Construction progress monitoring system

Abstract

To provide a construction progress grasping system which manages progress states of wall constructions and ceiling constructions further accurately and easily.SOLUTION: A construction progress grasping system 1 comprises: means 42 which detects constructions including wall constructions and ceiling constructions from identification slips arranged at different positions in a construction site and photographed images in the construction site photographed by a photograph-recording device 3 held or attached by a site worker, estimates progress states of these constructions, and detects the identification slips; a construction range estimation unit 45 which estimates a position and a range where the construction is being performed on the basis of photographed positions of the photographed images calculated from movement information of the site worker acquired by a pedestrian autonomous navigation means 2, assuming, to be a reference point, a position where the identification slip detected immediately before photographing of the photographed image where the construction is detected; and an information display unit 46 which creates display data by superimposing the progress states of the constructions, on the positions and the ranges of the constructions estimated by the construction range estimation unit 45, on a construction drawing of a building, and displays the display data on a display device 5.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a construction progress monitoring system for managing the progress of wall and ceiling construction work at a construction site. [Background technology]

[0002] At construction sites for buildings such as buildings and condominiums, it is necessary to grasp the progress of the construction. For this reason, for example, a site supervisor walks around the construction site to grasp the progress of each part of the site, but this takes time and effort. In response to this, for example, Patent Document 1 discloses a configuration for grasping the progress of work related to oneself based on the grasped multiple tasks performed in each room of a building and the currently performed tasks in each room. In this configuration, multiple read objects showing information on multiple tasks are read, and the states of the multiple tasks corresponding to the multiple read objects are identified based on the read states, and status-related information related to the states of the multiple tasks performed in multiple target areas is stored in a storage means based on the identified statuses of the multiple tasks. When one or more tasks out of the multiple tasks are selected, information corresponding to the status of at least the task selected in the selection information for each of the multiple target areas is displayed based on the status-related information stored in the storage means, with the selected task as a reference. Here, the task refers to interior work, door work, exterior work, and the like of a building under construction. In addition, the read object is, for example, a two-dimensional barcode, and a code sheet on which the two-dimensional barcode is displayed is provided individually in each room of the building. In the configuration disclosed in Patent Document 1, for example, after a worker has carried out work, he or she must fill in the two-dimensional barcode to indicate that the work has been completed, which makes operation difficult. In addition, in the configuration of Patent Document 1, as described above, it is necessary to provide a code sheet for each room in the building. This makes it difficult to manage the progress for each small part in the room. Therefore, with the configuration disclosed in Patent Document 1, it is difficult to manage the progress status for each part accurately and in detail. In the configuration of Patent Document 1, it is possible to perform progress management more accurately and in detail by providing a code sheet for each small part in the room. However, in this case, the number of code sheets to be managed increases, making operation even more difficult.

[0003] Patent Document 2 also discloses a configuration in which an identification code placed at a construction site and the construction site are photographed, and image data of the image showing the construction site is stored in a state in which it can be classified by construction site and in a state in which it is associated with information representing a specific step among multiple steps in the construction applied to the construction site based on information on the position and information on the construction acquired based on the image showing the identification code. Here, the identification code represents position information representing the position of the construction site in the building to be constructed and information on the construction applied to the construction site. The information on the construction includes information representing the type of construction and information representing the construction method. According to this configuration, by photographing the identification code placed at the construction site and then photographing the construction site, images based on the image data of the image showing the construction site are displayed in the order of steps for each construction site, and the progress of the construction at each construction site can be easily grasped. However, in the configuration disclosed in Patent Document 2, it is necessary to provide an identification code for each construction site, making it difficult to easily manage the progress of construction.

[0004] Patent Document 3 discloses a configuration in which images of a construction site are stored in association with an identifier that identifies the date and time of shooting and the location of shooting, the proportion of each architectural element in the captured image is predicted by a prediction model, an element model in which each architectural element predicted by the prediction model is recorded in the attribute is identified, a process associated with the identified element model is identified, and progress management data that records the proportion of the architectural elements included in the captured image together with the identified process is recorded in a progress information storage unit. A GPS (Global Positioning System) is used to identify the location of the captured image. In a configuration such as that of Patent Document 3, radio waves from GPS satellites cannot be received inside a building, and the position of the shooting location cannot be obtained, or even if it can be obtained, there is a possibility that the obtained position will have a large error. For this reason, progress management data may not be created accurately. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Patent Publication No. 2021-12677 [Patent Document 2] JP 2020-30519 A [Patent Document 3] Patent Publication No. 2021-47715 Summary of the Invention [Problem to be solved by the invention]

[0006] The problem to be solved by the present invention is to provide a construction progress monitoring system that can more accurately and simply manage the progress of wall construction and ceiling construction. [Means for solving the problem]

[0007] In order to solve the above problems, the present invention employs the following means. That is, the construction progress assessment system of the present invention is a construction progress assessment system for managing the progress of wall construction and ceiling construction within a construction site, and includes: an identification tag that is placed at different positions within the construction site and has identification information recorded thereon; a pedestrian autonomous navigation means that is configured to be held or worn by a site worker or a mobile body configured to be movable within the construction site and acquires movement information of the site worker or the mobile body, an imaging and recording device that is configured to be held or worn by the site worker or the mobile body and captures images within the construction site; and a construction progress assessment system that detects construction works including the wall construction and the ceiling construction from images captured by the imaging and recording device, estimates the progress of these works, and records the identification information. an information integration unit that calculates the shooting position of the photographed image in which the construction work is detected based on the movement information, using the position at which the identification tag detected immediately before the photographed image was placed as a base point, and associates the shooting position with the progress of the construction work corresponding to the photographed image and stores them in a database; a construction range estimation unit that estimates the position and range where the construction work is being carried out based on the shooting position; and an information display unit that creates display data by overlaying the progress of the construction work on the position and range of the construction work estimated by the construction range estimation unit on a construction drawing of a building, and displays the display data on a display device. According to this configuration, when a site worker or a moving body holds or wears a pedestrian autonomous navigation means and an image recording device and moves within the construction site, the image recording device captures the image within the construction site. The pedestrian autonomous navigation means obtains movement information indicating the position and route of the site worker or the moving body when moving within the construction site. The image recording device captures the image within the construction site, thereby obtaining captured images of the construction work, including the wall construction work and the ceiling construction work, within the construction site. The detection means detects the construction work, including the wall construction work and the ceiling construction work, within the construction site from the captured image of the image recording device, and estimates the progress of the construction work. The detection means detects the identification tag placed within the construction site. The information integration unit calculates the image capture position of the captured image based on the movement information of the site worker or the moving body, for the captured image in which the construction work is detected, using the position where the identification tag detected immediately before the capture of the captured image as the base point. This makes it possible to identify the image capture position with high accuracy, even for high-rise buildings and large-scale structures. The information integration unit also associates the photographing position of the photographed image calculated as described above with the progress of the construction work corresponding to the photographed image and stores them in the database. The construction range estimation unit also estimates the position and range of the construction work based on the photographing position. The information display unit displays the progress of various construction works, including wall construction and ceiling construction, superimposed on the position and range of the construction work estimated by the construction range estimation unit on the construction drawing of the building. In this way, the progress of the wall construction and ceiling construction is superimposed on the building drawing, so that the position and progress of the construction work can be easily grasped. In particular, in the above-mentioned configuration, the identification tag that is used as the base point for identifying the shooting position of the captured image only needs to be installed in places where on-site workers frequently pass, for example, on each floor of a building. In other words, there is no need to install an identification tag individually for each part, such as a wall or ceiling, that is the subject of construction or progress management, and therefore the number of identification tags installed can be reduced. In addition, since the progress status is automatically estimated by the detection means from the captured image, even if the work progresses, on-site workers are not required to report or input data into the system for progress management. This further improves ease of operation. As a result, it is possible to provide a construction progress monitoring system that can more accurately and simply manage the progress of wall construction and ceiling construction.

