Work management device

Abstract

To provide a work management device capable of improving work efficiency.SOLUTION: A work management device includes: a detection unit that detects, from an image obtained by capturing a worker working on a workpiece being carried in a predetermined direction, coordinates of a skeleton of the worker in the image; a determination unit that determines actions and positions in the predetermined direction of the worker from changes over time in the coordinates of the skeleton of the worker; a calculation unit that calculates a delay in the work of the worker based on the actions and positions in the predetermined direction of the worker; and a request unit that requests a helper to help the worker in accordance with the delay.SELECTED DRAWING: Figure 5

Description

[Technical field]

[0001] The present invention relates to a work management device. [Background technology]

[0002] For example, Patent Document 1 describes a method in which a worker on a production line in a factory presses a push button switch to display a request for help on a display such as an andon. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 6-208690 Summary of the Invention [Problem to be solved by the invention]

[0004] However, because the timing at which workers request help varies from one worker to another, it is difficult to call for help in a timely manner and improve work efficiency.

[0005] Therefore, the present invention has been made in consideration of the above problems, and an object of the present invention is to provide a work management device that can improve work efficiency. [Means for solving the problem]

[0006] The work management device of the present invention has a detection unit that detects the coordinates of a worker's skeleton in an image taken of a worker working on a workpiece being transported in a specified direction, a discrimination unit that discriminates the worker's movements and position in the specified direction from changes over time in the coordinates of the worker's skeleton, a calculation unit that calculates a delay in the worker's work based on the worker's movements and position in the specified direction, and a request unit that requests a helper to help the worker in accordance with the delay.

[0007] The above-mentioned work management device may further include a control unit that reduces a transport speed of the work in response to a delay in the work.

[0008] In the above work management device, the control unit may not need to reduce the transport speed of the work when the helper working on the work is captured in the image.

[0009] In the above-mentioned work management device, the detection unit detects the skeletal coordinates of the worker and the supporter from images taken of the worker and the supporter working on the workpiece, the discrimination unit discriminates the worker's movement and position in the specified direction from changes over time in the skeletal coordinates of the worker and discriminates the supporter's movement and position in the specified direction from changes over time in the skeletal coordinates of the supporter, and the calculation unit may calculate a delay in the work based on the worker's movement and position in the specified direction and the supporter's movement and position in the specified direction.

[0010] In the above-mentioned work management device, the discrimination unit may use a machine-learned discrimination model through supervised learning that inputs the coordinates of the worker's skeleton in the image and outputs the worker's movements, and discriminate the movements of the worker based on changes in the coordinates of the skeleton over time. Effect of the Invention

[0011] According to the present invention, the work efficiency can be improved. [Brief description of the drawings]

[0012] [Figure 1] FIG. 1 is a configuration diagram showing an example of a work management system. [Diagram 2] FIG. 2 is a diagram showing an example of an image captured by a camera device of a worker during work. [Diagram 3] FIG. 3 is a diagram showing an example of the correlation between the motion and position of a worker. [Figure 4] FIG. 4 is a diagram showing an example of an image captured by a camera device of a worker and a supporter during work. [Diagram 5] FIG. 5 is a diagram showing an example of calculation of delays in work by a worker and a helper. [Figure 6] FIG. 6 is a diagram showing another example of calculation of the delay in the work of the worker and the helper. [Figure 7] FIG. 7 is a flowchart showing an example of the operation of the work management server. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0013] (Example of the configuration of a work evaluation system) 1 is a configuration diagram showing an example of a work management system 9. As an example, the work management system 9 is installed in an assembly plant for automobiles or the like.

[0014] The work management system 9 includes a work management server 1, a plurality of camera devices 2, a plurality of displays 3, and a transport device 4, which are capable of communicating with each other via a LAN (Local Area Network) 90. The work management server 1 is an example of a work management device. The work management server 1 manages the work of workers. A camera device 2 and a display device 3 are installed at each worker's work area. The camera device 2 is an example of an imaging device, and captures images of people such as workers. The display device 3 displays the work status of the worker, such as whether or not help is needed. The transport device 4 is the drive source for the conveyor that transports the workpieces that each worker is working on.

