Construction management support device, construction management support program, and storage medium

Abstract

To provide a construction management support device, a construction management support program and a storage medium which generate a daily work report by using a voice recognition technique.SOLUTION: A construction management support device 100 comprises: a storage unit which stores worker specification information for recognizing a worker; an acquisition unit ACQ which acquires image information of object construction imaged by imaging means; an image analysis unit which refers to the worker specification information, analyzes the acquired image information and recognizes the worker; a measurement unit CNT which measures the work time of the recognized worker; and a display unit DIS which displays work information including the recognized worker and the work time associated with the worker.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a construction management support device, a construction management support program, and a storage medium, and in particular to a construction management support device, a construction management support program, and a storage medium that use image information to recognize workers, work content, or machinery, and display information related to the work. [Background technology]

[0002] Traditionally, cost management systems have been used for civil engineering and construction work. The on-site agent is the one who inputs the data. The on-site agent has many roles, such as holding meetings with the client, holding meetings with subcontractors, giving instructions on-site, coordinating the schedule with other contractors, managing the schedule, managing safety, and preparing and submitting documents related to the work. A typical cost management system involves entering the data of workers and machines into a specified spreadsheet, calculating the work hours and amount of work from the data, and then calculating and displaying the budget and cost. As a prerequisite for this cost management, labor management and machine management, which manage the work hours and amount of work of workers and machines, are required, but in a busy environment, it is difficult for on-site agents to find the time to input data into the system every day (preparing daily reports).

[0003] As a conventional technique, there is a "civil engineering work cost management device" (see Patent Document 1). The purpose of this is to "make it easy to input the amount of work completed at the site of civil engineering work, accurately grasp the cost of the civil engineering work, and constantly manage the progress." This conventional technique is a cost management technique that "inputs budget data from an execution budget file, creates an input form in a daily report format, an input form in a completed format containing budget data, or an outsourcing lump-sum input format containing budget data, and displays it on a display device, and inputs from an input device 5 the ratio of how much work has been completed compared to the planned amount of work as a ratio to the budget of resource usage, or inputs the quantity of resources used, and calculates the progress rate based on the budget and the results, and creates a cost management table and a progress status management table." [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent Application No. 9-340835 Summary of the Invention [Problem to be solved by the invention]

[0005] However, although the above-mentioned cost management system can automatically calculate values ​​and is convenient, it still does not solve the fundamental problem that the user must regularly and accurately input the work time, work content, and work volume. Such initial data depends on the user's input and its accuracy. If this initial data is entered accurately and appropriately regularly (for example, every day), there is no problem. This initial data is often entered by the person in charge of the site (site agent), but the current public construction sites often have complex labor forms such as joint ventures (joint venture companies or organizations for the sole purpose of the construction work), subcontractors, and sub-subcontractors, and in such a heterogeneous environment, it is difficult to guarantee accuracy and regular input. In other words, it would be convenient if it were possible to automatically and regularly recognize and input the work time and work volume of workers and machines, but such a technique has not been developed.

[0006] In addition, no mechanism or system has been developed that can verify whether the information entered by the site agent is accurate and appropriate, and whether it properly reflects the actual situation at the construction site.

[0007] On the other hand, voice recognition technology has improved in recent years, and it is now possible to recognize simple and routine conversations at a practical level. In addition, many practical AI engines have been developed.

[0008] Meanwhile, image recognition technology has progressed mainly in the camera and home electronics industries. For example, face recognition, facial expression recognition, person identification, object identification, and other technologies are increasingly being equipped in commercially available single-lens reflex cameras.

[0009] In view of this situation, the inventors of the present application have come up with the idea of ​​a technique that supports cost management by introducing voice recognition technology and AI techniques.The inventors of the present application have also come up with the idea of ​​a technique that supports cost management and construction management by using image recognition technology to recognize workers and machines, obtain work time and work content, and compare and interpolate numerical values ​​to more appropriately manage numerical values.

[0010] SUMMARY OF THE PRESENT EMBODIMENT An object of the present invention is to provide a construction management support device, a construction management support program, and a storage medium that generate a daily work report using voice recognition technology. [Means for solving the problem]

[0011] In order to solve the above-mentioned problems, a construction management support device according to a first invention is (a memory unit for storing the working budget;) A memory unit for storing a speech recognition model (including at least one of an acoustic model, a pattern match dictionary model, and a language model, and customized to correspond to construction words); (A memory unit for storing a daily work report generation algorithm for generating a daily work report based on construction information;) An acquisition unit that acquires voice; A voice recognition unit that uses the voice recognition model to generate text information based on the acquired voice and extracts construction information based on the generated text information; A generating unit that generates a daily work report including the extracted construction information based on the construction information (using a daily work report generating algorithm or using an actual budget); has.

[0012] Further, a construction management support device according to a second aspect of the present invention comprises: The acquisition unit, Get more images, The construction management support device, Further comprising an image recognition unit that generates image-derived construction information based on the acquired image, The processing unit, The construction information included in the daily work report is compared with the image-derived construction information to obtain matching image-derived construction information, and the construction information included in the daily work report is further associated with an image corresponding to the image-derived construction information that matches the construction information as evidence. It is characterized by:

[0013] Further, a construction management support device according to a third invention comprises: Further comprising a processing unit that associates the voice corresponding to the construction information included in the daily work report as evidence. It is characterized by:

[0014] Further, a construction management support device according to a fourth aspect of the present invention comprises: The construction information is At least one of the following is included: work content, machinery, tools, equipment, heavy machinery, vehicles, labor-hours information, in-house terminology, worker information (labor, target work (e.g., heavy machinery operator), and workload); It is characterized by:

[0015] Further, a construction management support device according to a fifth aspect of the present invention comprises: A trained AI model and A question generation unit that uses the trained AI model to generate question information (such as text information such as "Please input xxx" or a voice output command such as "Please talk about xxx") that prompts voice input about the construction work (when there are unclear or insufficient points in the contents of the daily work report, or simply to prompt input); Further comprising It is characterized by:

[0016] Further, a construction management support device according to a sixth aspect of the present invention comprises: It also has a trained AI model, At least one of the speech recognition unit, the processing unit, and the generation unit uses the trained AI model. It is characterized by:

[0017] Further, a construction management support device according to a seventh aspect of the present invention comprises: The acquisition unit, Further acquiring on-site audio information or on-site image information from a mobile monitoring device (e.g., a drone); The voice recognition unit or the image recognition unit, Further extracting on-site construction information from the acquired on-site voice information or on-site image information; The generation unit, Instead of generating a daily work report including the extracted construction information based on the construction information, Generate a daily work report including the construction information based on the extracted construction information and the extracted on-site construction information; It is characterized by:

[0018] Further, a construction management support device according to an eighth aspect of the present invention comprises: A command generating unit that generates command information for prompting a mobile monitoring device (such as a drone) to operate to acquire on-site audio information or on-site image information, The acquisition unit, Further acquiring on-site audio information or on-site image information from the mobile monitoring device (e.g., a drone); The voice recognition unit or the image recognition unit, Extracting on-site construction information from the acquired on-site voice information or on-site image information, The generation unit, Instead of generating a daily work report including the extracted construction information based on the construction information, Generate a daily work report including the construction information based on the extracted construction information and the extracted on-site construction information; It is characterized by:

[0019] Further, a construction management support device according to a ninth aspect of the present invention comprises: The system further includes a processing unit that associates, as evidence, voice, on-site voice information, or on-site image information corresponding to the construction information included in the daily work report. It is characterized by:

[0020] Further, a construction management support device according to a tenth aspect of the present invention comprises: A storage unit that stores machine identification information for recognizing a construction machine which is one of machines, tools, equipment, heavy machinery, or vehicles, an operation pattern of the construction machine, and a machine work conversion table including work content associated with the operation pattern of the construction machine; An acquisition unit that acquires image information of the target construction work captured by the imaging means; By referring to the machine identification information, the acquired image information is analyzed, the construction machine is recognized, and the operation pattern of the construction machine is also recognized. an image analysis unit that recognizes the operation pattern and identifies the work content of the construction machine by referring to a machine work conversion table based on the operation pattern; and The generation unit, The recognized construction machine and the work content of the construction machine are included in the daily work report. It is characterized by:

[0021] Further, a construction management support device according to an eleventh aspect of the present invention comprises: The apparatus further includes a memory unit for storing worker identification information for identifying a worker (such as the worker's physical build, height, weight, bib number, name tag, work clothes, helmet, accessories, hat, face, and other features such as shape, color, design, pattern, and QR code (two-dimensional bar code, etc.)), The acquisition unit, Further on-site image information is obtained from mobile surveillance devices (e.g. drones), The extraction unit: Recognizing a worker from the acquired site image information by referring to the worker identification information, and extracting site-derived construction information including the worker; The generation unit, Instead of generating a daily work report including the extracted construction information based on the construction information, Generate a daily work report including the construction information based on the extracted construction information and the extracted on-site construction information; It is characterized by:

[0022] Further, a construction management support device according to a twelfth aspect of the present invention comprises: A measurement unit that measures the work time of the recognized worker (for example, the time from when the worker starts to when he / she finishes the work, the time present in the image information, etc.); a calculation unit that refers to the work-per-hour information and calculates the output (amount of work) from the work time measured for each worker; Further comprising It is characterized by:

[0023] As described above, the solution of the present invention has been described as an apparatus, but the present invention can also be realized as a method, a program, or a storage medium having a program recorded thereon, which are substantially equivalent to the above, and it should be understood that these are also included within the scope of the present invention. Note that each step of the below-described method and program uses a processor such as a CPU or DSP for data processing as necessary, and stores input data, processed or generated data, etc. in a storage device such as a magnetic tape, HDD, or memory.

[0024] For example, a program according to a thirteenth aspect of the present invention for causing a computer to execute the method is as follows: A construction management support program that causes a processing device to function as the construction management support device according to the first to twelfth aspects of the present invention.

[0025] Furthermore, for example, the present invention is realized as a computer-readable recording medium, that is, a computer-readable storage medium storing a construction management support program according to a fourteenth aspect of the present invention. Effect of the Invention

[0026] According to the present invention, it is possible to generate a daily work report by utilizing voice recognition technology. [Brief description of the drawings]

[0027] [Figure 1] FIG. 1 is a block diagram showing an outline of a construction management support device according to a first embodiment of the present invention. [Diagram 2] FIG. 2 is a flowchart illustrating an example of a process executed by the construction management device illustrated in FIG. [Diagram 3] FIG. 3 is a diagram showing a schematic diagram of the process executed by the construction management device shown in FIG. 1 and the output result thereof. [Figure 4] FIG. 4 is a flowchart illustrating an example of a process executed by the construction management support device illustrated in FIG. [Diagram 5] FIG. 5 is a block diagram showing an outline of a construction management support device according to a second embodiment of the present invention. [Figure 6] FIG. 6 is a flowchart showing another example of the process executed by the construction management device shown in FIG. [Figure 7] FIG. 7 is a diagram showing a schematic diagram of work content specification information referred to in the process executed by the construction management device shown in FIG. 5 and its output result. [Figure 8] FIG. 8 is a diagram showing in schematic form the work content specification information referred to in the process executed by the construction management device shown in FIG. 5 and its output result. [Figure 9] FIG. 9 is a diagram illustrating individual identification information referenced in the process executed by the construction management device illustrated in FIG. [Figure 10] FIG. 10 is a diagram showing a schematic diagram of a state in which work contents (daily work report) are output from individual identification information referred to in the process executed by the construction management device shown in FIG. [Figure 11] FIG. 11 is a block diagram showing an outline of a construction management support device according to a third embodiment of the present invention. [Figure 12] FIG. 12 is a flowchart showing an example of a process executed by the construction management device shown in FIG. [Figure 13] FIG. 13 is a diagram showing a schematic diagram of the construction machine identification information referred to in the process executed by the construction management device shown in FIG. 11 and its output result. [Figure 14] FIG. 14 is a diagram showing in schematic form the construction machine identification information and the machine work conversion table referred to in the processing executed by the construction management device shown in FIG. [Figure 15] FIG. 15 is a diagram showing a schematic diagram of a technique for designating work area designation information that designates a work area that is a part of an image area. [Figure 16] FIG. 16 is a schematic diagram explaining a technique for measuring actual operating time by recognizing the vibration and movement of machines and tools. [Figure 17] FIG. 17 is a diagram for explaining a mechanism for outputting work information (daily output report) including planned and calculated output, using unit information and work progress information. [Figure 18] FIG. 18 is a schematic diagram illustrating one embodiment of the present invention. [Figure 19] FIG. 19 is a schematic diagram illustrating one embodiment of the present invention. [Figure 20] FIG. 20 is a schematic diagram illustrating one embodiment of the present invention. [Figure 21] FIG. 21 is a schematic diagram illustrating one embodiment of the present invention. [Figure 22] FIG. 21 is a diagram illustrating the work content specification information referenced in the process executed by the construction management device. [Figure 23] FIG. 23 is a schematic diagram illustrating one embodiment of the present invention. [Figure 24] FIG. 24 is a schematic diagram illustrating one embodiment of the present invention. [Diagram 25] FIG. 25 is a schematic diagram illustrating one embodiment of the present invention. [Figure 26] FIG. 26 is a schematic diagram illustrating one embodiment of the present invention. [Figure 27] FIG. 27 is a schematic diagram illustrating one embodiment of the present invention. [Figure 28] FIG. 28 is a schematic diagram showing the overall configuration of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0028] Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

[0029] <Embodiment 1> 1 is a block diagram showing an outline of a construction management support device according to embodiment 1 of the present invention. As shown in the figure, the construction management support device 100 (CMSD) has a control unit (CPU, arithmetic processing unit, processor) 110, an input unit IN, an output unit OUT, a communication unit COM, a memory unit 120, and a display unit DIS.