[0008] In one aspect of the present invention, the detection means estimates which of a plurality of construction types is being carried out as the progress status, and when creating the display data, the information display unit estimates and displays the completion rate for each of the plurality of construction types by dividing the quantity of construction work completed for that construction type by the total quantity for that construction type. According to this configuration, the detection means estimates which of the multiple construction types is being carried out as the progress status. Based on the estimation result, the completed quantity of each of the multiple construction types is divided by the total quantity of the construction type, so that the progress rate of the construction can be more easily grasped.

[0009] In one aspect of the present invention, the detection means has a trained model that has been trained using construction images taken of a construction site and the progress status corresponding to the construction images as learning data, and inputs the captured images into the trained model to estimate, with respect to the wall construction, the progress status as one of a plurality of construction types including wall underlayment work, insulation work, board work, and finishing work, or, with respect to the ceiling construction, to estimate, with respect to the ceiling construction, the progress status as one of a plurality of construction types including ceiling underlayment work, board work, equipment work, and fire-resistant coating work. According to this configuration, by subjecting wall construction images and ceiling construction images to machine learning, the specific construction type of wall construction and ceiling construction can be estimated from each captured image. Therefore, it becomes possible to grasp the progress of construction more precisely.

[0010] In one aspect of the present invention, the information display unit displays the progress status of the construction work by superimposing it on the wall and ceiling portions corresponding to the construction work on the construction drawing. According to this configuration, the progress of wall construction and ceiling construction can be displayed superimposed on the wall and ceiling parts corresponding to the construction on the construction drawing, allowing the progress of the wall construction and ceiling construction to be grasped more intuitively.

[0011] In one aspect of the present invention, the identification tag is provided for each floor in the building. For example, if a system were implemented to obtain movement information of field workers and the like using GPS, it would be difficult to obtain information regarding height with sufficient accuracy, making it difficult to improve the accuracy of the movement information, especially in high-rise buildings. In contrast, with the above-mentioned configuration, when calculating the photographing position of the photographed image based on the movement information of the field worker or the moving object with the position where the identification tag is placed as the base point, the height position where the photographed image was taken can be obtained by referring to the detection result of the identification tag. Therefore, even if the movement information does not include information on height obtained by the pedestrian autonomous navigation means, or even if there is such information, the accuracy of which is unreliable, it is possible to accurately grasp the height position within the building. Effect of the Invention

[0012] According to the present invention, it is possible to provide a construction progress monitoring system that can more accurately and simply manage the progress of wall construction and ceiling construction. [Brief description of the drawings]

[0013] [Figure 1] 1 is a block diagram showing a configuration of a construction progress monitoring system according to an embodiment of the present invention. [Diagram 2] 2 is a diagram showing an example of a construction site where positions of wall construction and ceiling construction are detected by the construction progress monitoring system of FIG. 1. FIG. [Diagram 3] FIG. 13 illustrates an example of a PDR record showing movement information of a field worker. [Figure 4] FIG. 13 is a diagram showing an example of a captured image of an identification tag. [Diagram 5]FIG. 11 is a diagram showing an example of a history of detection of an identification tag from a captured image. [Figure 6] FIG. 4 is a diagram showing an example of a captured image on which detection is performed by a detection unit. [Figure 7] A figure showing an example of a wall construction type estimation record in the detection means. [Figure 8] FIG. 11 is a diagram showing an example of integrated data created in an information integration unit. [Figure 9] 13 is a diagram showing the results of estimation of the position and range of wall construction by a construction range estimation unit. FIG. [Figure 10] 11 is an explanatory diagram showing a method for estimating the position and range of wall construction work by a construction work range estimation unit. FIG. [Figure 11] 13 is a diagram showing the results of estimation of the position and range of ceiling construction work by a construction work range estimation unit. FIG. [Figure 12] 1 is a flowchart showing the flow of a construction progress estimation method executed using the construction progress assessment system of the present embodiment. [Figure 13] FIG. 13 is a diagram showing an example of a wall construction type estimation record when an omnidirectional camera is used. [Figure 14] FIG. 13 is a diagram showing an example of integrated data when an omnidirectional camera is used. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0014] The present invention is a construction progress monitoring system that manages the progress of wall construction and ceiling construction, and detects the locations and types of various construction works being carried out in various parts of a building using an image recording device, a pedestrian autonomous navigation means for acquiring movement information of on-site workers or mobile objects, and a detection means for detecting the progress of construction work, and displays the progress of the construction work and the completion rate based on the progress on an information display unit. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, an embodiment for carrying out a construction progress monitoring system according to the present invention will be described with reference to the accompanying drawings. A block diagram showing the configuration of a construction progress monitoring system according to an embodiment of the present invention is shown in Fig. 1. Fig. 2 is a diagram showing an example of a construction site where the positions of wall construction and ceiling construction are to be detected by the construction progress monitoring system of Fig. 1. 1, the construction progress assessment system 1 mainly comprises a pedestrian autonomous navigation means 2, an image recording device 3, a management system main body 4, and a display device 5. The construction progress assessment system 1 manages the progress of construction work, including wall construction and ceiling construction work, within a construction site G. In this embodiment, the construction site G is a site where construction and construction of a building having multiple floors, such as a high-rise building, is carried out. As shown in FIG. 2, the pedestrian autonomous navigation means 2 and the photographing and recording device 3 are used by the site worker Q when he / she photographs the inside of the construction site G while moving within the construction site G of the building under construction. The pedestrian autonomous navigation means 2 and the photographing and recording device 3 are configured to be held or worn by the site worker Q. That is, the pedestrian autonomous navigation means 2 and the photographing and recording device 3 may be held by the site worker Q in his / her hand or by storing them in his / her pocket. Furthermore, the pedestrian autonomous navigation means 2 and the photographing and recording device 3 may be worn on a part of the body (for example, the head, etc.) of the site worker Q or on an outfit (a helmet or clothing). The site worker Q may also wear a helmet or a bag to which the pedestrian autonomous navigation means 2 and the photographing and recording device 3 are attached. The pedestrian autonomous navigation means 2 and the photographing and recording device 3 may be separate devices, or a device in which the pedestrian autonomous navigation means 2 and the photographing and recording device 3 are integrated may be used. In addition, the pedestrian autonomous navigation means 2 and the photographing and recording device 3 may be configured to be attachable to a moving body such as a drone or robot configured to move along a predetermined route within the construction site G, instead of a site worker Q.

[0015] The pedestrian autonomous navigation means 2 is a device that acquires movement information of the field worker Q by pedestrian autonomous navigation, i.e., PDR (Pedestrian Dead-Reckoning). As the pedestrian autonomous navigation means 2 described above, a dedicated device having the functions shown below may be used, or a smartphone, tablet terminal, or the like having an application program capable of realizing the functions shown below may be used. As shown in FIG. 1, the pedestrian autonomous navigation means 2 functionally includes a sensor 21, a movement information calculation unit 22, a data storage unit 23, and a data output unit 24. As the sensor 21, for example, an acceleration sensor, a gyro sensor, a geomagnetic sensor, etc. are used. After starting positioning, the sensor 21 detects changes in acceleration, angular velocity, geomagnetism, etc. that occur with the movement of the field worker Q holding or wearing the pedestrian autonomous navigation means 2. Note that the sensor 21 is not limited to an acceleration sensor, gyro sensor, or geomagnetic sensor, and may be one that detects other parameters as appropriate, as long as it can acquire movement information of the field worker Q within the construction site G.

[0016] The movement information calculation unit 22 generates movement information of the field worker Q by calculating the amount of movement from the positioning start position, i.e., the relative position to the positioning start position, when the field worker Q moves from the positioning start position, based on the detection data from the sensor 21. More specifically, the movement information calculation unit 22 measures the amount of movement from the position at the previous time, i.e., the relative position, at regular time intervals using the PDR and calculates the position by integrating the amount of movement. In addition, the movement information calculation unit 22 obtains the orientation of the field worker Q from the detection data from the sensor 21 at regular time intervals using the PDR. In the movement information of the field worker Q, information indicating the movement route (amount of movement) based on the relative position measured as described above, the orientation of the field worker Q at each position on the movement route, and time information at each position are associated with each other. FIG. 3 is a diagram showing an example of a PDR record showing movement information of a field worker. In the movement information shown in FIG. 3, for example, time information (date and time) and the position (X coordinate, Y coordinate) and orientation of the field worker Q at that time are associated with each other at preset time intervals, and are recorded as text data in which, for example, one set of associated correspondence relationships is written on one line. The data storage unit 23 stores the movement information of the site worker Q generated by the movement information calculation unit 22. The data output unit 24 outputs the movement information of the field worker Q stored in the data storage unit 23 to the management system main body 4. The form in which the data output unit 24 outputs the movement information may be, for example, data transfer via a wireless LAN (Local Area Network) such as Wi-Fi or Bluetooth (registered trademark), a mobile phone communication network, or data transfer via a connection cable or various types of portable memory.

