Work recording system and work recording method

The work recording system addresses inefficiencies in acquiring images for quality determination by incorporating an imaging unit and control unit to assign pass/fail information, thereby improving learning efficiency and quality control.

WO2025115325A1PCT designated stage expired Publication Date: 2025-06-05MITSUBISHI HEAVY IND LTD
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
PCT/JP2024/030664
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2023-11-28
Filing Date
2024-08-28
Publication Date
2025-06-05

AI Technical Summary

Technical Problem

Existing work recording systems face inefficiencies in acquiring images for machine learning, particularly for determining the quality of work locations, which hinders learning efficiency and quality control.

Method used

A work recording system that includes an imaging unit to capture work location images and a control unit to execute a pass/fail process based on these images, assigning pass/fail information to the images for recording and potential use in machine learning.

Benefits of technology

Enables the acquisition of images with pass/fail information, enhancing the efficiency of quality determination and machine learning processes, while improving the reliability of automatic image determination.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure JP2024030664_05062025_PF_FP_ABST
    Figure JP2024030664_05062025_PF_FP_ABST
Patent Text Reader

Abstract

Provided is a work recording system that records work performed on a work location by a worker, and that comprises: an imaging unit that captures an image of the work location to acquire a work image; and a control unit that executes quality processing for determining the quality of the work location on the basis of the work image. The control unit assigns quality information to the work image acquired by the imaging unit and records the work image to which the quality information has been assigned.
Need to check novelty before this filing date? Find Prior Art

Description

Work recording system and work recording method

[0001] The present disclosure relates to a work recording system and a work recording method.

[0002] A system is known in the art that provides a worker with images of past work results related to a current work item via a wearable device worn on the worker's head (see, for example, Patent Literature 1). In this system, images captured by an imaging unit are acquired as work results.

[0003] JP 2018-194914 A

[0004] An example of the work result is the quality of the work area where a worker has worked. The quality of the work area is guaranteed by the worker checking and inspecting the work area, and quality control is also performed by recording images of the work area. Incidentally, the quality of the work area can be guaranteed by assessing the images of the work area. In this case, images of the work area with guaranteed quality are required as training data for machine learning to perform image assessment. In this case, images must be acquired for machine learning, which reduces learning efficiency.

[0005] Therefore, an object of the present disclosure is to provide a work recording system and a work recording method that can record maintenance work and acquire images used for determining whether the work is good or bad.

[0006] The work recording system disclosed herein is a work recording system that records work performed by workers at a work location, and includes an imaging unit that images the work location and acquires work images, and a control unit that executes pass / fail processing to determine the pass / fail quality of the work location based on the work images, and the control unit assigns pass / fail information to the work images acquired by the imaging unit and records the work images with the pass / fail information assigned.

[0007] Furthermore, the work recording method disclosed herein is a work recording method executed by a work recording system that records work performed by workers at a work location, and the work recording system includes an imaging unit that images the work location and acquires a work image, and a control unit that executes pass / fail processing to determine whether the work location is pass / fail based on the work image, and the control unit assigns pass / fail information to the work image acquired by the imaging unit and records the work image with the pass / fail information assigned.

[0008] According to the present disclosure, it is possible to obtain images used for determining whether a product is good or bad, along with recording maintenance work.

[0009] Fig. 1 is a diagram of a work recording system according to this embodiment. Fig. 2 is an explanatory diagram showing the functions of the work recording system according to this embodiment. Fig. 3 is a flowchart relating to a work recording method according to this embodiment. Fig. 4 is a diagram of a work recording screen.

[0010] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the drawings. However, this disclosure is not limited to these embodiments. Furthermore, the components in the following embodiments include those that are easily replaceable by those skilled in the art, or those that are substantially identical. Furthermore, the components described below can be combined as appropriate, and when there are multiple embodiments, the respective embodiments can also be combined.

[0011] [Present Embodiment] A work recording system 10 according to this embodiment is a system for recording the work content of aircraft maintenance work. The work recording system 10 records the work content at a work location by a worker. The work recording system 10 may be used as a system for educating and training on maintenance work, or as a system for supporting the execution of maintenance work. The work recording system 10 will be described with reference to FIG. 1 .

[0012] (Work Recording System) FIG. 1 is a diagram of a work recording system according to this embodiment. As shown in FIG. 1, the work recording system 10 includes a control unit 31, a storage unit 32, a display unit 33, an input unit 34, and an imaging unit 35. The work recording system 10 may be a wearable terminal or a stationary terminal, and is not particularly limited. A worker terminal 39 is also connected to the work recording system 10 via a communication network 37. Although not shown, the worker terminal 39 includes a display unit and an input unit, similar to the work recording system 10, and enables the worker terminal 39 to input information for recording work.

