Work cycle analysis system

Abstract

To easily grasp work status.SOLUTION: A work cycle analysis system 100 comprises: an imaging unit 41 capable of imaging a work area where a work process is performed; an identification unit 32 for identifying a specific component in an image captured by the imaging unit 41; and a work process determination unit for determining the start or end of the work process included in a work cycle based on a change in a pixel ratio within the image of the specific component identified by the identification unit 32.SELECTED DRAWING: Figure 9

Description

[Technical Field]

[0001] The present invention relates to a work cycle analysis system. [Background technology]

[0002] Patent Document 1 discloses a work analysis system that detects the movements of a worker or a work machine and determines the work performed within a work area based on the detected movements. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2020-194243 Summary of the Invention [Problem to be solved by the invention]

[0004] The work analysis system described in Patent Document 1 is required to detect multiple work machines and workers and grasp the operating status of each of the detected work machines, etc., in order to determine the work performed within the work area. Therefore, to accurately determine the work status, it is necessary to improve, for example, the accuracy of recognizing the type of work machine and the accuracy of determining the operating status of the work machine. However, both of these measures increase the computational load, making it difficult to grasp the work status. Furthermore, to collect accurate data, it is necessary to place detectors that detect movement in appropriate locations and have people constantly check whether movement is being detected correctly, which increases the human workload required for data collection and may increase operating costs.

[0005] An object of the present invention is to easily grasp the working situation. [Means for solving the problem]

[0006] The present invention is a work cycle analysis system that analyzes the work cycle of a building constructed by carrying out a work cycle consisting of a plurality of work processes, and includes an imaging unit that can capture an image of a work area where the work processes are carried out, an identification unit that identifies specific components in the image captured by the imaging unit, and a work process determination unit that determines the start or end of a work process included in the work cycle based on a change in the pixel ratio in the image of the specific component identified by the identification unit. [Effects of the Invention]

[0007] According to the present invention, the work situation can be easily grasped. [Brief explanation of the drawings]

[0008] [Figure 1] 1 is a schematic diagram illustrating an example of a work site where a work process is analyzed by a work cycle analysis system according to an embodiment of the present invention. [Figure 2A] 2 is an enlarged view showing a part A in FIG. 1, and is a diagram for explaining the work process at the work site shown in FIG. 1. FIG. [Figure 2B] FIG. 2B is a diagram for explaining a work process subsequent to the work process shown in FIG. 2A. [Figure 2C] FIG. 2C is a diagram for explaining a work process subsequent to the work process shown in FIG. 2B. [Figure 2D] FIG. 2D is a diagram for explaining a work process subsequent to the work process shown in FIG. 2C. [Figure 3] 1 is a block diagram of a work cycle analysis system according to an embodiment of the present invention. [Figure 4] 1 is a flowchart showing a procedure for determining a work process by a work cycle analysis system according to an embodiment of the present invention. [Figure 5A] 2B is a diagram showing an image captured by an imaging unit in the work process shown in FIG. 2A. FIG. [Figure 5B] 2C is a diagram showing an image captured by an imaging unit in the work process shown in FIG. 2B. FIG. [Figure 5C]2D is a diagram showing an image captured by an imaging unit in the work process shown in FIG. 2C. FIG. [Figure 6] 10 is a graph showing changes in the area ratio of the guide rail in an image captured by an imaging unit. [Figure 7A] 2E is a diagram showing an image captured by an imaging unit in the work process shown in FIG. 2D. FIG. [Figure 7B] 7B is a diagram showing an image captured by an imaging unit in a work step subsequent to the work step shown in FIG. 7A. FIG. [Figure 7C] 7C is a diagram showing an image captured by an imaging unit in a work step subsequent to the work step shown in FIG. 7B. FIG. [Figure 8] 10 is a graph showing changes in the area ratio of reinforcing bars and the like in an image captured by an imaging unit. [Figure 9] 1 is a diagram for explaining determination of a work process by a work cycle analysis system according to an embodiment of the present invention. FIG. [Figure 10] This is a graph showing the change in the area ratio of reinforcing bars, etc. in the image captured by the imaging unit, and is a figure for explaining the determination of the start time of the pouring process. [Figure 11] 10A and 10B are diagrams for explaining a method for identifying a person from an image captured by an imaging unit. [Figure 12] 10 is a graph showing the change in the area ratio of reinforcing bars and the like in an image captured by an imaging unit and the number of people identified from the image. DETAILED DESCRIPTION OF THE INVENTION

[0009] Hereinafter, a work cycle analysis system according to an embodiment of the present invention will be described with reference to the drawings.

[0010] The work cycle analysis system 100 according to an embodiment of the present invention is a system for analyzing the work cycle of a building constructed by carrying out a work cycle consisting of multiple work processes, particularly a building constructed by repeatedly carrying out a work cycle, and automatically determines the start and end times of the work processes included in the work cycle.

[0011] In the following, a description will be given of a case where the structure to which the work cycle analysis system 100 is applied is a bridge 1 constructed by the cantilever erection method, as shown in Fig. 1. Note that the structure to which the work cycle analysis system 100 is applied is not limited to the bridge 1, and may be any structure that is constructed by carrying out a work cycle consisting of a plurality of work processes, such as a structure such as a dam or a tunnel, or a building such as an apartment building.

[0012] As shown in Figure 1, the cantilever erection method involves moving a mobile work vehicle 10 from the top end 2a of each pier 2 symmetrically around the pier 2 along the bridge axis direction while sequentially constructing blocks 3a to 3e separated by a predetermined construction block length (for example, 2 to 8 meters) in the bridge axis direction, gradually extending the main girder 3 in the bridge axis direction. Figure 1 shows the state in which, after the first block 3a has been constructed on the top end 2a of the pier 2, the second block 3b, third block 3c, fourth block 3d, and fifth block 3e have been constructed so that they extend in order.

[0013] As shown in Fig. 2A, the mobile work vehicle 10 is mainly composed of a truss-structured main body frame 12, a pair of upper beams 13 that are attached to the top of the main body frame 12 and extend in a direction perpendicular to the bridge axis direction, multiple suspension members 14 that extend downward from both ends of each upper beam 13, a pair of lower beams 15 that are supported by the suspension members 14 and extend in a direction perpendicular to the bridge axis direction, and a work platform 16 that is attached to the lower beams 15. Fig. 2A is an enlarged view of part A in Fig. 1.

[0014] The mobile work vehicle 10 also has multiple travel motors (not shown) attached to a pair of base frames 12a located at the bottom ends of the main frame 12, and by driving these travel motors, it is possible to move along a pair of guide rails 18 provided on the main girder 3. The pair of guide rails 18 are components for guiding the mobile work vehicle 10 to the work area where a new block will be installed, and are installed on the main girder 3 along the bridge axis direction.

[0015] Next, a general work cycle of the cantilever erection method will be described with reference to Figures 2A to 2D. Figure 2A shows the state after construction of the fifth block 3e has been completed.