[0017] The photographing and recording device 3 photographs the inside of the construction site G. The photographing and recording device 3 is, for example, a portable digital camera, a digital video (including a wearable camera that can be attached to the body), or the like. The photographing and recording device 3 may be a so-called omnidirectional camera (360-degree camera) that can photograph the surroundings in all directions of 360 degrees, but in this embodiment, a configuration is described in which the photographing and recording device 3 uses a (normal) camera that photographs mainly in one direction, rather than an omnidirectional camera. A configuration using an omnidirectional camera will be described later as a modified example. The photographing and recording device 3 may be a smartphone or tablet terminal equipped with a camera function. The photographing and recording device 3 functionally comprises a photographing unit 31, a photographing data storage unit 32, and a photographing data output unit 33. The photographing unit 31 includes an image pickup element using a CCD (Charge Coupled Device), a CMOS (Complementary Metal Oxide Semiconductor), or the like, and photographs at least one of still images and moving images.

[0018] The photographing data storage unit 32 stores the photographed image photographed by the photographing unit 31 as photographing data in association with the time when the photographed image was photographed, i.e., the photographing time. The photographing data storage unit 32 may use a memory built into the photographing and recording device 3, or a portable memory detachably provided in the photographing and recording device 3. The photographing data storage unit 32 may also transfer and store the photographing data in a smartphone or tablet terminal that can be connected to the photographing and recording device 3 via wireless LAN communication such as Wi-Fi. The photographing data output unit 33 outputs the photographing data of the construction site G stored in the photographing data storage unit 32 to the management system main body 4. As a form for outputting the movement information by the data output unit 24, similar to the data output unit 24, for example, data transfer via a wireless LAN such as Wi-Fi or a mobile phone communication network, as well as data transfer via a connection cable or various types of portable memories can be used.

[0019] The management system main body 4 estimates and manages the progress and volume of work within the construction site G based on the movement information of the site worker Q obtained by the pedestrian autonomous navigation means 2 and the photography recording device 3, and the photography data. The management system main body 4 is composed of a computer device such as a personal computer. The management system main body 4 functionally comprises a data input reception unit 41, a detection means 42, an information integration unit 43, a database 44, a construction area estimation unit 45, and an information display unit 46. The data input receiving unit 41 receives input of data output from the data output unit 24 of the pedestrian autonomous navigation means 2 and the photographing data output unit 33 of the photographing and recording device 3 .

[0020] The database 44 stores various data necessary for the management system main body 4 to estimate the progress of construction work within the construction site G. The database 44 stores data on the wall construction work and ceiling construction work carried out within the construction site G, and data on the identification tag M. As data related to wall construction and ceiling construction, for example, drawing data of each floor in the construction site G is stored in the database 44. As drawing data of each floor in the construction site G, there are CAD (Computer Aided Design) data and BIM (Building Information Modeling) data of the portion including the walls and ceiling of each floor. Specifically, as data related to wall construction, information on wall surface lines indicating the position of the wall surface, the type of material constituting each wall (structural material, base material, insulation material, finishing material, etc.), area, thickness, etc. are stored in the database 44. As data related to ceiling construction, for example, information on the type of ceiling panel, the installation position of equipment such as ducts and lighting, etc. are stored in the database 44. The identification tag M is placed within the construction site G as shown in FIG. 2. When the construction site G has multiple floors as in this embodiment, the identification tag M is provided on each floor. The identification tag M is, for example, a two-dimensional barcode, an AR marker, etc., and individual identification information (marker ID) is associated with and recorded for each identification tag M. As data related to the identification tag M, the correspondence between the identification information recorded in association with the identification tag M, information on the horizontal position and height position within the construction site G where the identification tag M is installed, and the direction (azimuth or angle) in which the identification tag M is displayed is stored in the database 44. Information on the horizontal and vertical positions within the construction site G where the identification tag M is installed may include the floor (story number) on which each identification tag M is installed, and the planar position on that floor (XY coordinates: absolute coordinates or relative coordinates on the image data of the design drawings).

[0021] The detection means 42 detects the identification tag M and construction work including wall construction and ceiling construction work from a plurality of captured images taken by the photographing unit 31 for each of the floors in the construction site G. When the captured images are moving images, the detection means 42 breaks down the moving images into a plurality of still images and executes the following processing for each still image. When the captured image to be processed contains identification tag M, the detection means 42 detects (identifies) the identification tag M from the captured image by a known image processing method or the like. When the captured image contains identification tag M, the detection means 42 acquires the shooting time when the captured image in which the identification tag M was detected was captured, and the file name of the captured image. Furthermore, the detection means 42 reads the identification information written on the identification tag M by processing the captured image in which the identification tag M was detected by an appropriate image processing means. 4 is a diagram showing an example of a captured image P of an identification tag. The detection means 42 detects the coordinates of corners Mc of the identification tag M in the captured image P. In this embodiment, the identification tag M has a rectangular shape, and therefore the detection means 42 detects the coordinates of each of the four corners Mc of the identification tag M. The detection means 42 creates an identification tag detection history, which is a history of when the identification tag M was detected. Fig. 5 is an example of the identification tag detection history created by the detection means 42. In the identification tag detection history, the shooting time (date and time) which is the time when the photographed image including the identification tag M was photographed, the identification information (ID) of the detected identification tag M, and the corner coordinates which are the coordinate values ​​of the pixel of the corner Mc of the identification tag M in the photographed image are associated with each other and are stored in the database 44 as text data in which, for example, one set of associated correspondence relationships is written on one line.

[0022] The detection means 42 judges and detects whether or not wall construction or ceiling construction is included in each image captured by the photographing unit 31, and if wall construction or ceiling construction is included, estimates the progress of the construction. The detection means 42 estimates which of a plurality of construction types is being carried out as the progress. For this purpose, the detection means 42 has a trained model 48 that has been deep-learned using a construction image of the construction site G and the type of construction corresponding to the construction image as learning data (teacher data). The trained model 48 has been deep-learned to estimate the type of wall construction or ceiling construction when wall construction or ceiling construction is photographed in the photographed image. The trained model 48 is composed of, for example, a convolutional neural network (CNN). The trained model 48 is used as a program module that is a part of artificial intelligence software, and has learned appropriate learning parameters. The detection means 42 executes the trained model 48 as a program on, for example, a CPU or GPU, and when a photographed image is input, estimates the type of construction of the wall construction or ceiling construction photographed in the photographed image. In this embodiment, the trained model 48 includes a trained model 48w for wall work and a trained model 48c for ceiling work, separately.

[0023] The trained model 48w for wall construction is deep-learned to detect wall construction being performed in a wall construction portion when the wall construction portion is captured in a photographed image and to estimate the type of construction of the detected wall construction. Specifically, the trained model 48w for wall construction is trained to appropriately detect which of a plurality of construction types the wall construction being performed in the wall construction portion captured in the photographed image is, including wall foundation work using LGS (Light gauge steel) or the like, insulation work, spraying work, board work, putty filling work, finishing work, and the like. In one captured image, multiple wall constructions may be captured overlapping each other in the horizontal direction of the captured image or in the depth direction perpendicular to the captured image. In this embodiment, in order to estimate all of the multiple wall constructions captured in one captured image, an object detection model such as faster R-CNN, YOLO, or SSD (Single Shot MultiBox Detector) is used as the trained model 48w for wall construction. In the detection means 42, the captured image is input to the trained model 48w for wall construction, and the type of wall construction in the captured image is estimated. As described above, since the trained model 48w for wall construction in this embodiment is an object detection model, it is possible that multiple wall construction projects and multiple types of wall construction projects are detected from one image, and it is also possible that no wall construction is detected in the captured image.

