movement detection process, the next process to be performed may be identified as finger movement detection process, and the camera may be controlled to capture images under the image conditions required for this finger movement detection process. This allows the camera to be appropriately controlled to capture images required for the process reliably.

[0068] (Embodiment 2) Next, embodiment 2 will be described. This embodiment is an example in which the second application process and video conditions are specified based on information about the worker. Note that this embodiment can be implemented in combination with embodiment 1, and the configurations shown in embodiment 1 may be used as appropriate.

[0069] 12 shows an example of the configuration of the monitoring server 100 according to some embodiments. In the example of Fig. 12, the monitoring server 100 includes a worker information acquisition unit 170 in addition to the configuration of Fig. 7.

[0070] The worker information acquisition unit 170 acquires information about the worker performing the work. The worker information acquisition unit 170 may acquire the information about the worker from a database in the monitoring server 100 or from an external system. For example, the information about the worker is the worker's status. The worker's status indicates the worker's level of experience in the work. For example, 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 according to years of service, etc. The status level may be rookie, veteran, etc.

[0071] The worker information acquisition unit 170 identifies the worker performing the work and acquires information about the identified worker. For example, the worker information acquisition unit 170 may detect a person from the video captured by the camera 101 and perform face recognition on the image of the detected person to identify the worker. The worker information acquisition unit 170 may acquire information about a work schedule and identify the worker currently performing the work from the work schedule. The worker information acquisition unit 170 may acquire card information input into a card reader or the like when the worker performs the work and identify the worker from the card information.

[0072] In this example, the monitoring server 100 may identify an item of interest based on information about the worker, or may identify video conditions based on information about the worker. Note that other configurations are the same as those in the example shown in the first embodiment.

[0073] In one example, the attention item identifying unit 120 identifies an attention item such as a second application process based on the acquired information about the worker. The attention item identifying unit 120 may identify an attention item based on the processing result of the application process and the information about the worker.

[0074] 13 shows an example of an application table that sets a second application process according to the worker's status. In the example of Fig. 13, a second application process is associated with each worker's status. For example, when the attention item identification unit 120 acquires the worker's status, it refers to the application table in Fig. 13 and identifies a second application process according to the worker's status.

[0075] For example, when the length of service is acquired as the worker's status, the attention item identification unit 120 determines whether the worker is a new employee or a veteran based on the length of service. The attention item identification unit 120 may compare the worker's length of service with a predetermined number of years of service. In this example, if the attention item identification unit 120 determines that the worker is a new employee, it is necessary to analyze in detail the work of new employees who are more likely to make errors, and therefore it identifies finger movement detection processing as the second application processing. Furthermore, if the attention item identification unit 120 determines that the worker is a veteran, it is not necessary to analyze in detail the work of veteran employees who are less likely to make errors, and therefore it identifies hand movement detection processing as the second application processing.

[0076] For example, when a proficiency level is acquired as the worker's status, the attention item identification unit 120 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 attention item identification unit 120 identifies finger gesture detection processing as the second application process because, as with a newcomer, it is necessary to analyze in detail the work of a worker with low proficiency. On the other hand, if the worker's proficiency level is higher than the predetermined proficiency level, the attention item identification unit 120 identifies hand gesture detection processing as the second application process because, as with a veteran, it is not necessary to analyze in detail the work of a worker with high proficiency.

[0077] For example, when an error frequency is acquired as the worker's status, the attention item identification unit 120 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 attention item identification unit 120 identifies finger movement detection processing as the second application process because, like a new employee, it is necessary to analyze in detail the work of a worker who makes more errors. On the other hand, if the worker's error frequency is lower than the predetermined error frequency, the attention item identification unit 120 identifies hand movement detection processing because, like a veteran, it is not necessary to analyze in detail the work of a worker who makes less errors.

[0078] In another example, the condition specifying unit 130 specifies the video conditions based on the acquired information about the worker. The condition specifying unit 130 may specify the video conditions based on information about the worker and a focus item such as a second application process. FIG. 14 shows an example of a video condition table that sets video conditions according to the worker's status. In the example of FIG. 14, video conditions are associated with each worker's status. For example, when the condition specifying unit 130 acquires the worker's status, it refers to the video condition table of FIG. 14 and specifies the video conditions according to the worker's status.

[0079] For example, if the condition specifying unit 130 determines that the worker is a new employee based on the length of service, it increases the zoom and brightness of the image around the worker's hand as a condition for the image, since it is necessary to accurately analyze the work of the new employee, which is likely to be error-prone.Also, if the condition specifying unit 130 determines that the worker is a veteran based on the length of service, it does not need to accurately analyze the work of the veteran, which is less likely to be error-prone, so it decreases the zoom and brightness of the image around the worker's hand as a condition for the image.