[0016] As shown in Figure 2A, once construction of the fifth block 3e is completed, a relocation process is carried out to relocate the mobile work vehicle 10 so that a new block (sixth block 3f) can be constructed by extending it from the fifth block 3e in the bridge axis direction.

[0017] In the process of relocating the mobile work vehicle 10, first, as shown in Figure 2B, a pair of guide rails 18 is pulled out onto the completed fifth block 3e. While the guide rails 18 are being pulled out, the mobile work vehicle 10 is left on the fourth block 3d with no load being applied to the guide rails 18.

[0018] Once the pair of guide rails 18 have been fixed onto the fifth block 3e, the travel motor (not shown) is driven to move the mobile work vehicle 10 along the guide rails 18, and as shown in Figure 2C, the mobile work vehicle 10 is moved onto the fifth block 3e. As a result, the work platform 16 of the mobile work vehicle 10 protrudes in the direction of the bridge axis from the fifth block 3e, for which construction has been completed.

[0019] Once the work platform 16 is in a state of projecting from the fifth block 3e in the bridge axis direction, as shown in Fig. 2D, the construction process of the projecting section is carried out on the work platform 16, in which the sixth block 3f is constructed so that it projects from the fifth block 3e. The construction process of the projecting section is a general reinforced concrete framework construction process, and is carried out in the following order, as will be described later: formwork installation process, reinforcement process, concrete pouring process, concrete curing process, and formwork dismantling process.

[0020] Once construction of the sixth block 3f is complete and the formwork is dismantled, the relocation process of the mobile work vehicle 10 is carried out again.

[0021] In this way, the bridge 1 constructed by the cantilever erection method is gradually constructed by repeatedly carrying out a work cycle that includes the process of relocating the mobile work vehicle 10 and the process of constructing the overhanging portion.

[0022] In addition, the relocation process of the mobile work vehicle 10 includes an extension process of extending the guide rails 18, a fixing process of fixing the extended guide rails 18, and a moving process of moving the mobile work vehicle 10 along the guide rails 18, and since the bridge 1 is constructed gradually as the relocation process of the mobile work vehicle 10 is repeatedly carried out, the relocation process of the mobile work vehicle 10 can be considered a work cycle in the construction of the bridge 1.

[0023] Similarly, the construction process for the overhanging section includes the formwork installation process, reinforcement process, concrete pouring process, etc., and since bridge 1 is gradually constructed as the construction process for the overhanging section is repeatedly carried out, the construction process for the overhanging section can be considered as a work cycle in the construction of bridge 1.

[0024] Here, in order to improve the construction efficiency of structures that are constructed by repeatedly carrying out a work cycle consisting of multiple work processes, such as the bridge 1 described above, it is necessary to accurately analyze the time required to carry out each work process that makes up the work cycle, and understand the work situation.

[0025] To grasp the work time, for example, it is possible to have on-site workers measure the time required to perform each work process, but this would mean that the start and end times of each work process would be determined subjectively, which could lead to variations depending on the worker measuring the time, making it difficult to accurately grasp the work situation.

[0026] Therefore, in this embodiment, in order to quantitatively and objectively grasp the work status of each work process, the time when each work process was carried out is automatically determined based on images taken of the work area where the work process is carried out.

[0027] Specifically, in order to analyze the work cycles that are repeatedly performed when constructing a bridge 1 (structure), the work cycle analysis system 100 includes a data collection unit 40 that collects data on the work area where the work process is carried out, and a cycle analysis unit 30 that analyzes the work cycles based on the data transmitted from the data collection unit 40, as shown in FIG. 3.

[0028] The data collection unit 40 has an imaging unit 41 capable of capturing images of the work area, an overhang sensor 42 capable of detecting the overhang length of the guide rail 18, and a data transmission unit 44 that transmits the image data captured by the imaging unit 41 and the distance data detected by the overhang sensor 42 to the cycle analysis unit 30.

[0029] The imaging unit 41 is a so-called digital camera, and is placed in a position where it can capture an image of the work area where work to construct the blocks that make up the main girder 3 is carried out.

[0030] Specifically, since the area above the work floor 16 is the work area where work to construct the blocks that make up the main girder 3 is carried out, the imaging unit 41 is fixed to a relatively high position on the mobile work vehicle 10 with the imaging direction facing downward, as shown in Figure 2A, for example, so that the entire work floor 16 can be imaged from above.

[0031] In order to ensure that the work area is sufficiently contained within the imaging range, it is preferable to place the imaging unit 41 vertically above the center of the work area, but the location where the imaging unit 41 is placed is not limited to this and it can be placed anywhere as long as almost the entire work area can be contained within the imaging range.

[0032] The overhang sensor 42 is a distance sensor such as a laser displacement sensor that can measure the length of the guide rail 18 pulled out from the mobile work vehicle 10 in the bridge axis direction, i.e., the distance between the tip of the guide rail 18 and the base frame 12a of the mobile work vehicle 10, and is attached to the tip of the guide rail 18, for example, as shown in Figure 2A.

[0033] As will be described later, the distance data detected by the overhang sensor 42 is used auxiliary when analyzing a work cycle, and therefore the overhang sensor 42 is not an essential component of the work cycle analysis system 100. The overhang sensor 42 may also be used to control the travel motor of the mobile work vehicle 10 when moving the mobile work vehicle 10 along the guide rails 18.

[0034] The data transmission unit 44 is a short-range wireless communication device such as Wi-Fi (registered trademark) or Bluetooth (registered trademark), or a general wireless communication device capable of sending and receiving data via an internet line, and as described above, transmits the image data captured by the imaging unit 41 and the distance data detected by the extension sensor 42 to the cycle analysis unit 30. Note that the data collection unit 40 may be connected to the cycle analysis unit 30 via a wired connection such as a LAN cable, in which case the image data captured by the imaging unit 41 is transmitted to the cycle analysis unit 30 via a wired connection.

[0035] The image data captured by the imaging unit 41 and the distance data detected by the protrusion sensor 42 may be transmitted to the cycle analysis unit 30 at any time, or may be transmitted at predetermined regular intervals, and the data collection unit 40 may be provided with a memory unit capable of temporarily storing the image data and distance data.

[0036] A data collection unit 40 configured in this manner is provided for each work area, i.e., for each area where a mobile work vehicle 10 is located. In other words, multiple data collection units 40 are provided for one cycle analysis unit 30.

[0037] The cycle analysis unit 30 includes an analysis unit 31 that analyzes data received from the data collection unit 40 via a data reception unit 35 , and a display unit 36 ​​that displays the results of the analysis by the analysis unit 31 .

[0038] The analysis unit 31 is composed of a computer equipped with a CPU (Central Processing Unit), ROM (Read Only Memory), RAM (Random Access Memory), and I / O interface (Input / Output Interface). The RAM stores data for CPU processing, the ROM stores CPU control programs and the like in advance, and the I / O interface is used for inputting and outputting information to and from devices such as the display unit 36 ​​and data receiving unit 35 connected to the analysis unit 31. The display unit 36 ​​is a general monitor device, and the data receiving unit 35 is a wireless communication device similar to the data transmitting unit 44 described above.