[0015] The work management server 1 has a CPU (Central Processing Unit) 10, a ROM (Read Only Memory) 11, a RAM (Random Access Memory) 12, a HDD (Hard Disk Drive) 13, and a communication port 14. The CPU 10 is electrically connected to the ROM 11, the RAM 12, the HDD 13, and the communication port 14 via a bus 19 so as to be able to input and output signals to and from each other.

[0016] The ROM 11 stores a program that drives the CPU 10. The RAM 12 functions as a working memory for the CPU 10. The communication port 14 is, for example, a wireless LAN card, and processes communication of the CPU 10 via the LAN 90.

[0017] When the CPU 10 reads a program from the ROM 11, it forms, as software functions, a device control unit 100, a skeleton detection unit 101, a task determination unit 102, a delay calculation unit 103, a conveyance speed control unit 104, and a support request unit 105. In addition, the HDD 13 stores detected skeleton data 130, learned skeleton data 131, task information 132, and a discrimination model 133.

[0018] The device control unit 100 instructs the skeleton detection unit 101, the task determination unit 102, the delay calculation unit 103, the conveying speed control unit 104, and the support request unit 105 to operate in accordance with a sequence defined in a program. Note that the device control unit 100, the skeleton detection unit 101, the task determination unit 102, the delay calculation unit 103, the conveying speed control unit 104, and the support request unit 105 are not limited to being realized by software, and may be realized by hardware such as an ASIC (Application Specified Integrated Circuit).

[0019] The skeleton detection unit 101 is an example of a detection unit. The skeleton detection unit 101 detects the coordinates of the skeleton of the worker in an image captured by the camera device 2, the image capturing the worker working on a workpiece being transported in a predetermined transport direction. The skeleton detection unit 101 receives image data from the camera device 2 via the LAN 90. The skeleton detection unit 101 detects the coordinates of the skeleton of the worker by image analysis using, for example, a general-purpose AI (Artificial Intelligence).

[0020] The skeleton detection unit 101 detects the coordinates of the evaluation worker's skeleton from the image of the evaluation worker and stores them in the HDD 13 as detected skeleton data 130. The detected skeleton data 130 indicates, for example, the coordinates of the evaluation worker's nose, left shoulder, right shoulder, left elbow, right elbow, left hand, right hand, left hip, and right hip in the image in chronological order for each frame number of the image. The coordinates are indicated by the X coordinate and the Y coordinate (x, y) when the horizontal and vertical directions of the rectangular frame of the image are defined as the X axis and the Y axis, respectively. There is no limitation on the position of the origin.

[0021] The task determination unit 102 is an example of a determination unit. The task determination unit 102 determines the motion of the worker and the position of the workpiece in the transport direction from the time change in the coordinates of the worker's skeleton. Examples of the worker's motion include an assembly motion for assembling parts to the workpiece, a welding motion for welding parts to the workpiece, and a painting motion for painting the workpiece. The position of the worker is indicated by the area in which the worker is working, among multiple areas obtained by dividing the frame of the image of the camera device 2 along a direction substantially perpendicular to the transport direction, for example.

[0022] The task discrimination unit 102 receives as input the coordinates of the worker's skeleton in the image and outputs the worker's motion. Using a discrimination model 133 that has been machine-learned through supervised learning, the task discrimination unit 102 discriminates the worker's motion based on the time-dependent change in the skeleton coordinates. The skeleton coordinates of the exemplary worker learned by the discrimination model 133 are stored in the HDD 13 as learned skeleton data 131. Like the detected skeleton data 130, the learned skeleton data 131 indicates the coordinates of the nose, left shoulder, right shoulder, left elbow, right elbow, left hand, right hand, left hip, and right hip of the exemplary worker in the image in chronological order for each frame number of the image. The discrimination model 133 discriminates the worker's motion based on the degree of similarity of the detected skeleton data 130 to the learned skeleton data 131.

[0023] The discrimination model 133 is stored in the HDD 13 as a set of various calculation modules and parameters. In the learning process of the discrimination model 133, the type of work motion is annotated using the time change in the skeletal coordinates of an exemplary worker during work as teacher data. The discrimination model 133 is a neural network that mathematically models human brain functions, and determines the weighting coefficients of the activation functions of the parts corresponding to neurons based on machine learning. In this way, the task discrimination unit 102 can discriminate the motions of the worker with high accuracy by using the discrimination model 133 that has been machine-learned through supervised learning.