[0024] When the trained model 48w for wall construction estimates multiple wall constructions from a single captured image, it simultaneously outputs, for each wall construction, coordinate information indicating the range of a bounding box corresponding to the portion estimated to be the wall construction in question. Fig. 6 is a diagram showing an example of a captured image P detected by the detection means 42. In Fig. 6, two wall construction projects are detected in the captured image P, and the construction ranges corresponding to each of these are shown enclosed by bounding boxes Bw. In this embodiment, a combination of the minimum X coordinate (horizontal direction) value Xmin, the maximum X coordinate value Xmax, the minimum Y coordinate (vertical direction) value Ymin, and the maximum Y coordinate value Ymax is used as the coordinate information indicating the range of the bounding box. In this embodiment, the origin coordinate of the image is the upper left, and therefore the minimum Y coordinate value Ymin is located above the maximum Y coordinate value Ymax. Alternatively, a combination of the minimum X coordinate value, the minimum Y coordinate value, the width of the bounding box in the X direction, and the height of the bounding box in the Y direction may be used as the coordinate information indicating the range of the bounding box. In this embodiment, each coordinate value of the bounding box is calculated by dividing the coordinate value of the actual pixel in the captured image by the resolution of the captured image. For example, if the number of pixels in the X direction of the captured image is 1280, and the pixel located at the coordinate 320 from the left end has a value of 320 / 1280=0.25 as its coordinate information.

[0025] The detection means 42 creates a wall construction type estimation record, which is a record of wall construction that has been estimated. FIG. 7 is a diagram showing an example of a wall construction type estimation record created by the detection means 42. As shown in FIG. 7, for a captured image in which some (i.e., one or more) wall constructions have been detected by the trained model 48w for wall construction, the captured image capture time (date and time), the file name (image file name) of the captured image, the number of wall constructions detected, and the combinations (wall construction type estimation results) of the identification information of the construction type of the detected wall construction (name, or a numerical value, character string, etc. indicating the construction type) and the coordinate information indicating the range of the bounding box corresponding to the wall construction, for the number of wall constructions, are associated with each other and stored in the database 44 as text data in which, for example, one set of associated correspondences is written in one line. For example, in the example on the first line of Figure 7 where the file name is img01.jpg, one wall construction project was detected in the file, which is LGS (wall underlayment work), and the coordinates of the bounding box corresponding to the wall construction project in img01.jpg are 0.96, 0.55, 0.09, and 1.00. In addition, in the example of the file name img02.jpg on the second line of Figure 7, three wall work projects were detected for that file, the first wall work project being board work, the second wall work project being LGS (wall underlayment work), and the third wall work project being board work.

[0026] The trained model 48c for ceiling work is deep-learned to detect ceiling work being performed in a ceiling work area when the ceiling work area is photographed in a photographed image and to estimate the type of work of the detected ceiling work. Specifically, the trained model 48c for ceiling work is trained to appropriately detect which of a plurality of types of work, including ceiling foundation work, board work, equipment work, fireproof coating work, putty filling work, finishing work, and not yet completed, the ceiling work being performed in the ceiling work area photographed in the photographed image is. Regarding ceilings, multiple ceiling construction works are not generally photographed overlapping each other in one photographed image, and it is sufficient to detect one type of ceiling construction work from one photographed image. Therefore, in this embodiment, as the trained model 48c for ceiling construction, instead of an object detection model like the trained model 48w for wall construction, an image recognition model such as VGG, Residual Network (Residual Network), Xception, etc. is used to reduce the calculation load. In this embodiment, the trained model 48c for ceiling construction is an image recognition model as described above. Since the image recognition model performs class classification, it performs inference as to which of a plurality of classes the input image corresponds to. Therefore, in this embodiment, the trained model 48c for ceiling construction is trained to classify, for example, areas where there is no ceiling or places where there is no need to estimate the progress status into classes such as "outdoors" and "others". This prevents, for example, the sky from being mistaken for some type of construction.

[0027] In the detection means 42, the captured image is input to a trained model 48c for ceiling work, and the type of work of the ceiling work in the captured image is estimated. The detection means 42 creates a ceiling work type estimation record (not shown) which is a record of estimated ceiling work. For a captured image in which ceiling work has been detected without being classified into classes such as "outdoor" or "other" by the trained model 48c for ceiling work, the ceiling work type estimation record stores in the database 44 the captured image shooting time, the file name of the captured image, and the identification information of the type of work (such as a name or a numerical value or character string indicating the type of work) of the detected ceiling work, which are associated with each other and are written, for example, as text data in which one set of associated correspondences is written on one line. The detection means 42 may also output numerical information indicating the reliability of the estimation results of the wall construction, ceiling construction, etc. Based on this numerical information, the reliability of the estimation results obtained by the management system main body 4 may be determined.

[0028] The information integration unit 43 calculates the shooting location of the captured image in which construction work was detected, associates the shooting location with the progress status of the construction work corresponding to the captured image, integrates them to create integrated data, and stores it in the database 44. Specifically, the information integration unit 43 acquires the shooting time and file name of the captured image that is registered in either or both of the wall work type estimation record and the ceiling work type estimation record, i.e., the captured image in which some wall work or ceiling work has been detected. The information integration unit 43 extracts the identification tag M detected at the time when the photographed image was taken or the closest past time, i.e., immediately before the photographed image was taken, from the identification tag detection history, and acquires the identification information of the identification tag M and the coordinates of the corner Mc in the photographed image. The information integration unit 43 acquires the horizontal position in the construction site G where the identification tag M is installed and the direction in which the identification tag M is posted, which correspond to the extracted identification information of the identification tag M, from the correspondence relationship related to the identification tag M stored in the database 44. The information integration unit 43 calculates the relative position from the identification tag M at which the photographed image was taken, taking into account the shape of the identification tag M in the photographed image. The information integration unit 43 estimates the photographing position and photographing direction of the photographed image in which the identification tag M was taken, for example, as an absolute position in the construction site G, from the relative position, the horizontal position of the identification tag M, and the direction in which the identification tag M is posted. As described above, the information integration unit 43 estimates the shooting position of the image in which the identification tag M detected just before was taken, using the position where the identification tag M detected just before was placed as the base point. The information integration unit 43 further calculates the shooting position of the image in which construction work was detected based on the movement information of the site worker Q generated by the movement information calculation unit 22, using the shooting position as the base point. More specifically, the information integration unit 43 acquires, for example, from the movement information of the site worker Q generated by the movement information calculation unit 22, the relative movement amount from the shooting time of the image in which the identification tag M detected just before was taken to the shooting time of the image in which wall construction work or ceiling construction work was detected, and adds this relative movement amount to the shooting position of the image in which the identification tag M detected just before was taken, thereby calculating the shooting position of the image in which construction work was detected as an absolute position within the construction site G, for example. In addition, the information integration unit 43 acquires the orientation when the image in which construction work was detected was taken based on the shooting time of the image in which wall construction work or ceiling construction work was detected in the movement information of the site worker Q.

[0029] When wall construction work is detected in the photographed image, the information integration unit 43 obtains from the wall construction type estimation record the number of detected wall construction works associated with the photographed image, and combinations of identification information for the construction type of the detected wall construction work and coordinate information indicating the range of the bounding box corresponding to the wall construction work, equal to the number of wall construction works. When ceiling work is detected in a captured image, the information integration unit 43 obtains identification information of the work type of the detected ceiling work associated with the captured image from the ceiling work type estimation record.

