Site management system and site management method

Abstract

To provide a site management system capable of performing risk management of a construction site.SOLUTION: The site management system 1 includes a region classification unit 11 that classifies a region for each element from an image in which a construction site is imaged together with a scale and at least one element of a person, a machine or equipment, and a material is shown, an analysis unit 12 that extracts a point at which the element is located in the image based on a correspondence relationship between a point set on the scale and the region, and analyzes a risk in the construction site based on the extracted point, and an evaluation unit 13 that outputs an evaluation result obtained by evaluating an analysis result.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a site management system and a site management method. [Background technology]

[0002] The construction industry is currently facing a continuous environment in which it must deal with issues such as labor shortages, changes in lifestyles, increasing disaster severity, and measures to address aging infrastructure. At the same time, the construction industry's responsibilities and social demands remain high, making it necessary to take measures to maintain production capacity. To address this issue, the introduction of DX (Digital Transformation) utilizing digital technologies such as robots and AI (Artificial Intelligence) is being promoted.

[0003] For example, 3D measurement, 3D models, and VR / AR technology have been introduced in the design and planning department. Machine control and compaction management systems have been introduced in the construction department. Displacement measurement systems and material property monitoring systems have been introduced in the construction management department. The use of such technologies and systems to manage construction sites is being promoted, and some of the technologies have been institutionalized.

[0004] Patent Document 1 discloses a technology for supporting decision-making by workers at production sites. This technology provides workers with construction record data of construction work performed by skilled workers who have performed construction work similar to the construction work in which the worker is involved at the production site, and who have similar characteristics to the worker in question.

[0005] Patent Document 2 discloses a technology for determining the likelihood of a risk occurring on a construction work floor at each construction site based on the state of tidiness of the construction work floor as seen in a floor image. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] Patent No. 7382434 [Patent Document 2] Japanese Patent Application Publication No. 2024-015539 Summary of the Invention [Problem to be solved by the invention]

[0007] Meanwhile, an overview of the current situation at construction sites working to promote digital transformation has revealed that the use of new technologies and systems has resulted in the occurrence of defects caused by human error. For example, overlooking discrepancies between design values ​​and actual values ​​(machine control), reduced accuracy of measurement values ​​(TS surveying), and overlooking machine malfunctions (weighing systems) have been observed at construction sites. The causes of these defects are all due to basic management behaviors such as insufficient human checks, oversights, and failure to notice defects. Furthermore, there are concerns that these errors are not only caused by new technologies, but also by a decline in the skills of workers themselves.

[0008] Small errors like those mentioned above were causing workers at construction sites to lose their sense of security. Furthermore, these errors were causing a stagnation in the use of new technologies and robotics technology and hindering the transition from legacy systems. From a robotics perspective, it is now considered essential to promote digital transformation in the future by implementing "mechanisms that connect people and new technology" that incorporate psychological security.

[0009] The technology disclosed in Patent Document 1 merely provides workers with construction record data of work performed by skilled workers, and is not applicable to risk management at construction sites that change under various circumstances. Furthermore, the technology disclosed in Patent Document 2 uses the tidiness of the construction work floor to determine the possibility of risk occurrence, but construction sites are not necessarily tidier when robots and other devices are actually operating at the construction site. Therefore, even with the technology disclosed in Patent Document 2, it was not possible to properly manage risk at a variety of construction sites.

[0010] The present invention has been made in view of the above circumstances, and aims to enable risk management at a variety of construction sites. [Means for solving the problem]

[0011] The site management system of the present invention comprises an area classification unit that photographs a construction site together with a scale and classifies the image, which shows at least one element of people, machinery or equipment, and materials, into areas for each element; an analysis unit that extracts points where elements are located in the image based on the correspondence between points set on the scale and the areas, and analyzes the risks of the construction site based on the extracted points; and an evaluation unit that outputs an evaluation result that evaluates the analysis result. [Effects of the Invention]

[0012] According to the present invention, information that evaluates the results of analysis of risks at a construction site can be obtained, making it possible to manage risks at a variety of construction sites. Problems, configurations, and effects other than those described above will become apparent from the following description of the embodiments. [Brief explanation of the drawings]