[0039] The analysis unit 31 is installed, for example, in a monitoring room set up near the work site to monitor the construction status of the bridge 1 (structure) or in a remote monitoring room set up in a remote location far away from the work site. Note that the cycle analysis unit 30 may be placed near the data collection unit 40 to implement edge computing, in which case only the necessary data from the data collected by the data collection unit 40 and the results calculated by the cycle analysis unit 30 is sent to a predetermined server or the like.

[0040] The analysis unit 31 has an identification unit 32 that stores the learning results of previously performed machine learning and identifies specific components in an image captured by the imaging unit 41 based on the learning results, and a work process determination unit 33 that determines the start or end of a work process included in a work cycle based on a change in the area proportion of the specific component identified by the identification unit 32 in the image.

[0041] The identification unit 32 and the work process determination unit 33 are shown as virtual units that represent some of the functions of a general personal computer used as the analysis unit 31, and do not mean that they physically exist. Furthermore, the identification unit 32 and the work process determination unit 33 do not need to be provided within one analysis unit 31, i.e., one personal computer, but may be provided in different servers, for example, and configured to send and receive data to and from each other via wired or wireless communication.

[0042] The recognition unit 32 stores the learning results of machine learning that have been performed in advance, and based on the stored learning results, the recognition unit 32 automatically identifies and extracts a specific component from the image data P captured by the imaging unit 41. A specific component is a component that is used in a specific work process among multiple work processes that make up a work cycle, and specific examples will be described later.

[0043] The learning results stored in the recognition unit 32 include the results of machine learning that was conducted in advance using image data of the work area actually captured during each work process and image data created assuming each work process as training data.

[0044] As a machine learning technique, for example, a technique called semantic segmentation is used, which is a type of segmentation using deep learning, and identifies the type of object displayed in each pixel and detects the pixel area where the object to be identified exists.In order to be able to identify specific components from the image data captured by the imaging unit 41, learning is performed to detect the pixel area where a specific component exists.

[0045] The machine learning method is not limited to semantic segmentation, and any method that can extract the area of ​​a specific component occupying image data can be used. For example, other well-known methods using deep learning, such as instance segmentation and panoptic segmentation, can be used.

[0046] Furthermore, when the area occupied by a specific component in the image data captured by the imaging unit 41 is identified by the identification unit 32 in which the above-mentioned learning results are stored, the image data may be subjected to known image processing in advance, such as processing to improve the accuracy of object identification, for example, binarization processing or processing to process the image data three-dimensionally using SfM (Structure from Motion).

[0047] The work process determination unit 33 calculates the pixel ratio of the number of pixels identified by the identification unit 32 as having the presence of a specific component to the total number of pixels in the image data, i.e., the area ratio of the area in which the specific component appears to the total area of ​​the image data, and determines the start and end of a work process included in a work cycle based on changes in the calculated area ratio. Note that the determination of the start and end of a work process may be made based on changes in the pixel ratio, and may also be made based on, for example, changes in the length in a predetermined direction of a figure formed by pixels identified as having the presence of a specific component, or changes in the length of one side or diagonal of that figure, instead of changes in the area ratio.

[0048] The work process determination unit 33 stores a threshold value for determining the start or end of a work process or an arithmetic formula for calculating the threshold value, and determines that the start or end of a work process is the point in time when the area ratio (pixel ratio) of a specific component in the image data exceeds a predetermined threshold value or falls below a predetermined threshold value.

[0049] Furthermore, if the work process determination unit 33 has already determined the start time or end time of the same work process, it calculates the time elapsed from the already determined start time or end time to the newly determined start time or end time as the cycle time of the work cycle.

[0050] The determination results and the calculated results thus obtained by the work process determination unit 33 are displayed on the display unit 36. The determination results may be transmitted to a predetermined server or the like via a communication unit (not shown) and disclosed to a user who accesses the server or the like.

[0051] Next, a work cycle analysis method performed by the above-described work cycle analysis system 100 will be described with reference to the flow chart shown in FIG.

[0052] First, in step S11, the imaging unit 41 images the work area at a predetermined time interval (for example, every few minutes to every hour), and in the subsequent step S12, the image data captured by the imaging unit 41 is transmitted to the cycle analysis unit 30. Note that the frequency with which the image data is transmitted may be every time the imaging unit 41 images the work area, or may be at a predetermined time interval (for example, every few hours to every day) that is sufficiently longer than the time interval at which the work area is imaged, or may be when the work scheduled for one day at the work site is completed.

[0053] When the cycle analysis unit 30 receives the image data, it causes the identification unit 32 to identify a specific component in the image data (step S13).

[0054] Once a specific component is identified in the image data, the process proceeds to step S14, where the cycle analysis unit 30 causes the work process determination unit 33 to calculate the area ratio (pixel ratio) of the specific component in the image data and compare the calculated area ratio with a predetermined threshold. As described above, the work process determination unit 33 determines that the start or end of a work process included in the work cycle occurs when the calculated area ratio exceeds or falls below the predetermined threshold.

[0055] The result determined by the work process determination unit 33 in this way is output to the display unit 36 ​​etc. in the following step S15. The result determined by the work process determination unit 33 also includes the above-mentioned cycle time.

[0056] Once the determination result is output in this manner, the control flow temporarily ends.

[0057] By checking the start and end times of the work process and the cycle time of the work cycle including the work process displayed on the display unit 36, etc., the manager who manages the progress of the work can quantitatively and objectively grasp the work status of each work process.

[0058] Next, with reference to Figures 5 and 6, we will explain the case where a relocation process for a mobile work vehicle 10, which is made up of multiple work processes, is a work cycle analyzed by the above-mentioned work cycle analysis system 100. Figures 5A to 5C are diagrams showing image data of the work area captured by the imaging unit 41, with the lower side in the diagram being the overhang direction of the main girder 3, i.e., the work area where new blocks will be constructed sequentially. Also, Figure 6 is a graph showing the results of the work cycle analysis.

[0059] 5A is an image captured by the imaging unit 41 of the state shown in FIG. 2A above, that is, the state after construction of the fifth block 3e is completed and immediately before the start of the relocation process of the mobile work vehicle 10. For this reason, the subject of the first image data P11 is almost entirely the top surface of the fifth block 3e.

[0060] 5B is an image of the state shown in FIG. 2B described above, that is, the state in which the pair of guide rails 18 is pulled out onto the completed fifth block 3e, captured by the imaging unit 41. Therefore, the subject of the second image data P12 is the upper surface of the fifth block 3e and the pair of guide rails 18.