[0030] The information integration unit 43 associates these pieces of data with each other and integrates them to create integrated data. Fig. 8 is a diagram showing an example of integrated data created by the information integration unit 43. As shown in Fig. 8, the integrated data includes the photographing time (date and time) of the photographed image registered in either one or both of the wall work type estimation record and the ceiling work type estimation record, that is, for a photographed image in which some wall work or ceiling work has been detected, the file name (image file name), the photographing position and orientation (XY coordinates, orientation) of the photographed image, the identification information (marker ID) of the identification tag M detected immediately before, the number of wall works detected, the combination (wall work estimation result) of the identification information of the work type of the detected wall work and the coordinate information indicating the range of the bounding box corresponding to the wall work, and the identification information of the work type of the detected ceiling work (ceiling work estimation result) are associated with each other, and are stored in the database 44 as text data in which, for example, one set of associated correspondence is written in one line. In cases where no wall construction was detected in a captured image, but ceiling construction was detected (not classified into a class such as "outdoor" or "other"), for example, the number of wall construction instances will be recorded as 0 for that image, no identification information for the type of wall construction and no coordinate information indicating the range of the bounding box will be recorded, and only information relating to the ceiling construction will be associated with that image. In addition, in cases where wall construction is detected in a captured image but ceiling construction is not detected (classified into classes such as "outdoor" or "other"), for example, information regarding ceiling construction is not associated with the captured image, and only information regarding wall construction is associated with the captured image.

[0031] The construction area estimation unit 45 refers to each association in the integrated data one by one, and estimates the location and area where construction work is being performed based on the shooting position of the captured image in which the wall construction work and ceiling construction work are taken. The construction area estimation unit 45 estimates the location and area where the wall construction work and ceiling construction work are being performed within the construction site G based on the shooting position and shooting direction of the captured image in which the area where the wall construction work and ceiling construction work are being performed is captured among the multiple captured images. Fig. 9 is a diagram showing the results of estimation of the position and range of wall construction by the construction range estimation unit. In Fig. 9, the position and range Wm where wall construction is being carried out are displayed, for example, by coloring, on the construction drawing Z of the building. In addition, the position and range Wm where wall construction is being carried out are displayed, for example, by coloring, to show the type of construction of the wall construction. For example, with regard to wall construction, the construction area estimation unit 45 refers to the integrated data stored in the database 44 and estimates the location and area of ​​the wall construction by performing the following procedure for each correspondence in the integrated data (each line in the text data).

[0032] FIG. 10 is an explanatory diagram showing a method for estimating the position and range of wall construction by the construction range estimation unit. When the construction scope estimation unit 45 acquires one correspondence relationship in the integrated data, it acquires the identification information of the identification tag M in the integrated data, and acquires the height position (floor) where the identification tag M corresponding to the identification information is provided by referring to the database 44. The construction scope estimation unit 45 sets a reference point P0 corresponding to the shooting position and a reference direction D corresponding to the orientation on the construction drawing Z corresponding to the height position, based on the shooting position (XY coordinates) and orientation of the captured image in the integrated data.

[0033] Next, one combination of identification information of the construction type and coordinate information indicating the range of the bounding box is obtained from the integrated data. Of the coordinate information indicating the range of the bounding box, the minimum and maximum X coordinates are obtained, and it is calculated which range A1 the range between these minimum and maximum X coordinates is located in within an angle range A of 90° in total, 45° left and right from reference direction D. For example, when the minimum X coordinate is 0.10 and the maximum X coordinate is 0.45, then 0.1×90°=9°, and 0.45×90°=40.5°, and so it is clear that range A1 is between 9° and 40.5°. The construction range estimation unit 45 extends the line L1 corresponding to 9° and the line L2 corresponding to 40.5° from the reference point P0, and if the wall line S on the construction drawing Z is located between these lines L1 and L2, it estimates that the wall line S within the range A1 is a wall on which construction of the construction type whose identification information corresponds to the coordinate information indicating the range of the bounding box is being carried out on the construction drawing Z. The construction range estimation unit 45 registers in the database 44 the area Wm1 between the lines L1 and L2 on the reference point P0 side of the wall line S and the identification information of the construction type in association with each other. When the construction drawing Z is data that does not include the coordinate values ​​of the walls, such as an image, the pixel coordinate values ​​corresponding to the wall line S in the image are defined and recorded in an external file, etc., and this is referenced to determine the position of the wall line S in the construction drawing Z. When the construction drawing Z is data that includes the coordinate values ​​of the walls, such as BIM, the position of the wall line S in the construction drawing Z is determined by directly referencing the coordinate values ​​of the walls in the data. The construction area estimation unit 45 performs the above operation for each wall construction detected in the captured image, that is, for all combinations of construction type identification information and coordinate information indicating the range of the bounding box.

[0034] 10, the construction area estimation unit 45 associates the area Wm1 between the lines L1 and L2 on the reference point P0 side (i.e., the lower side) of the wall line S with the identification information of the construction type and registers it in the database 44. In this way, even if a reference point P3 is provided on the opposite side (i.e., the upper side) of the reference point P0 with respect to the wall line S during the above processing of other photographed images in the integrated data, and an area Wm2 is calculated so as to overlap with the area Wm1 between the lines L1 and L2 with respect to the same wall line S with respect to the reference point P3, these can be treated as different progress results. Alternatively, as simply shown in FIG. 11, the progress result of one wall construction may be associated with one wall line S, thereby reducing the amount of processing.

[0035] In addition, when performing the above-mentioned process, the wall line S is searched for by moving away from the reference point P0, and the start point P1 from which the search for the wall line S starts can be set according to the maximum Y coordinate value Ymax of the coordinate information indicating the range of the bounding box. That is, the relationship between the distance from the reference point P0 and the maximum Y coordinate value Ymax can be examined in advance, and the start point P1 on the construction drawing Z can be determined by a regression curve or the like using the maximum Y coordinate value Ymax as a variable. This prevents, for example, when the wall lines S exist in parallel and the wall on the near side is located closer to the start point P1, from erroneously applying progress information about the wall on the near side to another wall on the far side.

[0036] The construction area estimation unit 45 stores in the database 44 the information indicating the estimated location and area where the wall construction is being carried out, the height position (floor), and the construction type of the wall construction in association with each other.

[0037] Fig. 11 is a diagram showing the result of estimation of the position and range of ceiling work by the work range estimation unit. In Fig. 11, the position Cm where the ceiling work is being performed is displayed, for example, by coloring, on the construction drawing Z of the building. In addition, the position Cm where the ceiling work is being performed is displayed, for example, by color coding, to show the type of work of the ceiling work. For ceiling work, similarly to wall work, when the work scope estimation unit 45 acquires one correspondence in the integrated data, it acquires the identification information of the identification tag M in the integrated data, and acquires the height position (floor) where the identification tag M corresponding to the identification information is provided by referring to the database 44. The work scope estimation unit 45 sets a reference point corresponding to the shooting position and a reference direction corresponding to the orientation on the work drawing Z corresponding to the height position, based on the shooting position (XY coordinates) and orientation of the captured image in the integrated data. The construction range estimation unit 45 estimates that the area corresponding to a circle as shown in Figure 11, centered on the reference point on the construction drawing Z or a position a certain distance away from the reference point in the reference direction, is the position and range corresponding to the ceiling work in the integrated data. The work scope estimation unit 45 stores in the database 44 the information indicating the estimated location and scope of the ceiling work, the height (floor), and the type of the ceiling work in association with each other. In this embodiment, the position and range corresponding to the ceiling construction in the construction drawing Z are estimated as a detailed position as described above. Alternatively, for example, the construction drawing Z may be divided and partitioned in advance into rooms or areas of a certain range, and the room or the entire area to which the shooting position belongs may be estimated to be the position and range where the ceiling construction is being performed.

[0038] The information display unit 46 refers to the database 44, and for each floor, creates display data R as shown in Figs. 9 and 11 by overlaying the progress status (type of work) of the wall work and ceiling work on the position and range where the wall work and ceiling work are being performed, estimated by the work range estimation unit 45, on the construction drawing Z of the building corresponding to that floor, and outputs it to the display device 5 as an estimated result of the progress information of the work. For example, when the user of the display device 5 specifies a floor of the building, the information display unit 46 can be configured to obtain information indicating the position and range where the work is being performed, associated with that floor, and based on this, create display data by overlaying the construction drawing of the specified floor and the estimated result of the progress information of the work within the construction site G.