[0013] [Figure 1] 1 is a block diagram showing an example of the overall configuration of a site management system according to an embodiment of the present invention; [Figure 2] FIG. 2 is a block diagram illustrating an example of the hardware configuration of a computer according to an embodiment of the present invention. [Figure 3] 1 is a flowchart illustrating an example of a risk management method performed in a site management system according to an embodiment of the present invention. [Figure 4] FIG. 1 is a top view showing a worker capturing an image of a construction site according to an embodiment of the present invention. [Figure 5] 1A and 1B are diagrams illustrating examples of spiral frames and Fibonacci spirals according to an embodiment of the present invention. [Figure 6] FIG. 1 is a diagram showing an example of an image on which region classification is performed according to an embodiment of the present invention. [Figure 7]FIG. 10 illustrates the increment of the radius of a Fibonacci spiral according to one embodiment of the present invention. [Figure 8] FIG. 1 is a diagram depicting the radii of a Fibonacci spiral according to one embodiment of the present invention. [Figure 9] FIG. 1 illustrates a method for adjusting image ratio according to an embodiment of the present invention. [Figure 10] FIG. 10 is a diagram showing an example of an analysis format according to an embodiment of the present invention. [Figure 11] FIG. 10 is a diagram showing an example of a risk prediction value calculation table according to one embodiment of the present invention. [Figure 12] FIG. 4 is a diagram illustrating an example of an evaluation result display unit according to one embodiment of the present invention. [Figure 13] FIG. 10 is a diagram showing a display example of a scenario display section according to one embodiment of the present invention. [Figure 14] FIG. 10 is a diagram showing an example of an image on which a Fibonacci spiral is superimposed according to an embodiment of the present invention. [Figure 15] FIG. 10 is a diagram showing an example in which a captured image has a defect according to an embodiment of the present invention. [Figure 16] FIG. 10 is a diagram showing an example in which a region classification unit according to the first modification of the present invention automatically determines the type of element and performs region classification. [Figure 17] FIG. 10 is a diagram showing a modification of the positions of markers set on a Fibonacci spiral according to the second modification of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0014] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. In this specification and drawings, components having substantially the same functions or configurations are designated by the same reference numerals, and redundant description will be omitted.

[0015] [Current situation at the construction site] First, we will explain the current state of risk management at construction sites. Various construction robots are used at construction sites. The Ministry of Land, Infrastructure, Transport and Tourism defines a construction robot as "a technology that adds some kind of new mechanism or control or information processing function to machinery and equipment used at construction and inspection sites, thereby achieving work support, automation, and remote control, and enabling performance improvements and problem-solving in areas such as efficiency, precision, and safety." Possible construction robots include robots that carry objects and robots that drive piles. In the following explanation, construction robots will be referred to as "robots."

[0016] Previously, it was not possible to logically extract the causes of errors in collaborative environments with robots (construction sites). Here, an error refers to the situation leading up to an accident at a construction site. By sharing the causes of errors among workers, it is important to promote their decision-making and lead to error prevention measures in advance.

[0017] Furthermore, what is necessary for collaboration between humans and robots is proper etiquette from humans to things (proper operation). It is also desirable for people to be able to learn from things (realizations from interaction). Therefore, a systematic review of the matching of people and technology is required, and from the perspective of psychological security, it is important to incorporate methods of matching people and technology into construction sites.

[0018] Safety at construction sites is assessed from the perspective of "how much risk there is of an accident." However, accidents at work sites often occur in unpredictable places. For this reason, simply having a system predict danger at construction sites will not solve the problem. Therefore, the site management system described below provides awareness based on danger predictions, ultimately allowing people to make optimal decisions.

[0019] [One embodiment] 1 is a block diagram showing an example of the overall configuration of a site management system 100 according to an embodiment. The site management system 100 includes an information processing terminal 1 and a nudge server 10 that are capable of communicating with each other via the Internet.

[0020] In the following explanation, it is assumed that the site management system 100 is used by workers working at a construction site such as a building site. Workers include, for example, construction site supervisors (unskilled workers), workers, operators, managers (experts), as well as the management department of the organization to which the builder belongs. Note that the terms unskilled and skilled may be used interchangeably instead of workers.

[0021] Nudge is defined as, for example, "an approach based on behavioral science findings that encourages people to take desirable actions." Nudges are used as triggers to make workers aware of errors at construction sites. Construction sites where the site management system 100 is used include not only collaborative environments with robots, but also collaborative environments with devices other than robots (cranes, forklifts, etc.). Robots are assumed to be, for example, heavy machinery (whether automatic or manual) used at construction sites.

[0022] Workers use a camera to capture images of the construction site as a record of inspection work. Conventionally, it has been considered advisable to capture images of the construction site using a field of view that includes criteria for judgment. However, in the past, images were captured based merely on the standard of "visually balanced," leading to a discrepancy between the skill level of the photographer and the person evaluating the captured images. Therefore, in this embodiment, by establishing clear rules for the capture of images, it becomes possible for workers to consciously capture images and quantitatively evaluate them. The rules refer to the check method in the inspection work technique for extracting errors (ensuring a safe construction environment) and the method for setting control points.