[0030] The memory unit 120 stores worker identification information WDI for recognizing workers (humans (moving objects other than objects and animals), worker physique, height, weight, bib number, name tag, work clothes, clothing, outfit, helmet, accessories, hat, face, and other characteristics such as shape, color, design, pattern, etc.). It also stores the daily work report DD, execution budget EV, and voice recognition model VOR (including at least one of an acoustic model, a pattern matching dictionary model, and a language model, and customized to correspond to words related to the construction work) generated by the construction management support device 100. Furthermore, the memory unit 120 stores the trained AI model LAM.

[0031] The control unit 110 has an acquisition unit ACQ that acquires voice (and further acquires image information of the target construction work captured by the imaging means), an image recognition unit IMA (also called an image analysis unit) that refers to the worker identification information, analyzes the acquired image information, and recognizes the worker, and a measurement unit CNT (including a clock function) that measures the work time of the recognized worker (for example, the time from when the worker starts to finish the work, or the time present in the image information, etc.). The display unit DIS displays the recognized worker and work information (such as a daily labor report) including the work time associated with the worker.

[0032] In addition, the control unit 110 further has a voice recognition unit VR that uses a voice recognition model VOR to generate text information based on the acquired voice and extracts construction information based on the generated text information, and a generation unit GEN that generates a daily work report including the construction information based on the extracted construction information.

[0033] The acquisition unit ACQ further acquires images captured by the mobile monitoring device DR and the construction monitoring camera CCM (for example, via the network NET). The control unit 110 further includes an image recognition unit IMA that generates image-derived construction information based on the acquired images, and a processing unit PRO that compares the construction information included in the daily work report with the image-derived construction information to obtain matching image-derived construction information, and further associates the construction information included in the daily work report with an image corresponding to the image-derived construction information that matches the construction information as evidence.

[0034] The processing unit PRO further associates, as evidence, the voice corresponding to the construction information included in the daily work report.

[0035] The acquisition unit ACQ of the construction management support device 100 acquires image information (still images, videos, two images captured by two cameras, etc.) captured by the construction site camera CCM installed at the construction site via the communication unit COM and the network NET. The construction management support device 100 is also connected to the terminal PC1, mobile terminal PDA1, and mobile phone terminal MS1 held by the site manager, and is capable of exchanging information. The output unit OUT can output information stored in the device and information generated to the printer PRN. The display unit DIS can also display information stored in the device and information generated. The input unit IN accepts operation instructions and information inputs input via the mouse MUS and keyboard KBD, and passes the accepted information and data to the acquisition unit ACQ.

[0036] Alternatively, a touch panel may be provided on the display unit DIS so that it also functions as an input unit. The input unit and display unit may be provided on, for example, a mobile terminal PDA1 (or other terminal), and input and display may be performed on a terminal connected to the device via a network. This also applies to the following embodiments.

[0037] Each functional unit included in the control unit is preferably realized by a program module loaded into the memory space of the control unit. Usually, software that causes the processor to function as each unit of the device is downloaded from a storage unit or a website, and the software is installed on a PC and started, causing the PC to operate as the device. Note that each functional unit provided in the control unit is merely a set of functionally coherent steps, and it is possible to combine multiple functional units into one functional unit, incorporate part of it into another functional unit, or divide it into another independent functional unit.

[0038] It should be noted that the generated / extracted information, intermediate data, and acquired data can be transmitted to the outside, displayed on the display unit, and stored in the storage unit in the other embodiments described below. It is preferable that the present device is constructed on a computer, such as a general-purpose computer, a specific-purpose computer, a server, a PC, or a mobile terminal such as a smartphone, or that a program module that realizes (executes) the functions and processing procedures (methods) of the present device on a computer is stored in a CPU or storage unit of the computer, or is read from an external server or storage, and this is also true for each of the following embodiments. Also, each functional unit may be distributed to separate computers or devices connected by a network. Also, a form in which multiple functional units are combined into one, or part of the processing steps is assigned to another functional unit, is also possible. This is also true for other embodiments.

[0039] Construction information is At least one of the following is included: work content, machinery, tools, equipment, heavy machinery, vehicles, labor-hours information, in-house terminology, worker information (labor, target work (e.g., heavy machinery operator), and workload); It is characterized by:

[0040] The storage unit 120 further includes a trained AI model LAM, which is read (loaded) into the control unit 110 (processor) and functions as an AI (artificial intelligence). The control unit 110 further includes a question generation unit QUG that uses the trained AI model LAM to generate question information (such as text information such as "Please input xxx" or a voice output command such as "Please talk about xxx") to prompt voice input about the construction work (when there are unclear or insufficient points in the contents of the daily work report, or simply to prompt input).

[0041] It is also characterized in that at least one of the voice recognition unit VR, the processing unit PRO, and the generation unit GEN uses the trained AI model LAM.

[0042] In addition, the acquisition unit ACQ Further acquiring on-site audio information or on-site image information from a mobile monitoring device (e.g., a drone); The voice recognition unit VR or the image recognition unit IMA, Further extracting on-site construction information from the acquired on-site voice information or on-site image information; The generation unit GEN: Instead of generating a daily work report including the extracted construction information based on the construction information, A daily work report including construction information is generated based on the extracted construction information and the extracted construction information from the site.

[0043] In addition, the control unit 110 A command generating unit OGEN that generates command information for prompting a mobile monitoring device DR (such as a drone) to operate to acquire on-site audio information or on-site image information, The acquisition unit ACQ Further acquiring on-site audio information or on-site image information from a mobile monitoring device DR (such as a drone); The voice recognition unit VR or the image recognition unit IMA, Extracting on-site construction information from the acquired on-site voice information or on-site image information, The generation unit GEN: Instead of generating a daily work report including the extracted construction information based on the construction information, A daily work report including construction information is generated based on the extracted construction information and the extracted construction information from the site.

[0044] Fig. 2 is a flow chart showing an example of processing executed by the construction management support device shown in Fig. 1. As shown in the figure, in step S11, the storage unit 120 stores worker identification information WDI (human (moving object other than object or animal), worker's physique, height, weight, bib number, name tag, work clothes, clothing, outfit, helmet, accessories, hat, face, and other features such as shape, color, design, pattern, etc.) for identifying the worker. Next, in step S12, the acquisition unit ACQ acquires image information of the target construction captured by a camera (not shown) connected to this device or an imaging means mounted on a construction site camera CCM connected to an external network.

[0045] In step S13, the image analysis unit IMA refers to the worker identification information, analyzes the acquired image information, and recognizes (identifies) the worker. It is preferable to use a known technique for image recognition technology of people such as workers. It is possible to recognize and identify an individual (worker) by features such as face, height, and body shape. Alternatively, it is possible to recognize and identify a worker by features such as clothing, hat, bib number, pattern, and color. This image recognition is preferably performed in real time, but it can also be performed periodically at a predetermined time such as 1 minute or 5 minutes. In step S14, the measurement unit CNT measures the work time of the recognized worker. This is the time from when the worker appears in the image until the person disappears for a predetermined time or more (20 minutes, 1 hour, etc.) as the work time, or the time when the person appears as the work time. Finally, in step S15, the display unit DIS displays the recognized worker and work information (such as a daily labor report) including the work time associated with the worker. Once the storage step such as step S11 is performed, it is not necessary to perform it again until the information is changed. Alternatively, the storage step becomes unnecessary by installing a ROM or flash memory storing the information in the device or computer. The storage step is similar in other embodiments.