[0039] When creating the display data R, the information display unit 46 estimates the rate of completion for each of a plurality of types of work by dividing the completed quantity of that type of work by the total quantity of that type of work. The information display unit 46 calculates the completed quantity of each type of wall or ceiling work from the progress (type of work) of wall or ceiling work estimated as described above based on the type, area, and thickness of the materials constituting the walls and ceilings registered in the database 44. For example, if a certain type of work is estimated for a certain wall, the information display unit 46 assumes that the process corresponding to that type of work has been completed, thereby estimating the completed quantity of each type of work for the wall or ceiling. The information display unit 46 estimates the completion rate (%) of the work by dividing the cumulative completed quantity of that type of work estimated for all wall and ceiling work in the building by the total quantity of that work calculated from the work breakdown sheet, BIM, etc. The information display unit 46 calculates the amount of work by multiplying the rate of work completed calculated as above by the fee for the work. Here, for example, by inputting the corresponding contractor information for each part into BIM, it is possible to estimate the amount of work completed by each contractor. In this manner, the information display unit 46 calculates the volume rate and volume, and stores the calculated volume rate and volume in the display data R so that they can be displayed by the display device 5, for example.

[0040] The display device 5 may be a monitor device or the like provided in the management system main body 4, or may be a smartphone, a tablet terminal, or the like capable of wireless data communication with the information display unit 46. The information display unit 46 transmits and displays the display data R to the display device 5 via, for example, a communication cable, a wireless LAN such as Wi-Fi, data transfer via a mobile phone communication network, or the like. On the display device 5, the progress of the construction work is superimposed on the construction drawing of the building, on the position and range of the construction work estimated by the construction range estimation unit 45, based on the display data R transmitted from the information display unit 46. In addition to this, the display device 5 may simultaneously display the completion rate and the completion amount calculated in the information display unit 46.

[0041] FIG. 12 is a flowchart showing the flow of a method for estimating the progress of wall and ceiling construction work, which is executed using the construction progress monitoring system of this embodiment. As shown in FIG. 12, in order to estimate the progress of wall construction and ceiling construction using the construction progress monitoring system 1 in this embodiment, first, an identification tag M is placed within the construction site G (step S11). For example, if the construction site G has multiple floors, the identification tag M is installed on each floor (floor). Also, it is preferable to install the identification tag M near stairwells or elevators, which are areas of high traffic within the construction site G and tend to be the starting or ending point of people's movements on each floor. Also, the identification tag M may be installed in multiple locations on each floor. It is desirable to attach the identification tag M to the surface of a pillar or wall under construction. In particular, when the building to be constructed has multiple floors, a piece of paper with the floor number written on it is often attached to a pillar or the like on each floor. In such a case, by attaching the identification tag M near the floor number display, the identification tag M can be installed at the same time as the floor number display is posted. In this way, there is no need to go around the building to install the identification tag M. Furthermore, if the identification tag M is attached with a magnet or adhesive, it will be easy to remove. After the identification tags M are installed, information about the identification tags M is registered in the database 44 of the management system main body 4 (step S12). Specifically, as described above, the database 44 stores the identification information of each identification tag M in association with information about the horizontal position and height position (floor) within the construction site G where the identification tag M is installed, and the direction (azimuth or angle) in which the identification tag M is posted. The above-mentioned step S12 needs to be performed once at the construction site G in conjunction with the installation of the identification tag M in step S11.

[0042] In step S13, the site worker Q holds or wears the pedestrian autonomous navigation means 2 and the image recording device 3, and photographs the inside of the construction site G while walking and moving within the construction site G. At this time, it is preferable that the site worker Q moves within each floor from a position near where the identification tag M is installed on each floor as a starting point. At this time, the pedestrian autonomous navigation means 2 generates movement information of the field worker Q based on the detection data from the sensor 21. Furthermore, if the identification tag M is located ahead of the photographing unit 31 of the photographing and recording device 3, the photograph of the identification tag M is taken by the photographing and recording device 3. Considering the purpose of detecting the progress of the construction work in the construction site G by the construction progress assessment system 1, it is desirable for the on-site worker Q to move within the construction site G so that as many of the planned construction work locations within the construction site G as possible are photographed by the photographing and recording device 3. However, during regular patrols of the construction site G for the purpose of safety estimation or the like, the on-site worker Q may hold or wear the pedestrian autonomous navigation means 2 and the photographing and recording device 3, so that the movement within the construction site G for detecting the progress of the construction work within the construction site G also serves as the patrol. Alternatively, it is also possible to improve the comprehensiveness of the photographing by the photographing and recording device 3 within the construction site G by having a large number of on-site workers Q always carry the pedestrian autonomous navigation means 2 and the photographing and recording device 3 in a manner that does not interfere with each of their work.

[0043] As a result, in the pedestrian autonomous navigation means 2, the sensor 21 and the movement information calculation unit 22 acquire movement information of the site worker Q within the construction site G, and the movement information is stored in the data storage unit 23 in association with time information. In addition, in the photography and recording device 3, the photography unit 31 photographs the construction site G located around the movement path of the field worker Q, and the photographed image is stored in the photography data memory unit 32 as photography data associated with the photography time, which is the time when the photographed image was photographed. Next, the movement information of the site worker Q obtained by the pedestrian autonomous navigation means 2 and the photographed data of the inside of the construction site G taken by the photographing and recording device 3 are transferred from the data output unit 24 and the photographed data output unit 33 to the management system main body 4 (step S14). Note that this step S14 may be performed after the completion of the above step S13, or may be performed at any time while the above step S13 is being executed, using communication via wireless LAN or the like.

[0044] When the management system main body 4 receives the movement information of the site worker Q transferred in step S14 and the input of the photographed data within the construction site G, it stores these in the database 44. The management system main body 4 detects the progress of the construction work contained in the photographed images of the photographed data (step S15). To this end, first, if the captured image is a moving image, the detection means 42 of the management system main body 4 decomposes the moving image into multiple still images, and executes the following processes using each of the decomposed still images as the captured image. Then, the detection means 42 detects the identification tag M from the captured image to be processed by a known image processing method or the like. The detection means 42 creates an identification tag detection history, which is a history of when the identification tag M has been detected.

[0045] The detection means 42 also inputs the captured images to a trained model 48w for wall work and a trained model 48c for ceiling work. The detection means 42 uses the trained model 48w for wall work to detect the presence or absence of wall work in each captured image, and estimates the type of work for the wall work. The detection means 42 uses the trained model 48c for ceiling work to detect the presence or absence of ceiling work in each captured image, and estimates the type of work for the ceiling work. The detection means 42 creates a wall construction type estimation record, which is a record of the wall construction that has been estimated. Similarly, the detection means 42 creates a ceiling work type estimation record, which is a record of the ceiling work that has been estimated.

[0046] Next, the information integration unit 43 of the management system main body 4 generates integrated data on the progress of the wall construction and ceiling construction (step S16). Specifically, the information integration unit 43 acquires the shooting time and file name of the captured image that is registered in either or both of the wall work type estimation record and the ceiling work type estimation record, i.e., the captured image in which some wall work or ceiling work has been detected. As already explained, the information integration unit 43 extracts from the identification tag detection history the identification tag M detected at the same time as the photographing time of the photographed image or the closest past time, i.e., immediately before the photographing of the photographed image, and acquires the identification information of the identification tag M and the coordinates of the corner Mc in the photographed image. The information integration unit 43 calculates the relative position from the identification tag M at which the photographed image was taken, taking into account the shape of the identification tag M in the photographed image, based on the coordinates of the corner Mc in the photographed image at which the identification tag M was taken. The information integration unit 43 estimates the photographing position and photographing direction of the photographed image at which the identification tag M was taken as an absolute position within, for example, the construction site G, from the relative position, the horizontal position of the identification tag M, and the orientation of the display of the identification tag M. As described above, the information integration unit 43 estimates the shooting position of the captured image in which the most recently detected identification tag M was captured, using the position where the most recently detected identification tag M was placed as a base point. The information integration unit 43 further calculates the shooting position of the captured image in which construction work was detected, based on the movement information of the site worker Q generated by the movement information calculation unit 22, using the shooting position as a base point. In addition, when wall construction work is detected in the photographed image, the information integration unit 43 obtains from the wall construction type estimation record the number of detected wall construction works associated with the photographed image, and combinations of identification information for the construction type of the detected wall construction work and coordinate information indicating the range of the bounding box corresponding to the wall construction work, equal to the number of wall construction works. Furthermore, when ceiling work is detected in the captured image, the information integration unit 43 obtains identification information of the work type of the detected ceiling work associated with the captured image from the ceiling work type estimation record. The information integration unit 43 associates these pieces of data with each other and integrates them to create integrated data.