[0023] In production management, there is a rule known as the 4Ms. 4M stands for Man, Machine, Material, and Method. Experts ensure productivity by simulating problem discovery and problem-solving, while considering the correlation between the 4Ms and considering trade-offs and synergies.

[0024] At a construction site, the source of an error can be determined by the "condition" of the combination of each of the 4M elements. However, the method does not appear in the image. Therefore, workers use images of the construction site to determine whether there is a possibility of a deficiency in the three elements of Man, Machine, and Material.

[0025] <Configuration example of information processing terminal> The information processing terminal 1 is, for example, a notebook PC (Personal Computer) or a tablet terminal carried by an employee. The information processing terminal 1 includes an imaging unit 2 and an information display unit 3.

[0026] The imaging unit 2 is, for example, a camera. The imaging unit 2 captures an image of an object at a construction site 24 included within the angle of view together with a Fibonacci spiral 22a depicted in a spiral frame 22 (see FIG. 4 described later), which will be described later, and outputs the captured image to the nudge server 10. This image includes the construction site 24 (see FIG. 4 described later) together with the Fibonacci spiral 22a. The image captured by the imaging unit 2 is expected to be a still image, but may also be a video.

[0027] The information display unit 3 displays information such as the evaluation result output from the evaluation unit 13 of the nudge server 10 (such as the evaluation result display unit 70 shown in FIG. 12, which will be described later). Therefore, based on the information displayed on the information display unit 3, the worker can perform operations to resolve errors at the construction site.

[0028] The information processing terminal 1 may be used offline. Images of the construction site captured by the information processing terminal 1 in an offline state may be sent to the nudge server 10 when the information processing terminal 1 is connected to the Internet.

[0029] <Example of nudge server configuration> The nudge server 10 is constructed as an on-premise server or a cloud server. The nudge server 10 is connected to external organizations, namely, a construction company and a site management office installed at the construction site. The construction company and site management office staff can check the images and evaluation results output from the evaluation unit 13 of the nudge server and share the information with workers at the construction site.

[0030] The nudge server 10 includes an area classification unit 11, an analysis unit 12, an evaluation unit 13, and a recording unit 14. The region classification unit 11 classifies the image input from the imaging unit 2 into regions for each element. The region classification unit 11 classifies an image that shows any of the elements Man (person), Machine (machine or equipment), and Material (material) into regions for each element, and treats parts of the image that do not contain any element as unclassifiable. Depending on the image, the region classification unit 11 may only be able to classify into two or one type of region.

[0031] The analysis unit 12 extracts control points where elements are located in the image based on the correspondence between the control points (an example of a point) set in the Fibonacci spiral 22a and the area, and analyzes the risk of the construction site 24 based on the extracted control points. For example, the analysis unit 12 extracts elements necessary for analyzing the image as element information, severity (marker points), and risk assessment values. The control points will be described later with reference to FIG. 5.

[0032] The analysis unit 12 extracts the elements Man (people), Machine (machine or equipment), and Material (materials) from the 4Ms as element information. When the analysis unit 12 extracts elements from an image, it uses control points. For this reason, the analysis unit 12 extracts the elements (people, machines, materials) of marker points with control points M50 (R15.1), M78 (R36.6), and M100 (R51.5) as severity levels, and assigns a weighting coefficient to each element. The coefficients are set as follows, in order of risk in the event of a disaster: Man (0.78), Machine (0.5), and Material (0.23).

[0033] Furthermore, the analysis unit 12 extracts elements that appear at the positions of the markers on the Fibonacci spiral 22a and calculates the occupancy rate of each element in the image. The analysis unit 12 then calculates a risk assessment value by multiplying the sum of the occupancy rates multiplied by the weights by the highest severity among the severity levels calculated for each marker on which the element appears. Details of this process will be described later with reference to FIG. 10. Specifically, the analysis unit 12 calculates the risk assessment value by multiplying the vulnerability value (occupancy rate × severity) by the product of the two highest threat (marker) severity levels. The occupancy rate represents the area occupied by each element (person, material, or equipment) in the image. The analysis unit 12 extracts a risk prediction model by regression analysis using values ​​calculated using the threshold value of the risk assessment value as training data. For example, the analysis unit 12 performs regression analysis based on elements extracted from the image to extract a risk prediction model composed of elements with high influence.

[0034] The analysis unit 12 calculates a risk prediction value by substituting the elements reflected at the positions of the markers attached to the Fibonacci spiral 22a into the risk prediction model. The influence is a value calculated by multiplying the coefficient by the range.

[0035] The evaluation unit 13 outputs an evaluation result obtained by evaluating the analysis result of the analysis unit 12. Here, the evaluation unit 13 compares the risk prediction value with a threshold value (for example, "5"). If the risk prediction value is equal to or greater than the threshold value, the evaluation unit 13 outputs information indicating that the risk of the construction site 24 is low, and if the risk prediction value is less than the threshold value, the evaluation unit 13 outputs information indicating that the risk of the construction site 24 is high.