[0080] For example, when the proficiency level of a worker is lower than a predetermined level, the condition specifying unit 130 increases the frequency of image capture (frame rate, etc.) and resolution as image conditions because it is necessary to analyze the work of a less skilled worker with high accuracy, just like a newcomer. On the other hand, when the proficiency level of a worker is higher than the predetermined level, the condition specifying unit 130 decreases the frequency of image capture and resolution as image conditions because it is not necessary to analyze the work of a more skilled worker with high accuracy, just like a veteran.

[0081] For example, if the worker's error frequency is higher than a predetermined error frequency, the condition specifying unit 130 increases the frequency and resolution of the video as a video condition because, like a new employee, it is necessary to accurately analyze the work of a worker who makes more errors. On the other hand, if the worker's error frequency is lower than the predetermined error frequency, the condition specifying unit 130 decreases the frequency and resolution of the video as a video condition because, like a veteran, it is not necessary to accurately analyze the work of a worker who makes less errors.

[0082] As described above, the second application process and the video conditions may be identified based on the information of the worker performing the work. For example, the required process and video conditions change depending on the worker's status, so the second application process and the video conditions can be identified appropriately.

[0083] (Third Embodiment) Next, a third embodiment will be described. This embodiment is an example of performing distributed processing using edge devices and a cloud server. Note that this embodiment can be implemented in combination with the first or second embodiment, and the configurations shown in the first or second embodiment may be used as appropriate.

[0084] Fig. 15 shows an example configuration of a monitoring system 1 according to some embodiments. In the example of Fig. 15, the monitoring system 1 includes an edge device 200 and a cloud server 300 in addition to a monitoring server 100. For example, the edge device 200 is located at a manufacturing site (edge) close to the camera 101. The monitoring server 100 and the cloud server 300 are located at a remote location (cloud) away from the manufacturing site. Note that either or both of the monitoring server 100 and the cloud server 300 may be located at the same manufacturing site as the edge device 200.

[0085] The edge device 200 and the monitoring server 100 are communicably connected via a network NW1. The edge device 200 and the cloud server 300 are communicably connected via a network NW2. The network NW1 and the network NW2 may be the same network.

[0086] In the example of FIG. 15 , the functions shown in FIG. 7 are distributed to each device. Note that the functions shown in FIG. 12 may also be distributed to each device. The configuration of each device in FIG. 15 is an example, and other configurations may also be used. The edge device 200 may include the first image acquisition unit 110, the second image acquisition unit 150, the hand motion detection unit 161, and the finger motion detection unit 162 shown in FIG. 7 , and may further include an image transmission unit 210. The image transmission unit 210 transmits the image acquired by the second image acquisition unit 150 to the cloud server 300 via the network NW2.

[0087] The monitoring server 100 includes the attention item identification unit 120, the condition identification unit 130, and the video acquisition control unit 140 shown in Fig. 7, and may further include a processing result acquisition unit 180. The processing result acquisition unit 180 acquires the processing result of the first application processing from the edge device 200 via the network NW1.

[0088] 7, the cloud server 300 may include the posture detection unit 163 and the gaze detection unit 164, and may further include a video receiving unit 310. The video receiving unit 310 receives the video transmitted from the edge device 200 via the network NW2.

[0089] 16 shows an example of the operation of the monitoring system 1 according to some embodiments. In the example of Fig. 16, similar to Fig. 8, the edge device 200 acquires video from the camera 101 (S101) and performs a first application process based on the acquired video (S102). For example, when performing hand gesture detection processing as the first application process, the hand gesture detection unit 161 performs the hand gesture detection processing on the acquired video and notifies the monitoring server 100 of the processing result via the network NW1.

[0090] Next, the monitoring server 100 acquires the processing result of the first application processing from the edge device 200 (S110). For example, the processing result acquisition unit 180 acquires the processing result from the hand gesture detection unit 161 of the edge device 200 via the network NW1.

[0091] 8 , the monitoring server 100 identifies an item of interest based on the acquired processing result of the first application processing (S103), and identifies the required video conditions according to the identified item of interest (S104). Furthermore, the monitoring server 100 controls the camera 101 to instruct it to capture video (S105a) or instructs the camera 101 to capture video (S105b) according to the video conditions. The video acquisition control unit 140 transmits an instruction to control the camera 101 via the edge device 200. The video acquisition control unit 140 may control the video acquired by the edge device 200 and transmitted to the cloud server 300.

[0092] Next, the edge device 200 acquires an image corresponding to the control from the camera 101 (S106) and transmits the acquired image to the cloud server 300 (S111). When the second image acquisition unit 150 acquires an image from the camera 101, the image transmission unit 210 transmits the image acquired by the second image acquisition unit 150 to the cloud server 300 via the network NW2.