[0024] The delay calculation unit 103 is an example of a calculation unit. The delay calculation unit 103 calculates a delay in the work of a worker based on the worker's motion and position in the conveying direction. The longer the worker is delayed in the work, the more downstream the worker works in the conveying direction. For this reason, the delay calculation unit 103 calculates the delay, for example, from the correlation between the type of motion and the position in the conveying direction, as described below.

[0025] The conveying speed control unit 104 is an example of a control unit. The conveying speed control unit 104 reduces the conveying speed of the workpiece in response to a delay in the work. This makes it easier for the worker to perform the work, and therefore reduces the delay.

[0026] The support request unit 105 is an example of a request unit. The support request unit 105 requests a supporter to support the worker in response to a delay in the work. For example, the support request unit 105 outputs a support request to the display device 3 via a LAN. In response to the support request displayed on the display device 3, the supporter goes to the work site corresponding to the display device 3 to support the work of the worker.

[0027] For example, when the delay exceeds a threshold, the support requesting unit 105 determines whether support is required based on the work information 132. The work information 132 includes, for example, the proficiency level and work load of each worker. In addition, for example, the support requesting unit 105 may request support for the worker who is the most delayed among multiple workers.

[0028] In this way, the delay calculation unit 103 calculates the delay in the work based on the worker's movement determined from the coordinates of the worker's skeleton and the position in the transport direction, and the support request unit 105 requests the supporter to support the worker according to the delay in the work. This makes it possible to request support in a timely manner according to the exact status of the work. Therefore, the work management server 1 can improve the work efficiency.

[0029] The conveying speed control unit 104 does not reduce the conveying speed of the workpiece when a helper working on the workpiece is captured in the image of the camera device 2. Therefore, after the helper joins the work, delays caused by the helper's work are reduced, and the conveying speed of the workpiece is maintained, thereby suppressing a decrease in work efficiency.

[0030] (Example image of a worker at work) 2 is a diagram showing an example of an image G1 captured by the camera device 2 of a worker Ha working. The camera device 2 captures an image of the worker Ha from the front. In the image G1, the horizontal direction is the X direction, and the vertical direction is the Y direction.

[0031] Image G1 captures a worker Ha performing an assembly operation of assembling a part C to a workpiece W. In this example, the workpiece W is an automobile, but is not limited to this. The workpiece W is placed on a conveyor R and is conveyed in a conveying direction that coincides with the X direction in FIG. 2.

[0032] The skeleton detection unit 101 detects the coordinates of the skeleton of the worker Ha from the image G1 by image analysis using a general-purpose AI, and stores them in chronological order as detected skeleton data 130. Examples of the skeleton coordinates include the coordinates of the nose position P0 of the worker Ha, the left shoulder position P23, the right shoulder position P13, the left elbow position P22, the right elbow position P12, the left hand position P21, the right hand position P11, the left hip position P32, and the right hip position P31.

[0033] The task discrimination unit 102 discriminates the task of the worker Ha based on the learned skeleton data 131 from the change over time in the coordinates of each part of the skeleton indicated by the detected skeleton data 130. The task discrimination unit 102 chronologically compares the coordinates of the skeleton of each exemplary task indicated by the learned skeleton data 131 with the coordinates of each part of the skeleton indicated by the detected skeleton data 130. The task discrimination unit 102 determines that the change over time in the positions P0, P11-P13, P21-P23, and P31-P33 of each part of the worker Ha is similar to the change over time in the positions P0, P11-P13, P21-P23, and P31-P33 of each part during the exemplary assembly task, and discriminates the assembly task of the part C of the worker Ha.

[0034] Furthermore, the task determination unit 102 determines the position of the worker Ha in the conveying direction from among areas #1 to #4 in the image G1 based on the coordinates of the skeleton. Areas #1 to #4 are set by dividing the area in the image G1 into substantially four equal parts along the Y direction perpendicular to the conveying direction. The task determination unit 102 determines the position of the worker Ha as area #3 because the positions P0, P11 to P13, P21 to P23, and P31 to P33 of each part of the worker Ha are within area #3. The task determination unit 102 does not necessarily need to use the coordinates of all skeletons to determine the position, and may determine, for example, areas #1 to #4 in which the coordinates (P13, P23, 31, P32) of the skeleton of the worker Ha's torso are present, as the position of the worker Ha.