[0013] The control unit 31 includes an integrated circuit such as a CPU (Central Processing Unit). The control unit 31 executes data processing for recording maintenance work. The storage unit 32 is any storage device such as a semiconductor storage device or a magnetic storage device. The storage unit 32 stores images captured by the imaging unit 35, information generated by various processes, etc. The display unit 33 is a display device such as a liquid crystal display. The input unit 34 is an input device such as a keyboard and a mouse. The imaging unit 35 is a camera, for example.

[0014] The work recording system 10 presents information about the planned maintenance work to the worker via the display unit 33. Furthermore, the work recording system 10 generates information about the progress of the maintenance work and information about the results of the maintenance work by receiving work record information about the maintenance work input by the worker. The work recording system 10 then outputs the generated information about the results of the maintenance work to a work record database (not shown).

[0015] Next, an overview of the functions of the work recording system 10 will be described with reference to FIG. 2. FIG. 2 is an explanatory diagram functionally illustrating the work recording system according to this embodiment. As shown in FIG. 2, the work recording system 10 acquires work request data, which is data on maintenance work to be performed. The work request data is information for supporting maintenance work, and is resource information that associates the work procedure for the maintenance work, the location required for the maintenance work, the equipment required for the maintenance work, the parts required for the maintenance work, the personnel required for the maintenance work, and the work time required for the maintenance work.

[0016] Upon acquiring the work request data, the work recording system 10 provides support for the maintenance work based on the work request data. The work recording system 10 presents the maintenance work procedure to the worker via the display unit 33. The work recording system 10 also generates work record data based on the information input by the worker and outputs the generated work record data to the work record database.

[0017] Furthermore, while supporting maintenance work, the work recording system 10 requests the worker to record images of the work location where the maintenance work is being performed. The work recording system 10 performs quality control processing on the work images acquired by the imaging unit 35 to determine whether the work location is quality control or not. The work recording system 10 then generates work record data including the work images after quality control determination. The work record data is information on the actual resources after the maintenance work has been performed, and has the same data format as the work request data. The work record data also includes know-how information entered by the worker.

[0018] (Work Recording Method) Next, a work recording method according to this embodiment will be described with reference to Fig. 3. Fig. 3 is a flowchart illustrating the work recording method according to this embodiment. The control unit 31 of the work recording system 10 executes a pass / fail process to determine whether the work location is pass / fail while supporting maintenance work.

[0019] The control unit 31 of the work recording system 10 determines whether or not there is an accumulation of work images to which pass / fail information has been assigned as pass / fail processing (step S1). Specifically, in step S1, the control unit 31 determines whether or not there are an enough number of work images accumulated to serve as judgment material for executing the automatic image judgment processing described below, sufficient for image judgment to be performed. If the control unit 31 determines that there is an accumulation of work images (step S1: Yes), it determines whether or not there is reliability in the image judgment processing (step S2). In step S2, the control unit 31 determines the reliability by determining whether or not the number of scores for the image judgment described below is equal to or greater than a predetermined score. If the control unit 31 determines that there is reliability (step S2: Yes), it proceeds to the automatic image judgment processing (steps S13 to S32) in which the images are automatically judged. On the other hand, if the control unit 31 determines that there is no accumulation of work images (step S1: No) or that there is no reliability (step S2: No), it proceeds to manual image determination processing (steps S3 to S8) in which an operator manually determines the images.

[0020] When the control unit 31 shifts to the manual image determination process, it requests the worker to acquire a work image of the work location (step S3). In step S3, the control unit 31 displays a work record screen 50 for acquiring the work image on the display unit 33.

[0021] Here, the work record screen displayed on the display unit 33 will be described with reference to Fig. 4. Fig. 4 is a diagram of the work record screen. As shown in Fig. 4, the work record screen 50 displays images captured by the imaging unit 35. The work record screen 50 also displays a pass / fail judgment button 51 for making a pass / fail judgment on the work area, and a fail judgment button 52 for making a fail judgment on the work area. The pass / fail judgment button 51 and the fail judgment button 52 are buttons for judging the pass / fail of the work area, and are also buttons for capturing images.

[0022] When the control unit 31 displays the work record screen 50 in step S3, it accepts an input operation for a pass / fail judgment (step S4). Then, the control unit 31 judges whether the work area is pass or not based on the operation of the pass / fail judgment button 51 or the fail judgment button 52 input via the input unit 34 (step S5). If the control unit 31 determines in step S5 that the pass judgment button 51 has been operated, it judges the work area to be pass (step S5: Yes). On the other hand, if the control unit 31 determines that the fail judgment button 52 has been operated, it judges the work area to be fail (step S5: No).