[0093] Next, the cloud server 300 receives the video from the edge device 200 (S112). The video receiving unit 310 receives the video transmitted from the edge device 200 via the network NW2. Next, the cloud server 300 performs the second application process based on the received video, similar to FIG. 8 (S107).

[0094] As described above, the functions of the monitoring server may be distributed between the edge device and the cloud server. For example, each processing unit (detection unit) for application processing may be distributed between the edge device and the cloud server, and application processing may be performed by either device. Even with this configuration, as in the first embodiment, it is possible to appropriately control camera capture and reliably obtain the images necessary for processing. Furthermore, since the location where application processing is performed can be selected, the processing load can be distributed.

[0095] The present disclosure is not limited to the above-described embodiment, and can be modified as appropriate within the scope of the present disclosure.

[0096] 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 monitoring server, edge device, cloud server, etc. 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. 17. For example, a program for performing the method (control 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.

[0097] 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.

[0098] 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.

[0099] 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.

[0100] Some or all of the above embodiments may also be described as, but not limited to, the following supplementary notes: Some or all of the elements described in any supplementary note may be applied to various hardware, software, recording means for recording software, systems, and methods.

[0101] (Supplementary Note 1) A control system comprising: an acquisition means for acquiring a processing result of a first processing performed on a first video of an object; an identification means for identifying a second processing to be performed on a second video of the object based on the processing result; and a control means for controlling acquisition of the second video in accordance with conditions of the second video for performing the second processing. (Supplementary Note 2) The control system described in Supplementary Note 1, wherein the control means controls the camera in accordance with conditions for identifying a camera that will capture the second video or shooting conditions of the camera that will capture the second video. (Supplementary Note 3) The control system described in Supplementary Note 1 or 2, wherein the object is a worker performing work, the first processing is processing for analyzing the work, and the second processing is processing for analyzing the work in more detail than the first processing. (Supplementary Note 4) The control system described in Supplementary Note 3, wherein the identification means identifies the second processing based on information about the worker. (Supplementary Note 5) The control system according to Supplementary Note 3, wherein the control means controls acquisition of the second video in accordance with conditions for the second video based on information about the worker. (Supplementary Note 6) The control system according to Supplementary Note 4 or 5, wherein the information about the worker includes the status of the worker. (Supplementary Note 7) The control system according to any one of Supplements 1 to 6, comprising: a first processing means that performs the first processing; and a second processing means that performs the second processing, wherein the control means controls the second video that is acquired by the first processing means and transmitted to the second processing means. (Supplementary Note 8) The control system according to Supplementary Note 7, wherein the first processing means is included in an edge device disposed at an edge, and the second processing means is included in a cloud server disposed at a cloud. (Supplementary Note 9) A control device comprising: an acquisition means for acquiring a processing result of a first process performed on a first image capturing an object; an identification means for identifying a second process to be performed on a second image capturing the object based on the processing result; and a control means for controlling acquisition of the second image according to conditions of the second image for performing the second process.(Supplementary Note 10) The control device according to Supplementary Note 9, wherein the control means controls the camera according to conditions for identifying a camera that captures the second video or shooting conditions of the camera that captures the second video. (Supplementary Note 11) The control device according to Supplementary Note 9 or 10, wherein the target is a worker performing work, the first process is a process for analyzing the work, and the second process is a process for analyzing the work in more detail than the first process. (Supplementary Note 12) The control device according to Supplementary Note 11, wherein the identification means identifies the second process based on information about the worker. (Supplementary Note 13) The control device according to Supplementary Note 11, wherein the control means controls acquisition of the second video according to conditions for the second video that are based on information about the worker. (Supplementary Note 14) The control device according to Supplementary Note 12 or 13, wherein the information about the worker includes the status of the worker. (Supplementary Note 15) The control device according to any one of Supplements 9 to 14, comprising: a first processing means that performs the first processing; and a second processing means that performs the second processing, wherein the control means controls the second video acquired by the first processing means and transmitted to the second processing means. (Supplementary Note 16) The control device according to Supplementary Note 15, wherein the first processing means is included in an edge device disposed at an edge, and the second processing means is included in a cloud server disposed at a cloud. (Supplementary Note 17) A control method that acquires a processing result of a first processing performed on a first video image of an object, identifies a second processing to be performed on a second video image of the object based on the processing result, and controls acquisition of the second video image according to conditions of the second video image for performing the second processing. (Supplementary Note 18) The control method according to Supplementary Note 17, wherein the camera is controlled according to conditions for identifying a camera that will capture the second video image or a capturing condition of the camera that will capture the second video image. (Supplementary Note 19) The control method according to Supplementary Note 17 or 18, wherein the target is a worker performing work, the first process is a process for analyzing the work, and the second process is a process for analyzing the work in more detail than the first process.(Supplementary Note 20) The control method according to Supplementary Note 19, wherein the second process is identified based on information about the worker. (Supplementary Note 21) The control method according to Supplementary Note 19, wherein acquisition of the second image is controlled according to a condition for the second image based on information about the worker. (Supplementary Note 22) The control method according to Supplementary Note 20 or 21, wherein the information about the worker includes the status of the worker. (Supplementary Note 23) The control method according to any one of Supplements 17 to 22, wherein the second image is acquired by a first processing means that performs the first process and is transmitted to a second processing means that performs the second process. (Supplementary Note 24) The control method according to Supplementary Note 23, wherein the first processing means is included in an edge device arranged at an edge, and the second processing means is included in a cloud server arranged at a cloud.