[0061] 5C is an image captured by the imaging unit 41 of the state shown in Fig. 2C above, that is, the state in which the travel motor of the mobile work vehicle 10 is driven to move the mobile work vehicle 10 along the pair of guide rails 18 extended above the fifth block 3e. For this reason, the subject of the third image data P13 is almost entirely the upper surface of the work floor 16 of the mobile work vehicle 10.

[0062] In this way, in the relocation process of the mobile work vehicle 10, the pair of guide rails 18 are components used in specific work processes among the multiple work processes that make up the work cycle, specifically, the pull-out process of pulling out the guide rails 18, the fixing process of fixing the pulled-out guide rails 18, and the moving process of moving the mobile work vehicle 10 along the guide rails 18.

[0063] Therefore, in order to identify a pair of guide rails 18 as components corresponding to a specific component from image data P such as those shown in Figures 5A to 5C, the identification unit 32 stores the results of machine learning that was previously performed using, for example, images of a pair of guide rails 18 with different pull-out lengths as training data in order to identify the shape and position of the pair of guide rails 18.

[0064] Therefore, as shown by hatching in FIGS. 5A to 5C, the area in each image data in which the pair of guide rails 18 is captured is identified by the identifying unit 32.

[0065] The size of the area in which the pair of guide rails 18 is reflected in each image data gradually increases as the pair of guide rails 18 is pulled out onto the fifth block 3e from the state shown in Figure 5A, and reaches its maximum in the state shown in Figure 5B.

[0066] Furthermore, the size of the area in which the pair of guide rails 18 is captured in each image data gradually decreases as the mobile work vehicle 10 moves along the pair of guide rails 18 from the state shown in Figure 5B, and in the state shown in Figure 5C, it reaches its smallest size, equivalent to the state shown in Figure 5A.

[0067] In this way, the size of the area in which a pair of guide rails 18 (specific components) is captured is calculated by the work process determination unit 33 as the proportion of the number of pixels identified as containing a pair of guide rails 18 to the total number of pixels in the image data, i.e., the area in which the pair of guide rails 18 are captured to the total area of ​​the image data, and the change in the area proportion over time is graphed as shown in Figure 6.

[0068] Furthermore, since the area ratio of the guide rail 18 correlates with the detection value of the above-mentioned overhang sensor 42, which can detect the overhang length of the guide rail 18, the change over time in the area ratio of the guide rail 18 may be appropriately interpolated based on the change over time in the detection value of the overhang sensor 42.

[0069] In addition, the work process determination unit 33 has multiple thresholds for determining the start and end of each work process that makes up the relocation process of the mobile work vehicle 10, and determines the start and end of each work process by comparing these thresholds with the determined area ratio.

[0070] Specifically, the point at which the area proportion of the guide rail 18 in the image data reaches its maximum and then falls below the first threshold, i.e., the point at which the area proportion of the guide rail 18 begins to decrease, is considered to be the time when the mobile work vehicle 10 begins to move along the guide rail 18, and is therefore determined to be the start of the movement process of moving the mobile work vehicle 10 along the guide rail 18.

[0071] The first threshold value may be a preset value, but since the absolute value of the area ratio of the guide rail 18 in each work cycle varies to some extent, it is preferable that the first threshold value be calculated according to the maximum and minimum values ​​of the area ratio of the guide rail 18 in each work cycle.For example, the first threshold value is set to a value several percent smaller than the maximum value, with the difference between the maximum and minimum values ​​of the area ratio being 100%.

[0072] Furthermore, the point in time after the point determined to be the start of the movement process when the area ratio of the guide rail 18 falls below a second threshold value that is smaller than the first threshold value, i.e., the point in time when the area ratio of the guide rail 18 reaches approximately its minimum value, is considered to be the point in time when the subject of the image data is almost entirely occupied by the work floor 16 of the mobile work vehicle 10 as the mobile work vehicle 10 moves along the guide rail 18, and is therefore determined to be the end of the movement process of moving the mobile work vehicle 10 along the guide rail 18.

[0073] The second threshold may be a preset value, but similarly to the first threshold, the difference between the maximum and minimum values ​​of the area ratio is set to 100%, and the second threshold is set to a value several percent larger than the minimum value.

[0074] Furthermore, the point at which the area ratio of the guide rail 18 in the image data exceeds the third threshold, i.e., the point at which the area ratio of the guide rail 18 begins to increase, is considered to be the time at which the guide rail 18 begins to be pulled out onto the fifth block 3e where construction has been completed, and is therefore determined to be the start of the pull-out process of pulling out the guide rail 18.

[0075] The third threshold may be a preset value, but similarly to the first threshold, the difference between the maximum and minimum values ​​of the area ratio is set to 100%, and the third threshold is set to a value several percent larger than the minimum value. Note that the third threshold may also be set to the same value as the second threshold.

[0076] Furthermore, the point at which the area ratio of the guide rail 18 in the image data exceeds a fourth threshold value that is larger than the third threshold value, i.e., the point at which the area ratio of the guide rail 18 reaches almost its maximum value, is considered to be the point at which the guide rail 18 has been fully pulled out, and is therefore determined to be the end point of the pulling-out process of pulling out the guide rail 18.

[0077] The fourth threshold may be a preset value, or, like the first threshold, may be set to a value several percent smaller than the maximum value, where the difference between the maximum and minimum values ​​of the area ratio is 100%. Note that the fourth threshold may also be set to the same value as the first threshold.

[0078] Once the start and end times of each work process are determined in this way, the time required for each work process is calculated.

[0079] Specifically, as shown in Figure 6, the time from the start to the end of the pull-out process in which the guide rail 18 is pulled out is calculated as the time required for the pull-out process, the time from the end of the pull-out process in which the guide rail 18 is pulled out to the start of the movement process in which the mobile work vehicle 10 is moved along the guide rail 18 is calculated as the time required for the fixing process in which the pulled-out guide rail 18 is fixed to the main girder 3, and the time from the start to the end of the movement process in which the mobile work vehicle 10 is moved along the guide rail 18 is calculated as the time required for the movement process.

[0080] Furthermore, when the work process determination unit 33 determines a new start time for the pull-out process, it calculates the time that has elapsed from the previously determined start time of the pull-out process to the newly determined start time of the pull-out process as the time (cycle time) required to carry out one cycle of the relocation process of the mobile work vehicle 10. Note that the time from the end of the transfer process to the start of the pull-out process may also be calculated as the work platform usage period during which construction work is carried out on the work platform 16 of the mobile work vehicle 10.

[0081] By checking the results calculated by the work process judgment unit 33 on the display unit 36, etc., the manager who manages the progress of work can quantitatively and objectively grasp the work status of each work process, and for example, can predict the start time of the next work process from the time required for each work process and make timely arrangements for the necessary personnel and materials. Also, for example, if there is a large difference in the time required for the same work process in previous and subsequent work cycles, it becomes possible to investigate the cause of the delay and consider countermeasures based on quantitative and objective data.