[0035] The delay calculation unit 103 calculates a delay in the work based on the movement and position of the worker Ha. For example, the delay calculation unit 103 calculates the delay from the correlation between the movement and position of the worker Ha.

[0036] Fig. 3 is a diagram showing an example of the correlation between the actions and positions of worker Ha. As an example, worker Ha performs actions #1 to #4 in this order. Fig. 3 shows the actions #1 to #4 of worker Ha and areas #1 to #4 in chronological order when there is no delay.

[0037] When worker Ha performs actions #1 to #4 in areas #1 to #4 in sequence, respectively, the work delay is always 0. However, when worker Ha performs actions #1 to #4 in areas #1 to #4 downstream of the corresponding areas #1 to #4, it is determined that a work delay has occurred. For example, when worker Ha performs action #2 in area #3, the work delay is 1 (=3-2), and when worker Ha performs action #1 in area #3, the work delay is 2 (=3-1). Also, when worker Ha performs action #2 in area #4, the work delay is 2 (=4-2), and when worker Ha performs action #1 in area #4, the work delay is 3 (=4-1).

[0038] Therefore, the delay calculation unit 103 calculates the work delay (ji) from the action #i and area #j (i, j: positive integers) of the worker Ha when j>i holds. Note that, when j≦i holds, the delay calculation unit 103 calculates the work delay as 0. When the work delay is greater than a threshold value (>0), the support request unit 105 requests a supporter to support the work.

[0039] (Example image of workers and supporters working) Fig. 4 is a diagram showing an example of an image G2 captured by the camera device 2 of the worker Ha and the supporter Hb during work. The camera device 2 captures the worker Ha and the supporter Hb from the front. In Fig. 4, the same components as in Fig. 2 are given the same reference numerals, and their explanation will be omitted. As an example, the supporter Hb is performing an assembly operation to assemble a part D to a workpiece W in area #4 downstream of the worker Ha. It is assumed that the assembly of part D is a process that follows the assembly of part C.

[0040] The skeleton detection unit 101 detects the coordinates of the skeletons of the worker Ha and supporter Hb from image G2. Although the reference symbols are omitted, the coordinates of the nose position P0, left shoulder position P23, right shoulder position P13, left elbow position P22, right elbow position P12, left hand position P21, right hand position P11, left hip position P32, and right hip position P31 are also detected for the skeleton of supporter Hb. The coordinates of the skeleton of supporter Hb are saved as detected skeleton data 130.

[0041] The task determination unit 102 determines the motion and position in the conveying direction of the worker Ha from the time-dependent changes in the coordinates of the worker Ha's skeleton, and determines the motion and position in the conveying direction of the supporter Hb from the time-dependent changes in the coordinates of the supporter Hb's skeleton. The means for determining the motion and position of the supporter Hb is performed in the same manner as in the case of the worker Ha described above. The task determination unit 102 determines the assembling task of the worker Ha for part C and area #3 from the time-dependent changes in the coordinates of the worker Ha's skeleton, and determines the assembling task of the worker Hb for part D and area #4 from the time-dependent changes in the coordinates of the supporter Hb's skeleton.

[0042] The delay calculation unit 103 calculates a delay in the work based on the movement and position in the conveying direction of the worker Ha and the movement and position in the conveying direction of the supporter Hb. As an example, the delay calculation unit 103 detects the movement of the person working downstream in the conveying direction, out of the worker Ha and the supporter Hb, and the position of the person performing the movement of the last process, and uses these to calculate the delay. In this example, the delay is calculated based on the assembly movement of the part D by the supporter Hb and the area #4.

[0043] (Example of calculation of work delay) Fig. 5 is a diagram showing an example of calculation of delays in the work of a worker Ha and a supporter Hb. Reference characters G1a to G1d typically indicate images taken by the camera device 2 in time series (see arrows). In Fig. 5, components common to Fig. 4 are given the same reference characters and description thereof will be omitted.

[0044] As indicated by the symbol G1a, worker Ha is performing action #1 in area #3. The delay in the work at this time is calculated as 2 (=3-1) from the calculation method described above. Here, if the delay threshold for requesting help is 1, the request for help is displayed on the display 3 because the delay exceeds the threshold.