[0047] Next, the construction area estimation unit 45 refers to each association in the integrated data one by one, and estimates the location and area where the construction work is being performed based on the shooting position of the captured image in which the wall construction work and ceiling construction work are taken. The construction area estimation unit 45 estimates the location and area where the wall construction work and ceiling construction work are being performed within the construction site G based on the shooting position and shooting direction of the captured image in which the area where the wall construction work and ceiling construction work are being performed is captured among the multiple captured images. Specifically, when the construction scope estimation unit 45 acquires one correspondence relationship in the integrated data, it acquires the identification information of the identification tag M in the integrated data, and acquires the height position (floor) where the identification tag M corresponding to the identification information is provided by referring to the database 44. The construction scope estimation unit 45 sets a reference point corresponding to the shooting position and a reference direction corresponding to the orientation on the construction drawing Z corresponding to the height position, based on the shooting position (XY coordinates) and orientation of the captured image in the integrated data. The construction area estimation unit 45 estimates the position and area where wall construction and ceiling construction are being carried out, based on the reference points and reference directions set as described above. The construction scope estimation unit 45 stores in the database 44 information indicating the estimated location and scope of the wall and ceiling construction work, the height position (floor), and the type of construction work for the wall and ceiling work, in association with each other.

[0048] In this manner, after the association of various data in the database 44 in the management system main body 4 has been completed, when a user accesses the management system main body 4 and requests the display of information in the database 44 indicating the progress of construction work at the construction site G, the information display unit 46 creates display data R of information indicating construction progress information based on the various data required for displaying the information (step S17). When creating the display data R, the information display unit 46 estimates the rate of completion for each of a plurality of types of work by dividing the completed quantity of that type of work by the total quantity of that type of work. The information display unit 46 calculates the amount of work by multiplying the rate of work by the fee for the construction work. The information display unit 46 stores the calculated volume rate and volume in the display data R, for example. The information display unit 46 transmits the display data R to the display device 5 for display (step S18). On the display device 5, the progress of the construction work is superimposed on the construction drawing of the building, on the position and range of the construction work estimated by the construction range estimation unit 45, based on the display data R transmitted from the information display unit 46. In addition to this, the display device 5 may simultaneously display the completion rate and the completion amount calculated in the information display unit 46.

[0049] The construction progress assessment system 1 as described above is a construction progress assessment system 1 that manages the progress of wall construction and ceiling construction within a construction site G, and includes an identification tag M that is placed at different positions within the construction site G and has identification information recorded thereon, a pedestrian autonomous navigation means 2 that is configured to be held or worn by a site worker Q or a mobile body configured to be movable within the construction site G and acquires movement information of the site worker Q or the mobile body, an imaging and recording device 3 that is configured to be held or worn by the site worker Q or the mobile body and captures images within the construction site G, and detects construction work including wall construction and ceiling construction from images captured by the imaging and recording device 3 and estimates the progress of these works, and identification tag M; an information integration unit 43 which calculates the shooting position of a photographed image in which construction work has been detected based on movement information, using the position at which the identification tag M detected immediately before the photographed image was placed as a base point, and stores the shooting position and the progress of the construction work corresponding to the photographed image in association with each other in a database 44; a construction range estimation unit 45 which estimates the location and range where the construction work is being carried out based on the shooting position; and an information display unit 46 which creates display data R by overlaying the progress of the construction work on the location and range of the construction work estimated by the construction range estimation unit 45 on the construction drawing Z of the building, and displays the display data R on the display device 5. According to this configuration, when the site worker Q or the moving body holds or wears the pedestrian autonomous navigation means 2 and the photographing and recording device 3 and photographs the inside of the construction site G with the photographing and recording device 3 while moving within the construction site G, the pedestrian autonomous navigation means 2 obtains movement information indicating the position and route of the site worker Q or the moving body when moving within the construction site G. Furthermore, the photographing and recording device 3 photographs the inside of the construction site G, and obtains photographed images of the construction work including the wall construction and ceiling construction within the construction site G. The detection means 42 detects the construction work including the wall construction and ceiling construction within the construction site G from the photographed images of the photographing and recording device 3, and estimates the progress of the construction work. Furthermore, the detection means 42 detects the identification tag M placed within the construction site G. For the photographed images in which construction work is detected, the information integration unit 43 calculates the photographing position of the photographed image based on the movement information of the site worker Q or the moving body, using the position where the identification tag M detected immediately before the photographing of the photographed image as the base point. As a result, the photographing position of the photographed image can be identified with high accuracy even in high-rise buildings and large-scale structures. The information integration unit 43 also associates the photographing position of the photographed image calculated as described above with the progress of the construction work corresponding to the photographed image, and stores them in the database 44. The construction range estimation unit 45 also estimates the position and range of the construction work based on the photographing position. The information display unit 46 displays the progress of various construction works, including wall construction and ceiling construction, superimposed on the position and range of the construction work estimated by the construction range estimation unit 45 on the construction drawing Z of the building. In this way, the progress of the wall construction and ceiling construction work is superimposed on the building drawing, so that the position and progress of the construction work can be easily grasped. In particular, in the above-mentioned configuration, the identification tag M, which is used as a base point for identifying the photographing position of the photographed image, may be provided only in places where the site worker Q frequently passes, for example, on each floor of a building. In other words, it is not necessary to provide an identification tag M individually for each part, such as a wall or ceiling, that is the subject of construction, and therefore the number of identification tags M to be installed is reduced. In addition, since the progress is automatically estimated by the detection means from the photographed image, the site worker Q is not required to report or input data into the system for managing the progress, even if the work progresses. This further improves ease of operation. As a result, it is possible to provide a construction progress monitoring system 1 that can more accurately and simply manage the progress of wall construction and ceiling construction.

[0050] In addition, the detection means 42 estimates which of a plurality of construction types is being carried out as the progress status, and the information display unit 46 displays the progress status and, for each of the plurality of construction types, estimates and displays the completion rate by dividing the quantity of construction work completed for that construction type by the total quantity for that construction type. According to this configuration, the detection means 42 estimates which of the multiple construction types is being carried out as the progress status. Based on the estimation result, the completed quantity of each of the multiple construction types is divided by the total quantity of the construction type, so that the completion rate of the construction can be more easily grasped.

[0051] The detection means 42 also has a trained model 48 that has been trained using construction images taken of the construction site and the progress status corresponding to the construction images as learning data, and inputs the captured images into the trained models 48 (trained model 48w for wall work, trained model 48c for ceiling work) to estimate the progress status of wall work as one of a number of construction types including wall underlayment work, insulation work, board work, and finishing work, or to estimate the progress status of ceiling work as one of a number of construction types including ceiling underlayment work, board work, equipment work, and fire-resistant coating work. According to this configuration, by subjecting wall construction images and ceiling construction images to machine learning, the specific construction type of wall construction and ceiling construction can be estimated from each captured image. Therefore, it becomes possible to grasp the progress of construction more precisely.

[0052] In addition, the information display unit 46 displays the progress of the construction work on the construction drawing Z by superimposing it on the wall and ceiling parts corresponding to the construction work. According to this configuration, the progress of wall construction and ceiling construction can be displayed superimposed on the wall and ceiling parts corresponding to the construction on the construction drawing Z, allowing the progress of the wall construction and ceiling construction to be grasped more intuitively.

[0053] Moreover, an identification tag M is provided for each floor of the building. For example, when the pedestrian autonomous navigation means 2 acquires movement information using GPS, it is not easy to acquire information regarding height with sufficient accuracy, and therefore it is difficult to improve the accuracy of the movement information, especially when the building is a high-rise building. In contrast, with the above-mentioned configuration, when calculating the photographing position of the photographed image based on the movement information of the field worker Q or the moving object with the position where the identification tag M is placed as the base point, the height position where the photographed image was taken can be obtained by referring to the detection result of the identification tag M. Therefore, even if the movement information does not include information on height acquired by the pedestrian autonomous navigation means 2, or even if there is such information, the accuracy of which is unreliable, it is possible to accurately grasp the height position within the building.