[0082] In the above example, the start and end times of each work process are determined based on the area ratio of the area in which the guide rail 18 (specific component) is captured to the area of ​​the entire image data, but the start and end times of each work process may also be determined based on the change in length in a specified direction of the figure formed by pixels identified as containing the guide rail 18 (specific component), for example, the change in length of the figure in the vertical direction of the image data shown in Figures 5A to 5C.

[0083] Furthermore, the start and end times of each work process may be determined simultaneously (in real time) with the acquisition of image data by the imaging unit 41. In this case, it becomes possible to easily grasp the progress of the work, i.e., which work process is currently in progress. The start and end times of each work process may be determined at predetermined time intervals (e.g., every few hours to a day) that are sufficiently longer than the time interval at which image data is acquired by the imaging unit 41, or may be determined when the work scheduled for one day at the work site is completed. In order to improve the accuracy of the determination, it is preferable that the start and end times of each work process be determined after the imaging unit 41 has acquired image data for a sufficiently long period of time (e.g., every few hours to a day) that allows the maximum and minimum values ​​of the area ratio of the guide rail 18 (specific component) to be determined.

[0084] Next, with reference to Figures 7 and 8, we will explain the case where a construction process for an overhanging section consisting of multiple work processes is a work cycle analyzed by the above-mentioned work cycle analysis system 100. Figures 7A to 7C are diagrams showing image data of the work area captured by the imaging unit 41, and similar to Figures 5A to 5C, the lower side in the diagram is the overhang direction of the main girder 3, that is, the work area where new blocks are constructed sequentially. Also, Figure 8 is a graph showing the analysis results of the work cycle.

[0085] Here, the image data P21, P22, and P23 shown in Figures 7A to 7C are images captured by the imaging unit 41 of the state shown in Figure 2D above, i.e., the work being carried out sequentially to construct a new sixth block 3f so that it protrudes from the fifth block 3e.

[0086] Specifically, the first image data P21 shown in Fig. 7A is an image captured by the imaging unit 41 of a state in the middle of work to place the reinforcing bars 22 and the sheath tubes 24 in the formwork 20 installed in the work area in the formwork installation process, and the second image data P22 shown in Fig. 7B is an image captured by the imaging unit 41 of a state in which the work of placing the reinforcing bars 22 and the sheath tubes 24 in the formwork 20 has been completed. For this reason, compared to the first image data P21, the subject of the second image data P22 is almost entirely occupied by the reinforcing bars 22 and the sheath tubes 24.

[0087] 7C is an image captured by the imaging unit 41 of concrete being poured into the formwork 20 after the arrangement of the reinforcing bars 22 and the sheath tube 24 has been completed. Therefore, the subject of the third image data P23 is almost entirely the surface of the poured concrete, i.e., the top surface of the sixth block 3f.

[0088] In this way, in the construction process of the protrusion section, the reinforcing bars 22 and the sheath pipes 24 are components used in specific work processes among the multiple work processes that make up the work cycle, specifically, the reinforcement process and the concrete pouring process.

[0089] Therefore, in order to identify the reinforcing bars 22 and sheath tubes 24 as components corresponding to specific components from image data P such as those shown in Figures 7A to 7C, the identification unit 32 stores the results of machine learning that was previously performed using, for example, images of crossed reinforcing bars 22 and images of arranged sheath tubes 24 as training data to identify the shape and position of the reinforcing bars 22 and sheath tubes 24.

[0090] Therefore, as shown in Figures 7A and 7B, the recognition unit 32 will recognize areas in each image data in which the intersecting reinforcing bars 22 are captured and areas in which the sheath tube 24 extending along the protrusion direction is captured.

[0091] The size of the area in which the reinforcing bars 22 and the sheath pipes 24 appear in each image data gradually increases as the reinforcement work progresses from the state shown in FIG. 7A, and reaches a maximum in the state shown in FIG. 7B.

[0092] Furthermore, the size of the area in which the reinforcing bars 22 and sheath tubes 24 appear in each image data gradually decreases as concrete is poured into the formwork 20 from the state shown in FIG. 7B, and disappears in the state shown in FIG. 7C.

[0093] In this way, the size of the area in which the reinforcing bars 22 and sheath tubes 24 (specific components) are captured is calculated by the work process determination unit 33 as the proportion of the number of pixels identified as containing the reinforcing bars 22 or sheath tubes 24 to the total number of pixels in the image data, i.e., the area in which the reinforcing bars 22 or sheath tubes 24 are captured to the total area of ​​the image data, and the change in the area proportion over time is graphed as shown in Figure 8.

[0094] In addition, the work process determination unit 33 has multiple thresholds for determining the start and end of each work process that makes up the construction process of the overhang section, and determines the start and end of each work process by comparing these thresholds with the determined area ratio.

[0095] Specifically, the point at which the area ratio of the reinforcing bars 22 and sheath tubes 24 in the image data exceeds the first threshold and becomes maximum, i.e., the point at which the area ratio of the reinforcing bars 22 and sheath tubes 24 stops increasing, is considered to be the time when the placement of the reinforcing bars 22 and sheath tubes 24 has been completed, and is therefore determined to be the end of the reinforcement process.

[0096] The first threshold value may be a preset value, but since the absolute value of the area ratio of the reinforcing bar 22 and the sheath tube 24 in each work cycle will fluctuate to some extent, it is preferable that the first threshold value be calculated according to the maximum and minimum values ​​of the area ratio of the reinforcing bar 22 and the sheath tube 24 in each work cycle.For example, the first threshold value is set to a value several percent smaller than the maximum value, with the difference between the maximum and minimum values ​​of the area ratio being 100%.

[0097] Furthermore, the point in time when the area ratio of the reinforcing bars 22 and the sheath tubes 24 exceeds a second threshold value that is smaller than the first threshold value before the point in time determined to be the end of the reinforcement process, i.e., the point in time when the area ratio of the reinforcing bars 22 and the sheath tubes 24 begins to increase, is considered to be the time when the placement of the reinforcing bars 22 and the sheath tubes 24 within the formwork 20 begins, and is therefore determined to be the start of the reinforcement process.

[0098] The second threshold may be a preset value, but similarly to the first threshold, the difference between the maximum and minimum values ​​of the area ratio is set to 100%, and the second threshold is set to a value several percent larger than the minimum value.

[0099] Furthermore, the point in time after the point in time determined as the end of the reinforcement process when the area ratio of the reinforcing bars 22 and the sheath tubes 24 falls below a third threshold value that is smaller than the first threshold value, i.e., the point in time when the area ratio of the reinforcing bars 22 and the sheath tubes 24 becomes almost zero, is considered to be the point in time when the reinforcing bars 22 and the sheath tubes 24 placed in the formwork 20 are almost buried in the poured concrete, and is therefore determined as the end of the concrete pouring process. Note that, since the concrete pouring process starts not long after the reinforcement process is completed, the point in time when the area ratio of the reinforcing bars 22 and the sheath tubes 24 in the image data exceeds the first threshold value and becomes maximum is determined as the end of the reinforcement process and the start of the concrete pouring process.