[0045] Next, as indicated by the symbol G1b, the helper Hb arrives at the work site in response to the request and performs the subsequent operation #2 in area #4 downstream of the worker Ha. Also, the worker Ha is continuing the operation #1 in area #3. The delay in the work at this time is calculated as 2 (=4-2) from the downstream area #4 of the areas #3 and #4 in which the work is being performed, and the subsequent operation #2 of the operations #1 and #2.

[0046] Next, as shown by the symbol G1c, the supporter Hb finishes action #2 and performs action #3, and the worker Ha finishes action #1 and performs action #3 together with the supporter Hb. The work delay at this time is calculated as 1 (=4-3) from the downstream area #4 of the areas #3 and #4 in which work is being performed, and action #3.

[0047] Next, as shown by the reference symbol G1d, both the worker Ha and the supporter Hb finish the action #4 and perform the action #4. The delay in the work at this time is calculated as 0 (=4-4).

[0048] Fig. 6 is a diagram showing another example of calculation of the delay in the work of the worker Ha and the supporter Hb. Reference characters G2a to G2c typically indicate images of the camera device 2 in time series (see arrows). In Fig. 6, the same reference characters are used for configurations common to Fig. 5, and description thereof will be omitted.

[0049] As indicated by the symbol G2a, worker Ha is performing action #2 in area #4. The delay in the work at this time is calculated as 2 (=4-2) from the calculation method described above. Here, if the delay threshold for requesting help is 1, the request for help is displayed on the display 3 because the delay exceeds the threshold.

[0050] Next, as indicated by the symbol G2b, the helper Hb arrives at the work site in response to the request and performs the subsequent operation #3 in the area #3 upstream of the worker Ha. Also, the worker Ha is continuing the operation #2 in the area #4. The delay in the work at this time is calculated as 1 (=4-3) from the downstream area #4 of the areas #3 and #4 in which the work is being performed, and the subsequent operation #3 of the operations #1 and #2.

[0051] Next, as shown by the symbol G2c, both worker Ha and helper Hb are performing action #4 in area #4. The work delay at this time is calculated as 0 (=4-4) based on area #4, which is downstream of areas #3 and #4 where work is being performed, and action #4.

[0052] In this way, the delay calculation unit 103 calculates the delay in the work based on the movement and position in the transport direction of the worker Ha and the movement and position in the transport direction of the supporter Hb. Therefore, the work management server 1 can indicate the degree to which the delay has been eliminated by the supporter Hb joining the work. At this time, the delay calculation unit 103 may output the delay to the display 3. Note that the method of calculating the delay in the work of the worker Ha and the supporter Hb is not limited to this example.

[0053] (Operation of the work management server) 7 is a flowchart showing an example of the operation of the work management server 1. This operation is executed in parallel for each of a plurality of workers. First, the skeleton detection unit 101 acquires an image from the camera device 2 via the LAN 90 (step St1). Next, the skeleton detection unit 101 detects the coordinates of the skeletons of one or more people in the image (step St2).

[0054] The task determination unit 102 determines whether or not there is a supporter in the image based on the skeleton detection result (step St3). If there is no supporter (No in step St3), the task determination unit 102 determines the worker's motion and position in the conveying direction (step St4). Next, the delay calculation unit 103 calculates the task delay based on the worker's motion and position in the conveying direction (step St5).

[0055] Next, the support request unit 105 compares the work delay with a threshold value TH_L (step St6). If the work delay ≦ TH_L is true (No in step St6), the device control unit 100 determines whether all work is completed based on the results of the worker's actions (step St13). If work remains (No in step St13), each action from step St1 onwards is performed again, and if all work is completed (Yes in step St13), the operation of the work management server 1 ends.

[0056] If the task delay > TH_L holds (Yes in step St6), the support request unit 105 judges whether support is necessary based on the task information 132 (step St7). At this time, the support request unit 105 calculates the degree of need for support using a predetermined calculation formula from the worker's proficiency and task load included in the task information 132, for example, and judges that support is necessary if the degree is equal to or higher than a certain level (Yes in step St7), and judges that support is unnecessary if the degree is less than the certain level (No in step St7).

[0057] If support is not required (No in step St7), the operation of step St13 is performed. If support is required (Yes in step St7), the support request unit 105 requests support from the supporter by displaying on the display 3 that support is required (step St8). The supporter sees the display on the display 3 and heads to the work site.