[0054] (Modification of the embodiment) The construction progress monitoring system of the present invention is not limited to the above-described embodiment explained with reference to the drawings, and various modifications are possible within the technical scope. For example, as described in the above embodiment, an omnidirectional camera can be used as the image capturing and recording device 3. In this case, the 360° captured image by the omnidirectional camera is divided and converted into four images, an image in the angle range of 0° to 90° on the front side T1 of the site worker Q, an image in the angle range of 90° to 180° on the right side T2 of the front side T1, an image in the angle range of 180° to 270° on the opposite side T3 of the front side T1, and an image in the angle range of 270° to 360° (0°) on the left side T4 of the front side T1, in Fig. 10, and each of these images is subjected to the same processing as in the above embodiment. In this way, when an omnidirectional camera is used, one image is divided into four images, and wall construction is detected in each of these four images. Therefore, it is necessary to identify which of the four images the image in which wall construction was detected corresponds to. Therefore, when an omnidirectional camera is used, it is desirable to add an identifier for this identification to the wall construction type estimation record.

[0055] FIG. 13 is a diagram showing an example of wall construction type estimation record when an omnidirectional camera is used. In the above embodiment, as shown in FIG. 7, the wall construction type estimation result of the wall construction type estimation record is expressed by a combination of the number of wall constructions detected, and the identification information of the construction type of the detected wall construction and the coordinate information indicating the range of the bounding box corresponding to the wall construction, as many as the number of wall constructions. Instead, in the wall construction type estimation result of the wall construction type estimation record of this modified example, as shown in FIG. 13, the number of wall constructions detected, and the identification information of the construction type of the detected wall construction, the coordinate information indicating the range of the bounding box corresponding to the wall construction, and the image identification number, as many as the number of wall constructions. The image identification number can be set to 0 when the image corresponds to the front side T1 of the 360° photographed image by the omnidirectional camera, 1 when the image corresponds to the right side T2, 2 when the image corresponds to the opposite side T3, and 3 when the image corresponds to the left side T4. In the case of FIG. 13, in the example in which the file name is img01.jpg in the first line, the image identification number is 1, so it can be seen that this corresponds to the right side T2. Similarly, in the example of the file name img02.jpg on the second line, the image identification number is 2, which indicates that this corresponds to the opposite side T3. Fig. 14 is a diagram showing an example of integrated data when an omnidirectional camera is used. As with the wall construction type estimation record, an image identification number is added to the integrated data obtained by integrating this with other data.

[0056] In the above embodiment, the construction area estimation unit 45 acquires the minimum and maximum X coordinates from the coordinate information indicating the range of the bounding box, and calculates which range A1 the range between the minimum and maximum X coordinates is located in within the angle range A of 90° in total, 45° to the left and right of the reference direction D. When performing this calculation, the construction area estimation unit 45 needs to acquire the image identification number and consider the direction in which the target image faces. For example, when the image identification number is 0, the construction area estimation unit 45 uses the calculated range A1 as is without rotating it, when the image identification number is 1, the calculated range A1 is rotated 90° to the right and used, when the image identification number is 2, the calculated range A1 is rotated 180° to the right and used, and when the image identification number is 3, the calculated range A1 is rotated 270° to the right and used. For example, the range of 9° to 40.5° exemplified in the above embodiment can be rotated 90° to become a range of 99° to 130.5° relative to the reference direction D if the image identification number corresponding to the image is 1.

[0057] Regarding ceiling construction, when an omnidirectional camera is used, ceiling construction can be detected based on the image on the zenith side, and the type of ceiling construction can be estimated.

[0058] (Other Modifications of the Embodiment) In addition, an object detection model may be used as the trained model 48c for ceiling work to detect multiple ceiling works and multiple types of ceiling work from a single image. Alternatively, an image recognition model may be used as the trained model 48w for wall construction, thereby achieving a configuration that does not require a large amount of calculation. In addition, when identifying the shooting position of each captured image, the height (level) of the shooting position may be simultaneously obtained along with the shooting position of the captured image based on the identification information of the identification tag M detected immediately before, and the height may be registered in the integrated data instead of the identification information of the identification tag M. Furthermore, in the above embodiment, the association of each piece of data is expressed by text data, but instead, it may be realized by a relational database such as SQL. In the above embodiment, the detection means 42 detects the identification tag M as described above, and detects the wall construction and ceiling construction, and estimates the progress status inside the management system main body 4. Instead of this, the detection means 42 may be functionally divided into the photographing and recording device 3 and the management system main body 4 so that the detection of the identification tag M by the detection means 42 is performed by the photographing and recording device 3, and the detection of the wall construction and ceiling construction, and the estimation of the progress status is performed by the management system main body 4, and these may be configured to operate as the detection means 42 as a whole. Alternatively, the detection means 42 may be functionally divided into the photographing and recording device 3 and the management system main body 4 so that the detection of the identification tag M by the detection means 42 is performed by the management system main body 4, and the detection of the wall construction and ceiling construction, and the estimation of the progress status is performed by the photographing and recording device 3, and these may be configured to operate as the detection means 42 as a whole. Furthermore, the entire function of the detection means 42 may be configured to be executed by the photographing and recording device 3. In addition, the configurations described in the above embodiments can be selected or changed as appropriate without departing from the spirit of the present invention. [Explanation of symbols]

[0059] 1 Construction progress monitoring system 48 Trained model 2 Pedestrian autonomous navigation method 48c Trained model for ceiling construction (trained model) 3. Photographing and recording device 48w Pre-trained model for wall construction (Pre-trained model) 5 Display device G Construction site 42 Detection means M Identification tag 43 Information integration unit P Photographed image 44 Database Q Field worker 45 Construction scope estimation section R Display data 46 Information display section Z Construction drawing

Claims

1. A construction progress monitoring system for managing the progress of wall construction and ceiling construction at a construction site, Identification tags are placed at different positions within the construction site and have identification information recorded thereon; A pedestrian autonomous navigation means includes an autonomously operating sensor, and is configured to be held or attached by a site worker or a mobile body configured to be movable within the construction site, and acquires movement information of the site worker or the mobile body; An image recording device configured to be held or attached by the site worker or the moving body and adapted to capture images of the inside of the construction site; a detection means for detecting the wall construction and the ceiling construction from the image captured by the recording device and estimating the progress of the construction work, using a trained model in which a construction image of the construction site and the progress status corresponding to the construction image are trained as training data, and for detecting the identification tag by image processing the captured image; an information integration unit that acquires the position of the identification tag detected immediately before the image in which the construction work was detected was taken, calculates the image position of the image taken after the movement based on the movement information using the acquired position of the identification tag as a base point, associates the image position with the progress of the construction work corresponding to the image, and stores the associated data in a database; A construction area estimation unit that estimates the location and area where the construction work is being carried out from the integrated data based on the shooting position; an information display unit that creates display data by superimposing the progress of the construction work on the position and the range of the construction work estimated by the construction range estimation unit on a construction drawing of a building, and displays the display data on a display device; A construction progress monitoring system comprising:

2. The detection means estimates which of a plurality of construction types is being carried out as the progress status, The construction progress monitoring system described in claim 1, characterized in that when creating the display data, the information display unit estimates the completion rate for each of the multiple construction types by dividing the quantity of construction work completed for that construction type by the total quantity for that construction type.

3. The detection means inputs the captured image into the trained model, With respect to the wall construction, the progress status is estimated to be one of a plurality of construction types including wall foundation construction, insulation construction, board construction, and finishing construction, or The construction progress monitoring system according to claim 1 or 2, characterized in that, with regard to the ceiling construction work, the progress is estimated as one of a plurality of construction types including ceiling underlayment work, board work, equipment work, and fire-resistant coating work.

4. The construction progress monitoring system according to any one of claims 1 to 3, characterized in that the information display unit displays the progress status of the construction work by superimposing it on the wall and ceiling parts corresponding to the construction work on the construction drawing.

5. 5. The construction progress monitoring system according to claim 1, wherein the identification label is provided for each floor of the building.

Images