[0100] The third threshold may be a preset value, but similarly to the first threshold, the difference between the maximum and minimum values ​​of the area ratio is set to 100%, and the third threshold is set to a value several percent larger than the minimum value. Note that the third threshold may also be set to the same value as the second threshold.

[0101] Once the start and end times of each work process are determined in this way, the time required for each work process is calculated.

[0102] Specifically, as shown in Figure 8, the time from the start to the end of the reinforcement process is calculated as the time required for the reinforcement process, and the time from the end of the reinforcement process to the end of the concrete pouring process is calculated as the time required for the concrete pouring process.

[0103] Furthermore, when the work process determination unit 33 determines a new start time for the reinforcement process, it calculates the time elapsed from the previously determined start time of the reinforcement process to the newly determined start time of the reinforcement process as the time required to perform one cycle of the construction process of the overhang section (cycle time). Note that the time from the end of the concrete pouring process to the start time of the reinforcement process may be calculated as the time required for other processes including the concrete curing process, formwork dismantling process, and formwork installation process.

[0104] By checking the results calculated by the work process judgment unit 33 on the display unit 36, etc., the manager who manages the progress of work can quantitatively and objectively grasp the work status of each work process, and for example, can predict the start time of the next work process from the time required for each work process and make timely arrangements for the necessary personnel and materials. Also, for example, if there is a large difference in the time required for the same work process in previous and subsequent work cycles, it becomes possible to investigate the cause of the delay and consider countermeasures based on quantitative and objective data.

[0105] In the above example, the case has been described in which the specific components identified by the identification unit 32 are the reinforcing bars 22 and the sheath tubes 24 used in the reinforcement arrangement process, but the specific components identified by the identification unit 32 may be either the reinforcing bars 22 or the sheath tubes 24 used in the reinforcement arrangement process. However, in the reinforcement arrangement process, there are times when the sheath tubes 24 are placed above the reinforcing bars 22 and times when the reinforcing bars 22 are placed above the sheath tubes 24. Therefore, if only either the reinforcing bars 22 or the sheath tubes 24 are identified as the specific components, the area ratio will fluctuate, and as a result, there is a risk that the work process determination unit 33 will not accurately determine the work process.

[0106] Therefore, in the above example, it is preferable that both the reinforcing bar 22 and the sheath tube 24 used in the reinforcing bar arrangement process are specific components that can be identified by the identification unit 32.

[0107] Furthermore, in the above example, the start and end times of each work process are determined based on the area proportion of the area in which the reinforcing bars 22 and sheath tubes 24 (specific components) appear to the area of ​​the entire image data, but the start and end times of each work process may also be determined based on the change in length in a specified direction of the figure formed by pixels identified as containing the reinforcing bars 22 and sheath tubes 24 (specific components), for example, the change in length of the figure in the left-right and up-down directions of the image data shown in Figures 7A to 7C, or the change in the length of the diagonal of the figure.

[0108] The start and end times of each work process may be determined simultaneously (in real time) with the acquisition of image data by the imaging unit 41. In this case, it is possible to easily grasp the progress of the work, i.e., which work process is currently in progress. The start and end times of each work process may be determined at predetermined time intervals (e.g., every few hours to a day) that are sufficiently longer than the time interval at which image data is acquired by the imaging unit 41, or may be determined when the scheduled work for one day is completed at the work site. In order to improve the accuracy of the determination, it is preferable that the start and end times of each work process be determined after the imaging unit 41 has acquired image data for a sufficiently long time (e.g., every few hours to a day) that allows the maximum and minimum values ​​of the area ratios of the reinforcing bars 22 and the sheath tube 24 (specific components) to be determined.

[0109] Furthermore, in the formwork installation process carried out before the reinforcement process, the formwork 20 is installed over an area where the reinforcing bars 22 and sheath pipes 24 will be arranged, and the area of ​​the formwork 20 reflected in the image data decreases as the reinforcing bars 22 and sheath pipes 24 are arranged in the reinforcement process. For this reason, instead of the reinforcing bars 22 and the sheath pipes 24, the formwork 20 may be used as a specific component, and the start and end times of the formwork installation process and the reinforcement process may be determined based on changes in the area ratio of the area reflected in the formwork 20 to the area of ​​the entire image data.

[0110] Furthermore, in the concrete pouring process carried out after the reinforcement process, concrete is poured over the area where the reinforcing bars 22 and sheath pipes 24 are arranged, and the area of ​​concrete reflected in the image data increases as the concrete pouring process progresses. For this reason, concrete may be used as a specific component instead of the reinforcing bars 22 and sheath pipes 24, and the start and end times of the concrete pouring process may be determined based on changes in the area ratio of the area reflected in the concrete to the area of ​​the entire image data.

[0111] Furthermore, the area of ​​the formwork 20 reflected in the image data decreases as the reinforcing bars 22 and sheath tubes 24 are placed, and the area of ​​the concrete reflected in the image data increases as the concrete pouring process progresses. Therefore, by identifying the formwork and concrete as specific components in advance, along with the reinforcing bars 22 and sheath tubes 24, it is possible to use the changes in the area proportions identified as formwork 20 and concrete as an auxiliary means when analyzing the reinforcement process and concrete pouring process based on the area proportions of the reinforcing bars 22 and sheath tubes 24 as described above, thereby improving the accuracy of analysis of the reinforcement process and concrete pouring process.

[0112] Next, with reference to Fig. 9, a case will be described in which a work cycle including the above-mentioned mobile work vehicle 10 relocation process and the above-mentioned overhang construction process is analyzed by the above-mentioned work cycle analysis system 100. Fig. 9 is a graph showing the results of the work cycle analysis.

[0113] The analysis of the work cycle including the process of relocating the mobile work vehicle 10 and the process of constructing the overhanging section is carried out by combining the results of the analysis of the process of relocating the mobile work vehicle 10 and the results of the analysis of the process of constructing the overhanging section. In this case, the specific components identified by the identification unit 32 are the guide rail 18, the reinforcing bar 22, and the sheath pipe 24.

[0114] Specifically, as shown in Figure 9, by aligning the time axes of the graph shown in Figure 6, which analyzes the relocation process of the mobile work vehicle 10, and the graph shown in Figure 8, which analyzes the construction process of the overhang section, it is possible to further determine the start and end times of the work process, which would not be possible if the relocation process of the mobile work vehicle 10 and the construction process of the overhang section were analyzed separately.

[0115] For example, during the period between the time determined to be the end of the moving process in which the mobile work vehicle 10 is moved along the guide rail 18 and the time determined to be the start of the reinforcement process, construction work on the protrusion can begin on the work floor 16 of the mobile work vehicle 10, and this period is therefore determined to be the period for the formwork installation process which is carried out before the reinforcement process.