[0023] If the control unit 31 determines in step S5 that the work area is good, it acquires a work image at the time the good judgment button 51 is operated, and assigns good information to the acquired work image and records it in the storage unit 32 (step S6). On the other hand, if the control unit 31 determines in step S5 that the work area is bad, it acquires a work image at the time the bad judgment button 52 is operated, and assigns bad information to the acquired work image and records it in the storage unit 32 (step S7).

[0024] After executing steps S6 and S7, the control unit 31 adds points to the score of the image judgment so as to increase the reliability of the automatic image judgment (step S8). After executing step S8, the control unit 31 ends the manual image judgment process.

[0025] On the other hand, when the control unit 31 transitions to the automatic image judgment process after executing step S2, it requests the worker to acquire work images of the work area (step S13). The control unit 31, triggered by the request in step S13, acquires the work image by having the imaging unit 35 capture an image of the work area and performing an imaging process using spatial reference points (step S14). In step S14, the control unit 31 uses the spatial reference points to capture an image so that the imaging range, such as the angle of view of the work area, is at a predetermined position. After acquiring the work image in step S14, the control unit 31 automatically judges the quality of the work area based on the acquired work image (step S15). In step S15, the control unit 31 performs quality judgment of the acquired work image based on the accumulated work images with quality information, using, for example, a learning model that performs machine learning using the work images with quality information as training data.

[0026] If the control unit 31 determines in step S15 that the work area is acceptable (step S15: Yes), it requests the worker to confirm whether the acceptability determination is correct, that is, whether the work area is acceptable (step S16). In step S16, the worker inputs whether the work area is acceptable or not through the input unit 34. The control unit 31 then determines whether the work area is acceptable or not based on the input from the input unit 34. If the control unit 31 determines that the work area is acceptable based on the confirmation from the worker (step S16: Yes), it adds acceptability information to the acquired work image and records it in the storage unit 32 (step S17).

[0027] After executing step S17, the control unit 31 determines that the result of the automatic image determination matches the result of the confirmation by the operator, and adds points to the score of the image determination so as to increase the reliability of the automatic image determination (step S18). After executing step S18, the control unit 31 ends the automatic image determination process.

[0028] On the other hand, in step S16, if the control unit 31 determines that the work area is not suitable based on confirmation from the worker (step S16: No), it adds non-suit information to the acquired work image and records it in the memory unit 32 (step S21).

[0029] After executing step S21, the control unit 31 determines that the result of the automatic image judgment differs from the result of the confirmation by the worker, and deducts points from the score of the image judgment so as to lower the reliability of the automatic image judgment (step S22). After executing step S22, the control unit 31 ends the automatic image judgment process.

[0030] Furthermore, if the control unit 31 determines in step S15 that the work area is not acceptable (step S15: No), it notifies the worker that the work image has been classified and determined to be abnormal (step S25). After executing step S25, the control unit 31 requests the worker to confirm whether the pass / fail determination is correct, that is, whether the work area is not acceptable (step S26). If the control unit 31 determines that the work area is not acceptable based on the confirmation from the worker (step S26: Yes), it assigns non-acceptance information to the acquired work image and records it in the storage unit 32 (step S27).

[0031] After executing step S27, the control unit 31 determines that the result of the automatic image determination matches the result of the confirmation by the operator, and adds points to the score of the image determination so as to increase the reliability of the automatic image determination (step S28). After executing step S28, the control unit 31 ends the automatic image determination process.

[0032] On the other hand, in step S26, if the control unit 31 determines that the work area is not bad, i.e., good, based on confirmation from the worker (step S26: No), it adds good information to the acquired work image and records it in the memory unit 32 (step S31).

[0033] After executing step S31, the control unit 31 determines that the result of the automatic image judgment differs from the result of the confirmation by the worker, and deducts points from the score of the image judgment so as to lower the reliability of the automatic image judgment (step S32). After executing step S32, the control unit 31 ends the automatic image judgment process.

[0034] As described above, the work recording system 10 and the work recording method described in this embodiment can be understood, for example, as follows.

[0035] The work recording system 10 according to the first aspect is a work recording system 10 that records work performed by workers at a work location, and includes an imaging unit 335 that images the work location and acquires work images, and a control unit 31 that executes pass / fail processing to determine whether the work location is pass / fail based on the work images, and the control unit 31 assigns pass / fail information to the work images acquired by the imaging unit 35 and records the work images with the pass / fail information assigned.