[0046] Fig. 3 is a diagram showing the process executed by the construction management device shown in Fig. 1 and the output result. As shown in the figure, in this example, the color and number of the bib are used as worker identification information. The color and number of the bib and the worker's name are registered in advance as information, and this device uses the color and number of the bib in the image information captured by the camera to identify which worker started and finished the work, and measures the work time. Then, it displays (outputs) work information (daily labor report) including the worker's name and the work time as work information.

[0047] As described above, the first embodiment focuses on determining how much work each worker did at a construction site using image recognition, that is, a technique for automatically creating a daily labor report. Next, a technique for automatically creating a daily labor report using voice will be described.

[0048] Fig. 4 is a flow chart showing an example of processing executed by the construction management support device shown in Fig. 1. As shown in the figure, in step S21, the storage unit 120 stores an execution budget, a voice recognition model (including at least one of an acoustic model, a pattern matching dictionary model, and a language model, and customized to correspond to words related to the construction), and a daily work report generation algorithm for generating a daily work report based on the construction information. Either the execution budget or the daily work report generation algorithm may be stored, but it is more preferable to store both. In step S22, the acquisition unit ACQ acquires the voice. Next, in step S23, the voice recognition unit VR uses a voice recognition model to generate text information based on the acquired voice, and extracts construction information based on the generated text information. Next, in step S24, the generation unit GEN (using a daily work report generation algorithm or an execution budget) generates a daily work report including the construction information based on the extracted construction information. Finally, in step S25, the daily work report is displayed.

[0049] FIG. 23 is a schematic diagram for explaining one embodiment of the present invention. As shown in the figure, the device acquires the voice VCE1 of a site manager or the like during a morning meeting or the like. The voice recognizes the following: "Two workers, Sato and Yamada, entering," and "One backhoe entering, operator: Takahashi." The device inserts (inputs) data from the voice recognition content into the corresponding portion of the execution budget PLN1 (schedule), generating work content SN1 (daily report). In this way, data from the execution budget can be used for necessary figures and unit prices, so a practical daily report can be generated even if there is insufficient data in the voice.

[0050] Fig. 24 is a schematic diagram for explaining one embodiment of the present invention. As shown in the figure, this device associates "Two workers, Sato and Yamada, enter" with the worker details item as evidence of the daily report. This makes it possible to eliminate falsification and inaccuracies in the daily report.

[0051] FIG. 25 is a schematic diagram for explaining one embodiment of the present invention. As shown in the figure, the present device acquires the voice VCE2 of a site manager or the like at an evening meeting or the like. The voice includes the changes for the day and the completed volume (amount of work done), and this allows the work hours, amount of work done, etc. to be corrected, as in the daily work report SN2, in response to the voice "Yamada went to site B in the afternoon. Manual excavation was 75% of the planned amount." Similarly, the work hours, amount of work done, etc. are corrected in response to the voice "Mechanical excavation was achieved twice as much as planned." These voices are also associated with the corrections as evidence.

[0052] FIG. 26 is a schematic diagram for explaining one embodiment of the present invention. As shown in the figure, the drone DR31 (mobile monitoring device) that receives a command from this device takes off from the drone base BS and starts flying to photograph the construction site. The reason for acquiring site images (site audio) is to acquire evidence of cost management and associate it with the corresponding part of the cost management data. In this example, the site agent inputs by voice that there are two workers and one backhoe, and while flying, the drone DR31 flies over the site to acquire supporting evidence. Then, the drone DR31 transmits the acquired site images (site audio) to this device. In this device, Worker: Recognize QR1 => Worker PS1 Worker: Recognize QR2 ==> Worker PS2 Heavy equipment: Recognize QR3 ==> Backhoe BH1 Construction volume: 100m3 recognized from images (or images and GPS (location signal)) Workers and heavy machinery are recognized from site images using identification information such as QR codes (registered trademark) and management tables (such as work content identification tables).

[0053] FIG. 27 is a schematic diagram for explaining one embodiment of the present invention. As shown in the figure, this device associates work content (daily report) with supporting evidence. The voice portion of the audio VCE1 relating to the worker is associated with the item of work content relating to the worker. Similarly, in the on-site image, the image portion relating to the worker is associated with the item of work content relating to the worker. In this way, two pieces of evidence are associated, so this device guarantees that the cost management data is very accurate and proper, and it is also easy for a third party (such as a government office or certification body) to verify it.

[0054] Fig. 28 is a schematic diagram showing the overall configuration of the present invention. This shows the overall configuration of the present invention, which makes it possible to provide a very effective cost management system for public works in the construction and civil engineering industries.

[0055] Next, in the second embodiment, a technique for realizing automatic management of work progress, rather than a daily labor report, will be described, which is mainly focused on determining what kind of work was performed and for how long at the construction site using image recognition. Unless otherwise specified, the functional blocks with the same names as those in FIG. 1 function and produce the same effects in FIG. 5. This is also true in other embodiments.

[0056] <Embodiment 2> 5 is a block diagram showing an outline of a construction management support device according to embodiment 2 of the present invention. As shown in the figure, the construction management support device 200 (CMSD) has a control unit (CPU, arithmetic processing unit, processor) 210, an input unit IN, an output unit OUT, a communication unit COM, a memory unit 220, and a display unit DIS.

[0057] The memory unit 220 stores work content identification information SCI for recognizing work content (characteristics such as the shape, color, design, pattern, etc. of a worker's physique, movement pattern, height, weight, bib number, name tag, work clothes, helmet, clothing, accessories, hat, face, tools, machines, instruments, vehicles, heavy machinery, etc., or combinations thereof, etc.). Work content identification information and worker identification information are similar, but the difference is that work content is associated with a worker's physique, movement pattern, height, weight, bib number, name tag, work clothes, helmet, clothing, accessories, hat, face, tools, machines, instruments, vehicles, heavy machinery, etc.

[0058] The control unit 210 has an acquisition unit ACQ that acquires image information of the target construction work captured by the imaging means, an image analysis unit IMA that refers to the work content identification information to analyze the acquired image information and recognize the work content, and a measurement unit CNT (including a clock function) that measures the work time of the recognized work content (for example, the time from when the worker starts the work to when he finishes it, the time present in the image information, etc.). The display unit DIS displays the recognized work content and work information (such as a work progress chart and daily labor report) including the work time associated with the work content.

[0059] FIG. 6 is a flow chart showing an example of a process executed by the construction management device shown in FIG. 5. As shown in the figure, in step S31, the storage unit 220 stores work content identification information for recognizing the work content. Next, in step S32, the acquisition unit ACQ acquires image information of the target construction captured by a camera (not shown) connected to this device or an imaging means mounted on a construction site camera CCM connected to an external network. In step S33, the image analysis unit IMA refers to the work content identification information, analyzes the acquired image information, and recognizes the work content. In step S34, the measurement unit CNT measures the work time of the recognized work content. This is the time from when something having a feature value indicating the work content appears in the image until when something having the feature value does not appear for a predetermined time or more (20 minutes, 1 hour, etc.) as the work time, or the time when the feature value appears as the work time. Finally, in step S35, the display unit DIS displays the recognized work content and work information including the work time associated with the work content (work progress information, a table showing the daily progress, daily labor report, etc.).