[0116] Furthermore, the period between the time determined to be the end of the concrete pouring process and the time determined to be the start of the pull-out process of pulling out the guide rail 18 is determined as the period for the concrete curing process and form dismantling process that follow the concrete pouring process. Since the time required for the concrete curing process is generally set in advance based on the type of cement and the outside air temperature, the time required for the form dismantling process can be determined by subtracting the set curing period from this period. Note that a tensioning process may be performed after the form dismantling process to tension the PC steel wires passed through the sheath pipe 24.

[0117] As a result, for example, as shown in Figure 9, the work process determination unit 33 will calculate the time required to perform one work cycle (cycle time) starting from the time determined to be the start time of the formwork installation process, i.e., the time determined to be the end time of the movement process, and the time required for each work process performed during one cycle.

[0118] By checking the results calculated by the work process judgment unit 33 on the display unit 36, etc., the manager who manages the progress of work can quantitatively and objectively grasp the work status of each work process, and for example, can predict the start time of the next work process from the time required for each work process and make timely arrangements for the necessary personnel and materials. Also, for example, if there is a large difference in the time required for the same work process in previous and subsequent work cycles, it becomes possible to investigate the cause of the delay and consider countermeasures based on quantitative and objective data.

[0119] According to the above embodiment, the following effects are achieved.

[0120] According to the above-described work cycle analysis system 100, the start or end of a work process included in a work cycle is determined based on a change in the pixel ratio (area ratio) of the specific components 18, 22, 24 identified by the identification unit 32 within the image.

[0121] In this way, the work status of the work process included in the work cycle can be understood from changes in the pixel ratio (area ratio) of specific components 18, 22, 24 in the image, so there is no need to identify multiple work machines or workers or analyze their movements, for example. This eliminates the need to set up complex recognition logic or judgment logic, making it relatively easy to understand the work status of a building.

[0122] Furthermore, particularly in the case of a structure that is constructed by repeatedly performing a work cycle, a manager who manages the progress of work on the structure can quantitatively and objectively grasp the work cycle and the work status of each work process by checking the analysis results of the cycle time and the time required for each work process analyzed by the above-mentioned work cycle analysis system 100. This makes it possible, for example, to predict the start time of the next work process from the time required for each work process and to arrange for the necessary personnel and materials in a timely manner, and if there is a large difference in the time required for the same work process between previous and subsequent work cycles, it becomes possible to investigate the cause of the delay and consider countermeasures based on quantitative and objective data, thereby improving the efficiency of work in constructing the structure.

[0123] The following modified examples are also within the scope of the present invention, and it is possible to combine the configuration shown in the modified example with the configuration described in the above embodiment, or to combine the configurations described in the different modified examples below.

[0124] In the above embodiment, the structure to which the work cycle analysis system 100 is applied is described as a bridge 1 constructed by the cantilever erection method, but the structures to which the work cycle analysis system 100 is applied are not limited to this. If the structure has specific components that are used in a specific work process among the multiple work processes that make up a work cycle, such as the above-mentioned guide rails 18 and reinforcing bars 22, it is possible to determine the start and end times of the work process using the above-mentioned work cycle analysis system 100.

[0125] For example, when a building is constructed by repeating a work cycle consisting of multiple work processes moving vertically upward, such as a steel-framed, reinforced concrete, or reinforced steel-reinforced concrete structure, it is possible to determine the start and end of the work processes included in the work cycle based on changes in the pixel proportion (area proportion) of specific components (e.g., columns, beams, reinforcing bars that make up slabs, and formwork installed when pouring concrete into these) in images taken from the side or above the work area.

[0126] Furthermore, in the above embodiment, the structure to which the work cycle analysis system 100 is applied is described as a structure constructed by repeatedly performing approximately the same work cycle, such as the bridge 1 constructed by the cantilever erection method, but the structure to which the work cycle analysis system 100 is applied is not limited to this, and may be, for example, a structure constructed by performing a work cycle consisting of multiple work processes only once. In this case as well, by analyzing the time required for each work process, etc., it is possible to improve the construction work efficiency of structures constructed by similar work cycles.

[0127] In the above embodiment, a method using machine learning has been described as a method for identifying a specific component from image data, but the method for identifying a specific component from image data is not limited to a method using machine learning, and any method that can identify a specific component from image data may be used, and known image recognition methods such as template matching and pattern recognition may be used. For example, in template matching, a template image of a specific component such as a guide rail 18 or a reinforcing bar 22 is prepared in advance, and the specific component is identified and extracted from the image data by comparing the template image with the image captured by the imaging unit 41.

[0128] In the above embodiment, the time when the area ratio of the reinforcing bars 22 and the sheath tube 24 in the image data exceeds the first threshold and reaches a maximum is determined as the end of the reinforcement process and the start of the concrete pouring process. Alternatively, to determine the start of the concrete pouring process, as shown in the graph of FIG. 10 , the time when the area ratio exceeds a fourth threshold when counting back from the time when the end of the concrete pouring process is determined because the area ratio falls below the third threshold may be determined as the start of the concrete pouring process. The fourth threshold may be a preset value, or, similar to the first threshold, may be set to a value several percent smaller than the maximum value, with the difference between the maximum and minimum area ratios being 100%. Note that the fourth threshold may also be set to the same value as the first threshold.

[0129] In this way, by determining the start time of the concrete pouring process by going back from the time when the end time of the concrete pouring process was determined, it becomes possible to accurately determine the start time of the concrete pouring process even if, for example, there is a certain period of time when no work is performed between the end time of the reinforcement process and the start time of the concrete pouring process. This allows the manager who manages the progress of work on the building to accurately understand the time required for the concrete pouring process and when the concrete pouring process was performed.

[0130] Furthermore, in the above embodiment, the identification unit 32 automatically identifies and extracts specific components from the image data P captured by the imaging unit 41 based on the stored learning results, and the work process determination unit 33 calculates the area ratio of the area in which the specific component appears to the entire area of ​​the image data, and determines the start and end of the work processes included in the work cycle based on changes in the calculated area ratio.

[0131] In addition, the identification unit 32 may further identify people (workers) from the image data P captured by the imaging unit 41 based on the stored learning results, and the work process determination unit 33 may determine the start and end of work processes included in the work cycle based on changes in the pixel ratio in the image data P of a specific component identified by the identification unit 32 and changes in the number of people identified by the identification unit 32.

[0132] Specifically, the identification unit 32 uses a detection algorithm that uses AI and neural network technology, for example, an object detection algorithm called YOLO (You Look Only Once), to identify and extract a person PE from image data P30 captured by the imaging unit 41, as shown in Fig. 11. YOLO is an algorithm that uses a CNN (Convolutional Neural Network), and the area in the image where the detection target is represented is indicated by a bounding box.

[0133] The image data P30 shown in Figure 11, like Figure 7A described above, is an image captured by the imaging unit 41 of the work in progress, in which reinforcing bars 22 and sheath tubes 24 are placed in the formwork 20 installed in the work area during the formwork installation process, and in this example, two people (workers) PE performing the work are identified.