[0036] With this configuration, work images with pass / fail information can be recorded, allowing the locations of maintenance work to be recorded. Furthermore, since the work images have pass / fail information attached, they can be used as training data for image evaluation. This allows images to be acquired for pass / fail evaluation, along with records of maintenance work.

[0037] As a second aspect, in the work recording system 10 relating to the first aspect, if the control unit 31 determines that the accumulation of recorded work images is not enough to perform a pass / fail judgment on the work area, it requests the worker to make a pass / fail judgment.

[0038] According to this configuration, it is possible to obtain a work image to which pass / fail information has been added by a worker while suppressing erroneous pass / fail judgment.

[0039] As a third aspect, in the work recording system 10 relating to the first or second aspect, if the control unit 31 determines that the accumulation of recorded work images is an accumulation that allows for a pass / fail judgment of the work area to be performed, it performs a pass / fail judgment based on the accumulated work images.

[0040] According to this configuration, since a sufficient number of work images to which pass / fail information has been added are accumulated, pass / fail judgment can be performed with high accuracy.

[0041] As a fourth aspect, the work recording system 10 relating to any one of the first to third aspects further comprises a display unit 33 that displays the image captured by the imaging unit 35, and an input unit 34 for inputting a pass / fail judgment of the work area, the input unit 34 having a pass judgment button 51 for making a pass / fail judgment of the work area, and a fail judgment button 52 for making a fail judgment of the work area, and the control unit 31 acquires the work image at the time of inputting the pass / fail judgment button 51 or the fail judgment button 52, and assigns the pass / fail information to the work image.

[0042] According to this configuration, by operating the pass / fail judgment button 51 or the fail judgment button 52, a work image to which pass / fail information has been added can be easily obtained.

[0043] A work recording method according to a fifth aspect is a work recording method executed by a work recording system 10 that records work performed by workers at a work location. The work recording system 10 includes an imaging unit 35 that images the work location and acquires work images, and a control unit 31 that executes pass / fail processing to determine whether the work location is pass / fail based on the work images. The control unit 31 assigns pass / fail information to the work images acquired by the imaging unit 35 and records the work images with the pass / fail information assigned.

[0044] With this configuration, work images with pass / fail information can be recorded, allowing the locations of maintenance work to be recorded. Furthermore, since the work images have pass / fail information attached, they can be used as training data for image evaluation. This allows images to be acquired for pass / fail evaluation, along with records of maintenance work.

[0045] 10 Work recording system 31 Control unit 32 Storage unit 33 Display unit 34 Input unit 35 Imaging unit 37 Communication network 39 Worker terminal

Claims

1. A work recording system that records work performed at a work location by a worker, comprising: an imaging unit that images the work location and obtains work images; and a control unit that executes pass / fail processing to determine the pass / fail quality of the work location based on the work images, wherein the control unit assigns pass / fail information to the work images acquired by the imaging unit, and records the work images with the pass / fail information assigned to them.

2. A work recording system as described in claim 1, wherein the control unit, when determining that the accumulation of recorded work images is not sufficient to perform a pass / fail judgment on the work area, requests the worker to make a pass / fail judgment.

3. A work recording system as described in claim 1, wherein the control unit, when determining that the accumulation of recorded work images is an accumulation that allows for a pass / fail judgment of the work area to be performed, performs a pass / fail judgment based on the accumulated work images.

4. A work recording system as described in claim 1, further comprising: a display unit which displays an image captured by the imaging unit; and an input unit which inputs a pass / fail judgment of the work area, wherein the input unit has a pass judgment button which makes a pass / fail judgment of the work area, and a fail judgment button which makes a fail judgment of the work area, and wherein the control unit acquires the work image at the time of inputting the pass / fail judgment button or the fail judgment button, and adds the pass / fail information to the work image.

5. A work recording method executed by a work recording system that records work performed at a work location by a worker, the work recording system comprising: an imaging unit that images the work location and obtains a work image; and a control unit that executes pass / fail processing to determine the pass / fail quality of the work location based on the work image, wherein the control unit assigns pass / fail information to the work image acquired by the imaging unit, and records the work image with the pass / fail information assigned to it.

Citation Information

Patent Citations

  • System, method, and program

    JP2018194914A

  • Work support device and work support program

    JP2012198644A

  • Operation confirmation system

    JP2015114676A

  • Work support system, kitchen support system, work support method, and program

    JP2019075009A

  • Determination system and determination method

    JP2022183771A