[0060] Fig. 7 is a diagram showing the work content identification information and its output result referenced in the process executed by the construction management device shown in Fig. 5. As shown in the figure, in this example, the bib number is used as the work content identification information SCI. As information, the bib number and the work content are registered in advance, and this device uses the bib number in the image information captured by the camera to identify and identify which work content has been started and finished, and measures the work time. Then, as work information, it displays (outputs) work information (daily labor report) including the work content and the work time.

[0061] In this example, the work content specification information SCI specifies two work contents. The work content specification information SCI10 associates "manual excavation" as the work content of bib numbers 11-19. Also, name information is associated with bib number 11, such as "Yamada Taro," and bib number 12, such as "Kato Shiro." The work content specification information SCI11 associates "manual loading" as the work content of bib numbers 21-29. Also, name information is associated with bib number 21, such as "Sato Taro," and bib number 22, such as "Takahashi Jiro."

[0062] This device analyzes the image information IMI and searches for the presence of the registered bibs 11-19, 21-29. If it recognizes the registered bibs 11, 12, it identifies the work content as "manual digging" and measures the work time for each individual (worker), and displays work information (work progress information) such as bib 11, work content "manual digging", and work time 7 hours. It is also possible to tally up and display the number of workers engaged in the work content, the total work time, etc. This example is applicable when a worker performs one work, but when one worker performs multiple works, it is not possible to identify the work content. The work content identification information that solves this problem is explained in the following figure.

[0063] FIG. 8 is a diagram showing the work content identification information and its output result referenced in the process executed by the construction management device shown in FIG. 5. As shown in the figure, in this example, the bib number and the machine (tool) are used as the work content identification information SCI. The bib number, the machine, and the work content are registered in advance as information, and this device recognizes the machine used (or nearby) by the worker wearing the bib number in the image information captured by the camera, and identifies and identifies which work content was started and finished, and measures the work time. Then, as work information, the device displays (outputs) work information (daily labor report) including the work content (and the worker's name) and the work time.

[0064] In this example, the work content specification information SCI specifies two work contents. The work content specification information SCI20 is associated with "manual digging" as the work content for the bib number 11-19 and the excavation drill TL20 as the machine. Name information is also associated, such as "Yamada Taro" for bib number 11 and "Kato Shiro" for bib number 12. The work content specification information SCI21 is associated with "manual loading (manual transport)" as the work content for the bib number 11-19 and the handcart TL21 as the machine (tool). Name information is also associated in a similar manner. In other words, the configuration makes it possible to identify different work contents even with the same bib number.

[0065] This device analyzes the image information IMI and searches for the presence of the registered bib number 11-19. If it recognizes the registered bib number 11, 12, it further recognizes whether the machine being used or operated is a registered excavation drill or handcart, and thereby identifies whether the work content is "manual excavation" or "manual loading", and measures the work time for each individual (worker), and displays work information (work progress information) such as bib number 11, work content "manual excavation", and work time 7 hours. It is also possible to tally and display the number of workers engaged in the work content, the total work time, etc. In this example, the work content is identified by "machine + bib number", but this is not limiting, and the work content can also be identified by "machine operation pattern" + "bib number", etc. For example, since the movement pattern of the machine differs depending on the operation mode even for the same machine, it is possible to identify the work content by registering this.

[0066] FIG. 9 is a diagram showing the individual identification information referred to in the process executed by the construction management device shown in FIG. 5. As shown in the figure, in this example, a QR code attached to the ceiling of a helmet or heavy equipment is used as the individual identification information PPI. As information, an identification code (such as a two-dimensional barcode) for identifying and specifying an individual, and information on the worker and heavy equipment are registered in advance, and this device recognizes the heavy equipment or machine or tool that is wearing the QR code (registered trademark) and the machine or tool that is used by the worker (or in its vicinity) using these QR codes (registered trademark) in the image information captured by the camera of a mobile monitoring device such as a drone or a fixed monitoring device, and thereby identifies and identifies which work content has been started and finished, and measures the work time. Then, as work information, work information (daily labor report) including the work content (and the names and number of workers) and the work time is displayed (output).

[0067] FIG. 10 is a diagram showing a schematic diagram of outputting work contents (daily work report) from individual identification information referred to in the process executed by the construction management device shown in FIG. 5. As shown in the figure, this device analyzes image information IMI and searches whether the recognized QR code (registered trademark) W1 and QR code (registered trademark) M1 exist (are registered). If the registered QR code (registered trademark) W1 and QR code (registered trademark) M1 are found, it further recognizes whether the machine being used and operated is a registered excavation drill or handcart, and thereby identifies whether the work content is "manual excavation" or "manual loading", and measures the work time by individual (by worker), and can display work information (work progress information) such as bib number 11, work content "manual excavation", and work time 7 hours. In this example, it is possible to recognize that the location where the worker with QR code (registered trademark) W1 is located is a manual excavation area, and recognize that the work content is "manual excavation". In the case of QR code (registered trademark) M1, it is recognized that there is a dump truck nearby and that the work is "digging and loading." The work time here may be measured from the image, or may be determined from the voice input of the site supervisor. The time may also be compared between the two and the accuracy may be displayed. If the time from the image is 7 hours and the time from the voice is also 7 hours, the accuracy may be displayed as 100%.

[0068] <Embodiment 3> 11 is a block diagram showing an outline of a construction management support device according to a third embodiment of the present invention. As shown in the figure, the construction management support device 300 (CMSD) has a control unit (CPU, arithmetic processing unit, processor) 310, an input unit IN, an output unit OUT, a communication unit COM, a memory unit 320, and a display unit DIS.

[0069] The memory unit 320 stores machine identification information MDI (such as characteristics of the machine, such as size, shape, tag, attachment, shape, color, design, pattern, etc.) for recognizing a construction machine which is one of machines, tools, equipment, heavy machinery, or vehicles, and a machine work conversion table MPT which includes work content (such as labor-hours code) associated with the construction machine.

[0070] The control unit 310 has an acquisition unit ACQ that acquires image information of the target construction work captured by the imaging means, an image analysis unit IMA that refers to machine identification information to analyze the acquired image information, recognizes the construction machine, and identifies the work content of the construction machine based on the construction machine by referring to a machine work conversion table, and a measurement unit CNT that measures the work time of the recognized construction machine (for example, the time from when the construction machine starts to finish work, the time present in the image information, etc.).The display unit DIS displays work information (daily construction report, daily labor report, etc.) including the recognized construction machine, the work content of the construction machine, and the work time.

[0071] Fig. 12 is a flow chart showing an example of processing executed by the construction management device shown in Fig. 11. As shown in the figure, in step S41, the storage unit 320 stores machine identification information (such as the size, shape, shape, color, pattern, etc. of the machine, tag, attachment, etc.) for recognizing a construction machine which is one of machines, tools, equipment, heavy machinery, or vehicles, and a machine work conversion table including work content (such as labor code) associated with the construction machine. Next, in step S42, the acquisition unit ACQ acquires image information of the target construction work captured by a camera (not shown) connected to this device or an imaging means mounted on a construction site camera CCM connected to an external network.