[0102] REFERENCE SIGNS LIST 1 Monitoring system 10 Control system 11 Acquisition unit 12 Identification unit 13 Control unit 20 Control device 30 Computer 31 Processor 32 Memory 100 Monitoring server 101 Camera 101a Bird's-eye view camera 101b Handheld camera 101c PTZ camera 110 First image acquisition unit 120 Attention item identification unit 130 Condition identification unit 140 Image acquisition control unit 150 Second image acquisition unit 160 Application processing unit 161 Hand gesture detection unit 162 Finger gesture detection unit 163 Posture detection unit 164 Gaze detection unit 170 Worker information acquisition unit 180 Processing result acquisition unit 200 Edge device 210 Image transmission unit 300 Cloud server 310 Image reception unit

Claims

1. A control system comprising: an acquisition means for acquiring a processing result of a first processing performed on a first image capturing an object; an identification means for identifying a second processing to be performed on a second image capturing the object based on the processing result; and a control means for controlling the acquisition of the second image in accordance with conditions of the second image for performing the second processing.

2. The control system according to claim 1, wherein the control means controls the camera according to conditions for identifying the camera that captures the second image or shooting conditions of the camera that captures the second image.

3. The control system of claim 1 or 2, wherein the target is a worker performing work, the first process is a process for analyzing the work, and the second process is a process for analyzing the work in more detail than the first process.

4. The control system according to claim 3, wherein the identifying means identifies the second process based on information about the worker.

5. The control system according to claim 3, wherein the control means controls acquisition of the second image in accordance with a condition for the second image based on information relating to the worker.

6. The control system according to claim 4 or 5, wherein the information about the worker includes a status of the worker.

7. A control system as described in any one of claims 1 to 6, comprising: a first processing means for performing the first processing; and a second processing means for performing the second processing, wherein the control means controls the second video acquired by the first processing means and transmitted to the second processing means.

8. The control system according to claim 7, wherein the first processing means is included in an edge device disposed at an edge, and the second processing means is included in a cloud server disposed in a cloud.

9. A control device comprising: an acquisition means for acquiring a processing result of a first processing performed on a first image capturing an object; an identification means for identifying a second processing to be performed on a second image capturing the object based on the processing result; and a control means for controlling the acquisition of the second image according to the conditions of the second image for performing the second processing.

10. The control device according to claim 9, wherein the control means controls the camera according to conditions for identifying the camera that captures the second image or shooting conditions of the camera that captures the second image.

11. The control device according to claim 9 or 10, wherein the target is a worker performing work, the first process is a process for analyzing the work, and the second process is a process for analyzing the work in more detail than the first process.

12. The control device according to claim 11, wherein the identification means identifies the second process based on information about the worker.

13. The control device according to claim 11, wherein the control means controls acquisition of the second image in accordance with a condition for the second image based on information relating to the worker.

14. The control device according to claim 12 or 13, wherein the information about the worker includes a status of the worker.

15. A control method comprising: acquiring a processing result of a first processing performed on a first image capturing an object; identifying a second processing to be performed on a second image capturing the object based on the processing result; and controlling acquisition of the second image according to conditions of the second image for performing the second processing.

16. The control method according to claim 15, further comprising controlling the camera according to a condition for identifying the camera that captures the second video or a shooting condition of the camera that captures the second video.

17. A control method as described in claim 15 or 16, wherein the target is a worker performing work, the first process is a process for analyzing the work, and the second process is a process for analyzing the work in more detail than the first process.

18. The control method according to claim 17, further comprising identifying the second process based on information about the worker.

19. The control method according to claim 17, further comprising controlling acquisition of the second image in accordance with a condition of the second image based on information about the worker.

20. The control method according to claim 18 or 19, wherein the information about the worker includes a status of the worker.

Images