[0134] The method for extracting people from image data P is not limited to this, and people may be extracted from image data P using a known method other than YOLO.

[0135] On the other hand, the work process determination unit 33 calculates the area ratio (pixel ratio) within the image data P of the specific component identified by the identification unit 32, and also calculates the number of people PE identified by the identification unit 32, and determines the start and end of the work process depending on changes in the area ratio and changes in the number of people PE.

[0136] Specifically, when analyzing the construction process of an overhanging portion, for example, the work process determination unit 33 determines the change over time in the area ratio of the area in which the reinforcing bars 22 and the sheath pipe 24 appear to the entire area of ​​the image data, as in the above embodiment, as shown in the graph in the upper part of Fig. 12, and also determines the change over time in the number of people PE identified by the identification unit 32, as shown in the graph in the lower part of Fig. 12. Note that the horizontal axes of the two graphs arranged vertically in Fig. 12 indicate the same time.

[0137] In addition, the work process determination unit 33 has determination conditions (thresholds) for determining the start and end of each work process that constitutes the construction process of the extension section from changes in the area ratio of the reinforcing bars 22 and the sheath tube 24, and also has determination conditions for determining the start and end of each work process that constitutes the construction process of the extension section from changes in the number of identified persons PE.

[0138] Therefore, for example, if it is known in advance that the number of workers working in the work area will be at its maximum at the start of the concrete pouring process during the construction process of the overhang section, by setting the maximum number of people PE identified by the identification unit 32 as the judgment condition for determining the start of the concrete pouring process, the work process judgment unit 33 will determine the start of the concrete pouring process as the point at which the maximum number of people PE identified by the identification unit 32 is reached during one cycle of the construction process of the overhang section, as shown in Figure 12.

[0139] As in the above embodiment, the start and end times of the reinforcement process and the end time of the concrete pouring process are determined based on the change in the area ratio of the reinforcing bars 22 and the sheath pipes 24 in the image data, and therefore the start time of the concrete pouring process may be determined to be the time when the number of people PE identified by the identification unit 32 is maximum within the period from the time when it is determined that the reinforcement process has been completed to the time when it is determined that the concrete pouring process has been completed.

[0140] In this way, by determining the start and end of a work process included in a work cycle based on changes in the number of people PE identified by the identification unit 32, it becomes possible to determine the start and end of a work process that is difficult to determine based solely on changes in the area proportion within the image data of a specific component identified by the identification unit 32.

[0141] Furthermore, by combining the results of determining the start and end of a work process based on changes in the pixel proportions in the image data P of a specific component identified by the identification unit 32 with the results of determining the start and end of a work process based on changes in the number of people PE identified by the identification unit 32, it is possible to improve the accuracy of determining the start and end of a work process; for example, the start time of the above-mentioned concrete pouring process can be derived from these two determination results (see Figures 10 and 12).

[0142] In order to improve the accuracy of person identification by the identification unit 32, it is preferable to install a plurality of imaging units 41 so that the work area can be photographed from a plurality of different directions.

[0143] Although the embodiments of the present invention have been described above, the above embodiments merely illustrate some of the application examples of the present invention, and it is not intended that the technical scope of the present invention be limited to the specific configurations of the above embodiments. [Explanation of symbols]

[0144] 100···Work Cycle Analysis System 1. Bridge (structure) 10. Mobile work vehicle 18. Guide rail (specific component) 22. Reinforcement (specific components) 24. Sheath tube (specific component) 30 Cycle Analysis Section 31...Analysis Department 32 Identification unit 33...Work process judgment department 40 Data collection section 41 Imaging unit

Claims

1. A work cycle analysis system for analyzing a work cycle of a building constructed by carrying out a work cycle composed of a plurality of work processes, an imaging unit capable of imaging a work area where the work process is performed; an identification unit that identifies a specific component in an image captured by the imaging unit; a work process determination unit that determines the start or end of the work process included in the work cycle based on a change in the pixel ratio within the image of the specific component identified by the identification unit, Work cycle analysis system.

2. the specific member is a member used in a specific work process among the plurality of work processes that constitute the work cycle, the work process determination unit determines the start or end of the specific work process based on a change in the pixel ratio of the specific component in the image. The work cycle analysis system of claim 1 .

3. The work cycle includes a reinforcing bar arrangement process for arranging reinforcing bars within the work area, The specific member is at least one of the reinforcing bar and the formwork that are placed in the work area in the reinforcing bar arrangement process, The work process determination unit determining the start and end of the reinforcement process based on a change in the pixel ratio of at least one of the reinforcing bars and the formwork in the image; The work cycle analysis system according to claim 1 or 2.

4. The work cycle includes a concrete pouring step of pouring concrete into the work area in which the reinforcing bars are placed, the specific member is at least one of the reinforcing bar and concrete placed in the work area during the concrete pouring step, The work process determination unit determining the start and end of the concrete pouring process based on a change in the pixel ratio of at least one of the reinforcing bars and the concrete in the image; The work cycle analysis system according to claim 1 or 2.

5. The work cycle includes a reinforcing bar arrangement process of arranging reinforcing bars within the work area and a concrete pouring process of pouring concrete into the work area in which the reinforcing bars have been arranged, the specific member is the reinforcing bar arranged in the reinforcing bar arrangement step, The work process determination unit A time point when the pixel ratio of the reinforcing bar in the image exceeds a second threshold is determined as the start of the reinforcement arrangement process; The time when the pixel ratio of the reinforcing bar exceeds a first threshold value that is greater than the second threshold value is determined as the end of the reinforcement arrangement process, The concrete pouring process is determined to be completed when the pixel ratio of the reinforcing bar falls below a third threshold value that is smaller than the first threshold value after the time when the reinforcement arrangement process is determined to be completed. The work cycle analysis system according to claim 1 or 2.

6. the work cycle includes a movement step of moving a mobile work vehicle used when performing a predetermined work step within the work area toward the work area, the specific member is a guide rail that guides the mobile work vehicle to the work area, The work process determination unit determining a time when the pixel ratio of the guide rail in the image reaches a maximum and then falls below a first threshold as the start of the moving process; a time point at which the pixel ratio of the guide rail falls below a second threshold value that is smaller than the first threshold value after the time point at which the start of the moving process is determined is the time point at which the moving process is ended; The work cycle analysis system according to claim 1 or 2.

7. The work cycle is repeated, the work process determination unit calculates, as the cycle time of the work cycle, the time elapsed from the already determined start time of the work cycle to the newly determined start time of the work cycle, or from the already determined end time of the work cycle to the newly determined end time of the work cycle. The work cycle analysis system according to claim 1 or 2.

8. the identification unit further identifies a person from the image captured by the imaging unit; the work process determination unit determines the start or end of the work process included in the work cycle based on a change in a pixel ratio within the image of the specific component identified by the identification unit and a change in the number of people identified by the identification unit. The work cycle analysis system according to claim 1 or 2.

Images