[0072] Next, in step S43, the image analysis unit IMA refers to the machine identification information, analyzes the acquired image information, recognizes the construction machine, and identifies (recognizes) the work content of the construction machine by referring to the machine work conversion table based on the construction machine. In step S44, the measurement unit CNT measures the work time of the recognized construction machine (for example, the time from when the construction machine starts working to when it finishes working (from when it starts moving until it stops working for a certain period of time, or until the operator leaves, etc.), the time present in the image information, etc.). For example, this is the time from when something with a feature value indicating the work content appears in the image to when the thing with the feature value does not appear for a predetermined time or more (20 minutes, 1 hour, etc.) as the work time, or the time when it appears as the work time. Finally, in step S45, the display unit DIS displays the recognized construction machine, the work content of the construction machine, and work information (daily construction report, daily labor report, etc.) including the work time.

[0073] FIG. 13 is a diagram showing construction machine identification information and its output result referenced in the process executed by the construction management device shown in FIG. 11. As shown in the figure, in this example, a set of "digging drill KD and shovel SH" (of course, one may be used) and a backhoe BH are used as construction machine identification information MDI. Identification information for identifying these machines and tools is registered in advance as information, and this device recognizes machines using the identification information from objects in the image information captured by the camera. Then, by referring to the machine work conversion table MPT having information relating machines to work contents, the device identifies the work contents from the recognized machine. Then, it identifies and identifies which machine started and ended which work contents, and measures the work time. Finally, it displays (outputs) work information (daily labor report) including the work contents (and the machines and tools used for the work, and the names of the workers) and the work times as work information.

[0074] The construction machine identification information MDI specifies two machines. The construction machine identification information MDI10 stores information for identifying an excavation drill KD10 and a shovel SH. The construction machine identification information MDI11 stores information for identifying a backhoe BH. Machines may be identified by features such as two-dimensional and three-dimensional shapes and sizes, but the simplest technique is to read numbers and letters such as bib number signs and machine model numbers, as in other embodiments.

[0075] The mechanical work conversion table MPT10 stores the work content "manual excavation" (or the unit code ABCXXX900) associated with the machine of MDI10. The mechanical work conversion table MPT11 stores the work content "digging and loading (mechanical excavation)" (or the unit code ABCYYY800) associated with the machine of MD11.

[0076] This device analyzes the image information IMI and searches whether there is a machine with information stored / registered in the construction machine identification information. If it recognizes a machine that is stored / registered, it further refers to the machine work conversion table, and thereby identifies whether the work content is, for example, "manual excavation" or "digging and loading", and measures the work time for each machine, and displays work information (work progress information) such as the machine used "digging drill KD1, shovel SH1", the work content "manual excavation", and the work time 7 hours. Similarly, as other machines, it displays work information (work progress information) such as the machine used "digging drill KD2, shovel SH2", the work content "manual excavation", and the work time 5 hours. Furthermore, it recognizes the machines used "backhoe BH1" and "backhoe BH2" as other machines, and displays work information (work progress information) such as the work content "digging and loading", and the work time 4 hours and 8 hours.

[0077] It is also possible to tally and display the number of workers engaged in the work content, the total working hours, etc. In this example, the work content was identified by "two machines (tools)", but this is not the only option; work content can also be identified by "the motion pattern of the person operating it (operation pattern)" + "machine". For example, even with the same machine, the pattern of how the machine moves differs depending on the operating mode, so by registering such motion patterns, it is possible to identify the work content.

[0078] FIG. 14 is a diagram showing the construction machine identification information and the machine work conversion table referred to in the process executed by the construction management device shown in FIG. 11. As shown in the figure, in this example, three types of "backhoe BH20, 21, 22" are used as the construction machine identification information MDI. Identification information for identifying the backhoe into three types (three kinds) is registered in advance as information, and this device recognizes the type of machine from the object in the image information captured by the camera using the identification information. Then, referring to the machine work conversion table MPT having information on the association between the machine and the work content, the work content corresponding to the recognized type of machine is identified from the type. Then, it is specified and identified which machine started and ended which work content, and the work time is measured.

[0079] In this example, the construction machine identification information MDI specifies three machines. The construction machine identification information MDI20 stores information for identifying the backhoe BH20. In this example, it is a backhoe and has a bucket, so it is recognized as a normal type backhoe BH20. The machine may be identified by features such as two-dimensional and three-dimensional shape and size, but the most convenient technique is to read numbers and letters such as the number plate and machine model number as in other embodiments and recognize it as a backhoe with a bucket. The construction machine identification information MDI21 stores information for identifying the backhoe BH21. In this example, the machine is the backhoe BH21, the identification information is the backhoe, and it is recognized by features such as the presence of the hook FK or the operation of the arm at a speed below a predetermined threshold (recognition of the crane mode). Furthermore, the construction machine identification information MDI22 stores information for identifying the backhoe BH22. In this example, the machine: backhoe BH22 has specific information: backhoe, and is recognized by features such as the presence of the blade DZA, or the forward and backward movement of the vehicle, and the raising and lowering of the arm.

[0080] The machine work conversion table MPT20 stores the work content "digging, loading, cutting" (or unit code ABCYY200) associated with the machine of MDI20. The machine work conversion table MPT21 also stores the work content associated with the machine of MD21: installation work ((rigging, rigging removal), (block installation, transportation, piping installation, etc.), (unit code: ABCYY210). The machine work conversion table MPT22 also stores the work content associated with the machine of MD22: earth removal work ((transportation of earth and sand, leveling work, etc.), (unit code: ABCYY220).

[0081] In this way, this device analyzes image information and searches for whether or not there is a machine with information stored and registered in the construction machine identification information. If it recognizes a stored and registered machine, it further refers to the machine work conversion table, and even if it recognizes a backhoe, it identifies the machine in more detail based on the type and operation of the machine, identifies the work content, measures the work time for each machine, and displays work information (work progress information) such as the machine used, work content, and work time.

[0082] FIG. 15 is a diagram showing a schematic diagram of a technique for specifying the work area designation information that designates the work area that is a part of the image area. As shown in the figure, the work area designation information that designates the work area D-IMI that is a part of the image area can be obtained from the image information IMI10. Alternatively, while looking at the display unit of this device, the operator can designate the desired area with a mouse or the like, and this can be used as the work area designation information that designates the work area D-IMI. The advantage of this is that when there are people, machines, etc. in the image information that are not directly related to the results of the work, it is possible to limit the area to only those involved in the work of the work, making it possible to more accurately recognize machines and workers, and as a result, it is possible to obtain more accurate work progress information. For example, when a construction worker leaves the work area, it is not involved in the work results, and therefore break times and the like are not counted as work time, making it possible to recognize the amount of work more accurately. In addition, there is an advantage that the amount of work of unused machines and the like in the image is not counted as work. In addition, although a lot of computing power is used for image processing, the amount of calculation can be reduced by limiting the area as described above, and there is also an advantage that the present invention can be executed even on equipment with a processor with low computing power. The work area can be specified using a mouse or the like, but it is preferable for the worker to carry a mobile device with GPS functionality, use GPS to locate three or more desired points, and specify the area by dividing the area into sections with these location information as vertices.