[0058] Next, the conveying speed control unit 104 compares the work delay with a threshold value TH_H (step St9). The threshold value TH_H is greater than the above threshold value TH_L. If the work delay ≦ TH_H is satisfied (No in step St9), the operation of step St13 is performed. If the work delay > TH_H is satisfied (Yes in step St9), the conveying speed control unit 104 instructs the conveying device 4 to reduce the conveying speed of the conveyor (step St10). After that, the operation of step St13 is performed.

[0059] The symbol Gv indicates the relationship between the work delay (horizontal axis) and the conveyor's transport speed (vertical axis). When the work delay is in the range of 0 to the threshold value TH_H, the transport speed control unit 104 maintains the transport speed at a constant value K. When the work delay exceeds the threshold value TH_L, the support request unit 105 requests support, and when the work delay exceeds the threshold value TH_H, the transport speed control unit 104 controls the transport speed to a value lower than the constant value K. At this time, for example, the correlation between the work delay and the transport speed is a linear relationship. Therefore, when the work delay is increasing before the helper arrives at the work site, the transport speed control unit 104 can suppress the work delay by reducing the transport speed.

[0060] The conveying speed control unit 104 may also change the timing at which the conveying speed starts to decrease depending on the proficiency of the worker. In this case, for example, the conveying speed control unit 104 adjusts the threshold value TH_H depending on the proficiency of the work information 132, as indicated by the arrow m. The conveying speed control unit 104 lowers the threshold value TH_H as the proficiency decreases. The dotted line is an example of the correlation between work delays and the conveying speed when the proficiency is low. As a result, the conveying speed decreases as the worker becomes less skilled, so that work delays can be more effectively suppressed.

[0061] Furthermore, if there is a helper (Yes in step St3), the work discrimination unit 102 discriminates the worker's motion and position in the conveying direction, and the helper's motion and position in the conveying direction (step St11). Next, the delay calculation unit 103 calculates the work delay based on the worker's motion and position in the conveying direction, and the helper's motion and position in the conveying direction (step St12). Thereafter, the operation of step St13 is performed. In this way, if there is a helper (Yes in step St3), the conveying speed control unit 104 does not reduce the conveying speed of the work, so that a decrease in work efficiency due to a decrease in the conveying speed is suppressed. In this way, the work management server 1 operates.

[0062] The above-described embodiment is a preferred example of the present invention, but the present invention is not limited to this embodiment and can be modified in various ways without departing from the scope of the present invention. [Explanation of symbols]

[0063] 1 Work management server (work management device), 10 CPU, 101 skeleton detection unit (detection unit), 102 work discrimination unit (discrimination unit), 103 delay calculation unit (calculation unit), 104 conveyance speed control unit (control unit), 105 support request unit (request unit), 130 detected skeleton data, 131 learned skeleton data, 133 discrimination model

Claims

1. a detection unit that detects, from an image captured of a worker working on a workpiece being transported in a predetermined direction, coordinates of the worker's skeleton in the image; a determination unit that determines a motion and a position in the predetermined direction of the worker from a time change of a coordinate of the skeleton of the worker; a calculation unit that calculates a delay in the work of the worker based on the motion of the worker and the position in the predetermined direction; A request unit that requests a supporter to support the worker in response to the delay. Work management device.

2. A control unit that reduces the conveying speed of the workpiece in response to the delay in the work. The work management device according to claim 1 .

3. The control unit does not reduce the transport speed of the workpiece when the supporter working on the workpiece is captured in the image. The work management device according to claim 2 .

4. The detection unit detects coordinates of skeletons of the worker and the supporter from images captured of the worker and the supporter working on the workpiece, respectively; the determination unit determines a motion and a position in the predetermined direction of the worker from a time change in a skeletal coordinate of the worker, and determines a motion and a position in the predetermined direction of the supporter from a time change in a skeletal coordinate of the supporter, the calculation unit calculates a delay in the work based on a movement of the worker and a position in the predetermined direction, and a movement of the supporter and a position in the predetermined direction; The work management device according to any one of claims 1 to 3.

5. the discrimination unit uses a discrimination model that has been machine-learned through supervised learning, which receives coordinates of a skeleton of the worker in the image as an input and outputs a motion of the worker, and discriminates the motion of the worker based on a change over time in the coordinates of the skeleton. The work management device according to any one of claims 1 to 3.

Images