[0083] A technique for finding an accurate amount of work has been explained in FIG. 15, but another technique for making the amount of work even more accurate will be explained below. FIG. 16 is a schematic diagram explaining a technique for measuring actual operation time by recognizing the vibration and movement of machines and tools. Many machines and tools used in construction vibrate violently or finely, so by recognizing this vibration using image recognition technology, it is possible to measure whether or not they are actually working based on this vibration. It is recognized whether or not the construction machine is vibrating, moving, or moving (recognizing vibration is most suitable), and the time the construction machine is vibrating, moving, or moving is further measured as actual operation time, and this is output as the actual operation time or actual work time of the machine or worker.

[0084] The acquisition unit shown in FIG. 11 acquires image information of the target construction work captured by the imaging means, work progress information including the worker of the target construction work, the work content and planned output associated with the worker, and the work rate information including the output per unit time for the worker. The image analysis unit also refers to the worker identification information to analyze the acquired image information and recognize the worker. The measurement unit measures the work time of the recognized worker (for example, the time from when the worker starts the work to when he finishes it, the time present in the image information, etc.). The calculation unit (not shown) refers to the work rate information to calculate the output (amount of work) from the measured work time for the worker. The display unit displays the worker of the target construction work work, the work content and planned output associated with the worker, and work information (such as daily output report) including the calculated output. It is possible to calculate the output not only for the worker but also for the construction machine and work content. Next, a detailed explanation will be given using the figures.

[0085] FIG. 17 is a diagram for explaining a mechanism for outputting work information (daily output report) including planned and calculated output using unit information and work progress information. Unless otherwise specified, the same symbols as those explained in FIG. 13 are the same and have the same functions. In this example, a backhoe is recognized, and furthermore, the unit information BKI30: backhoe: pile 0.5 m3 (per hour) of the unit code: ABCYY801 of the machine used related to the unit code: ABCYY800 of excavation and loading (mechanical excavation), which is the work content of the backhoe, is obtained (or stored in advance), and the work progress information YKD30: backhoe, worker A, planned output of excavation and loading 3.0 m3 (3 cubic meters) is obtained. Then, using this unit information, a calculated output of 2.5 m3 (calculated as 5 hours of work time x 0.5 m3 of work volume per unit time) is obtained. The calculated output of 3.0 m3, the planned achievement rate of about 80%, and the like are displayed as work information (daily output report). In this way, the unit-rate codes indicating the work machines and work content are mutually related through the code system of the unit-rate estimation system, so by utilizing this, it is possible to easily link the recognized machines, planned work volume, and calculated output.

[0086] Fig. 18 is a schematic diagram for explaining one embodiment of the present invention. As shown in the figure, a camera (imaging means) placed at the construction site captures the state of the construction, and a user at the site office can operate this device to refer to the daily labor report that is created automatically (or after a specific user operation is performed).

[0087] Fig. 19 is a schematic diagram for explaining one embodiment of the present invention. As shown in the figure, instead of the master tables shown in each of the above embodiments, it is possible to use master tables in the form of a worker master containing information about workers, a machine master containing information about construction machines, and a vehicle master containing information about vehicles.

[0088] Figure 20 is a schematic diagram for explaining one embodiment of the present invention. As shown in the figure, the construction area is realized by touch operations such as mouse or finger touch, drawing, etc. It is also possible to provide the device with an interface that displays candidate construction contents and allows the user to specify them.

[0089] Fig. 21 is a schematic diagram for explaining one embodiment of the present invention. As shown in the figure, several construction machines can be registered in the machine master (machine identification information) and automatically recognized, and their work time, etc. can be automatically tallied.

[0090] The effects of this device will be explained again. First, it enables accurate and timely understanding of cost management. Accurate daily cost input (input by image recognition) makes the following possible: You can see whether construction is progressing according to budget. - You can understand how much gross profit will be from the completed construction. - You can immediately see the current cost of uncompleted work. As mentioned above, the cost situation of the construction can be grasped immediately and action can be taken quickly.

[0091] By having machines (cameras) perform the work that was previously done by humans more accurately, labor costs can be reduced and administrative efficiency can be improved, resulting in increased profits through changes in personnel allocation. Also, by introducing the "cost management support system using image recognition" according to this invention, although costs will naturally rise temporarily due to the introduction of the system, it is possible to provide an extremely effective means of reducing costs and realizing profits that outweigh the increased costs.

[0092] Secondly, it makes it possible to improve the accuracy of in-house labor rates. Traditionally, veteran engineers have calculated highly accurate in-house labor rates based on their experience and intuition, which has led to no significant discrepancies in working budgets and estimates, contributing to the stability of business. In recent years, with the number of veteran engineers declining dramatically, the accuracy of in-house labor rates has become a major management issue. The "cost management support system using image recognition" of the present invention accumulates accurate daily input and is able to provide more objective and rational in-house labor rate data by analyzing the vast amount of actual data stored in a database, rather than relying on experience and intuition.

[0093] FIG. 22 is a diagram showing a schematic diagram of work content identification information referred to in the process executed by the construction management device. As shown in the figure, in this example, a two-dimensional barcode (QR code (registered trademark), etc., or the color, pattern, shape, or combination of these (these are called reflector patterns) of a reflector such as a vest or sticker) is used as the work content identification information SCI. As information, the two-dimensional barcode pattern and the work content are registered in advance, and the device uses these patterns in the image information captured by the camera to specify and identify which work content has been started and finished, and measures the work time. Then, as work information, work information (daily labor report) including the work content and the work time is displayed (output).

[0094] In this example, the work content identification information SCI specifies four work contents. The work content identification information SCI30 is associated with "manual excavation" as the work content of the pattern worker. The work content identification information SCI31 is associated with "manual loading" as the work content of the pattern worker. This identification information does not have information that identifies individual workers. Even with this configuration, it is possible to count the number of workers performing the work and the total work time.

[0095] Next, an example will be given that has information that identifies individual workers. The work content identification information SCI32 is associated with "Work ABC" as the work content of the pattern worker. The name information "Yamada Taro" is also associated. By associating even an individual's name with the reflector pattern, it is possible to manage labor at an individual level. The work content identification information SCI33 is associated with the work content of the machine (backhoe) of the pattern, "Excavation and loading (mechanical excavation)" (or the labor code ABCYYY800). In this way, the identification information is composed of information indicating the characteristics of the bib number, reflector, person, and machine (usually an image pattern extracted by performing image processing and its features, etc.) and the work content associated with these.

[0096] The advantage of 2D barcodes is that they are highly visible, most operating systems have APIs for reading 2D barcodes, and they are easy to obtain and recognize. Another advantage of reflectors is that while the feature values ​​of images (machines, people, objects, etc. that are the subject of surveillance when analyzing images) fluctuate greatly depending on the weather outdoors, the color, design, shape, and patterns of reflectors are relatively small in feature values, making it possible to recognize images with high accuracy. Also, reflectors are required to be worn by workers and machines at construction sites from the perspective of safety management, so they can be worn without feeling strange.

[0097] Although the present invention has been described based on the drawings and embodiments, it should be noted that those skilled in the art can easily make various modifications and corrections based on the present disclosure. Therefore, it should be noted that these modifications and corrections are included in the scope of the present invention. For example, the processes and functions included in each part and each step can be rearranged so as not to cause logical inconsistencies, and multiple means / parts and steps can be combined into one or divided. Alternatively, it should be noted that some components, functions, processes, steps, etc. of the device, method, program, etc. according to the present invention can be arranged on a remote server, etc. [Explanation of symbols]

[0098] 100...Construction management support device 110...Control unit 120...Storage section 200...Construction management support device 210...Control unit 220...Storage section 300...Construction management support device 310...Control unit 320...Storage section ABCXXX900...Step code ABCYY200...Step code ABCYYY800...Step code ACQ……Acquisition Department B……Yamada PM BH...backhoe BH1...backhoe BH2...backhoe BH20...backhoe BH20, 21, 22... Backhoe BH21...backhoe BH22...backhoe BKI30...Work rate information BS...Drone base CCM...Construction surveillance camera CCM...Construction site camera CNT...Measuring section COM……Communication Department D……Construction area DD……Work daily report DIS……Display section DR……Mobile monitoring device DR31...Drone DZA……earth removal board EV……Execution budget FK...Hook GEN……Generation section GPS...location IMA...Image Analysis Department IMA...Image Recognition Department IMI...Image information IMI10...Image information IN... Input section KBD...keyboard KD...Drill KD1...Drill KD10...Drill bit KD2...Drill LAM...Model MDI……Machine specific information MDI……Construction machinery specific information MDI10...Construction machinery specific information MDI11...Construction machinery specific information MDI20...Construction machinery specific information MDI21...Construction machinery specific information MDI22...Construction machinery specific information MPT... Machine Work Conversion Table MPT10...Machine work conversion table MPT11...Machine work conversion table MPT20...machine work conversion table MPT21...Machine operation conversion table MPT22...Machine operation conversion table MS1: Mobile phone terminal MUS...mouse NET...Network OGEN……Command generation unit OUT...Output section PC1...Terminal PDA1...Portable terminal PDA1, ... portable terminal PLN1... Execution budget PPI...Personally Identifiable Information PRN...Printer PRO……Processing section PS1……worker PS2...Worker QUG……Question generation unit SCI……Work content specific information SCI10...Work content specific information SCI11……Work content specific information SCI20...Work content specific information SCI21……Work content specific information SCI30...Work content specific information SCI31...Work content specific information SCI32...Work content specific information SCI33...Work content specific information SH...shovel SH1...shovel SH2...shovel SN1……Work details SN2……Work daily report TL20...Drill TL21...Car VCE1... Audio VCE2... Audio VOR...Voice recognition model VR...Voice recognition section W1……Code WDI... Worker Identification Information YKD30……Work progress information

Claims

1. A construction management support device, A storage unit for storing a speech recognition model; An acquisition unit that acquires voice; A voice recognition unit that uses the voice recognition model to generate text information based on the acquired voice and extracts construction information based on the generated text information; A generation unit that generates a daily work report including the extracted construction information based on the construction information, The daily work report including the construction information is a daily work report including the work time of each worker as the construction information. A construction management support device comprising:

2. The construction management support device according to claim 1, The acquisition unit, Get more images, The construction management support device, An image recognition unit that generates image-derived construction information based on the acquired image; A processing unit that compares the construction information included in the daily work report with the image-derived construction information to obtain matching image-derived construction information, and further associates the construction information included in the daily work report with an image corresponding to the image-derived construction information that matches the construction information as evidence. A construction management support device comprising:

3. The construction management support device according to claim 1, Further comprising a processing unit that associates the voice corresponding to the construction information included in the daily work report as evidence. A construction management support device comprising:

4. In the construction management support device according to any one of claims 1 to 3, The construction information is At least one of the following information is included: work content, machinery, tools, equipment, heavy machinery, vehicles, labor-hours information, company terminology, worker information, and work volume; A construction management support device comprising:

5. In the construction management support device according to any one of claims 1 to 4, A trained AI model, A question generation unit that generates question information that prompts voice input regarding construction work using a trained AI model; Further comprising A construction management support device comprising:

6. 3. The construction management support device according to claim 2, Further having a trained AI model, At least one of the speech recognition unit, the processing unit, and the generation unit uses the trained AI model. A construction management support device comprising:

7. 3. The construction management support device according to claim 2, The acquisition unit, Further acquiring on-site audio information or on-site image information from the mobile monitoring device; The voice recognition unit or the image recognition unit, Further extracting on-site construction information from the acquired on-site voice information or on-site image information; The generation unit, Instead of generating a daily work report including the extracted construction information based on the construction information, Generate a daily work report including the construction information based on the extracted construction information and the extracted on-site construction information; A construction management support device comprising:

8. 3. The construction management support device according to claim 2, a command generating unit that generates command information for prompting the mobile monitoring device to operate to acquire on-site audio information or on-site image information; The acquisition unit, Further acquiring on-site audio information or on-site image information from the mobile monitoring device; The voice recognition unit or the image recognition unit, Extracting on-site construction information from the acquired on-site voice information or on-site image information, The generation unit, Instead of generating a daily work report including the extracted construction information based on the construction information, Generate a daily work report including the construction information based on the extracted construction information and the extracted on-site construction information; A construction management support device comprising:

9. The construction management support device according to claim 1, The system further includes a processing unit that associates, as evidence, voice, on-site voice information, or on-site image information corresponding to the construction information included in the daily work report. A construction management support device comprising:

10. A construction management support device, A storage unit that stores machine identification information for recognizing a construction machine which is one of machines, tools, equipment, heavy machinery, and vehicles, an operation pattern of the construction machine, and a machine work conversion table including work content associated with the operation pattern of the construction machine; An acquisition unit that acquires image information of the target construction work captured by the imaging means; an image analysis unit that refers to the machine identification information, analyzes the acquired image information, recognizes the construction machine, and further recognizes the operation pattern of the construction machine, and identifies the work content of the construction machine based on the operation pattern by referring to the machine work conversion table; A generation unit that includes the recognized construction machine and the work content of the construction machine in the daily work report. A construction management support device comprising:

11. A construction management support device, A storage unit for storing worker identification information for identifying a worker; an acquisition unit that further acquires on-site image information from the mobile monitoring device; An extraction unit that recognizes a worker from the acquired site image information by referring to the worker identification information and extracts site-derived construction information including the worker; A generation unit that generates a daily work report including the construction information based on the extracted construction information and the extracted on-site construction information. A construction management support device comprising:

12. In the construction management support device according to any one of claims 1 to 11, A measurement unit that measures the work time of the recognized worker; a calculation unit that refers to the work-per-unit information and calculates the output from the work time measured for each worker; The construction management support device further comprises:

13. A construction management support program that causes a processing device to function as the construction management support device according to any one of claims 1 to 12.

14. A computer-readable recording medium, comprising: A computer-readable storage medium storing the construction management support program according to claim 13.

Images