[0036] The evaluation unit 13 displays the evaluation result of the image, categorized as "safe" or "cautious," on the screen of the information processing terminal 1 based on the risk prediction model extracted by the analysis unit 12, as shown in FIG. 12 (described later), to inform the user of the evaluation result. Furthermore, the evaluation unit 13 clearly indicates the control point and element (M100, machine, etc.) that has the greatest impact on the error, based on the degree of impact obtained from the analysis result. Therefore, the site management system 100 can encourage workers to become aware of the causes of errors.

[0037] 13, which will be described later, based on the relationship between the element appearing at the start position of the Fibonacci spiral 22a and other elements that have a high degree of influence, the evaluation unit 13 outputs information indicating a scenario for having the worker at the construction site 24 check the construction site 24. This allows the worker to check the construction site 24 according to the scenario and eliminate error factors.

[0038] The recording unit 14 records the evaluation results output by the evaluation unit 13. In addition, the images sent from the imaging unit 2 are also recorded in the recording unit 14. The worker can check the evaluation results, etc. recorded in the recording unit 14 via the evaluation unit 13 as needed.

[0039] <Example of computer hardware configuration> Next, the hardware configuration of the computer 50 of each device that constitutes the site management system 100 will be described.

[0040] 2 is a block diagram showing an example of the hardware configuration of the calculator 50. The calculator 50 is an example of hardware used as a computer capable of operating as the site management system 100 according to this embodiment. In the site management system 100 according to this embodiment, each functional block is configured by the calculator 50 (computer) executing a program, and the risk management method shown in FIG. 3 is realized by the cooperation of each functional block.

[0041] The computer 50 includes a CPU (Central Processing Unit) 51, a ROM (Read Only Memory) 52, and a RAM (Random Access Memory) 53, each connected to a bus 54. The computer 50 further includes a display device 55, an input device 56, a non-volatile storage 57, and a network interface 58.

[0042] The CPU 51 reads out program code of software that realizes each function according to this embodiment from the ROM 52, loads it into the RAM 53, and executes it. Variables, parameters, etc. generated during the calculation processing of the CPU 51 are temporarily written to the RAM 53, and these variables, parameters, etc. are read out by the CPU 51 as appropriate. However, an MPU (Micro Processing Unit) or a GPU (Graphics Processing Unit) may be used instead of the CPU 51, or the CPU 51 and a GPU (Graphics Processing Unit) may be used together. The functions of the area classification unit 11, analysis unit 12, and evaluation unit 13 of the nudge server 10 are realized by the CPU 51.

[0043] The display device 55 is, for example, a liquid crystal display monitor, and displays the results of processing performed by the computer 50 to the worker. The input device 56 is, for example, a keyboard, a mouse, etc., and allows the worker to input predetermined operations and give instructions. The information display function of the information display unit 3 of the information processing terminal 1 is realized by the display device 55. Although not shown, the information processing terminal 1 is provided with an input unit in the form of a touch panel display, and the function of this input unit is realized by the input device 56.

[0044] The nonvolatile storage 57 may be, for example, a hard disk drive (HDD), a solid state drive (SSD), a flexible disk, an optical disk, a magneto-optical disk, a CD-ROM, a CD-R, a magnetic tape, or a nonvolatile memory. The nonvolatile storage 57 stores an operating system (OS), various parameters, and programs for operating the computer 50. The ROM 52 and the nonvolatile storage 57 store programs and data necessary for the CPU 51 to operate. In other words, the ROM 52 and the nonvolatile storage 57 are used as examples of computer-readable, non-transitory storage media that store programs executed by the computer 50. The function of the recording unit 14 of the Nudge server 10 is realized by the nonvolatile storage 57.

[0045] For example, a network interface card (NIC) or the like is used as the network interface 58. The network interface 58 is capable of transmitting and receiving various types of data between devices via a local area network (LAN), a dedicated line, or the like connected to a terminal of the NIC.

[0046] <Example of processing in the on-site management system> Next, a specific example of processing by each functional unit of the site management system 100 will be described with reference to the flowchart of FIG. 3 and other drawings.

[0047] 3 is a flowchart showing an example of a risk management method performed by the site management system 100. This risk management method is performed by the information processing terminal 1 and the nudge server 10 in cooperation with each other.

[0048] First, the worker uses a dedicated scale to acquire data necessary for image analysis by the analysis unit 12. For example, the imaging unit 2 acquires image data of an image of the construction site 24 captured together with the spiral frame 22 shown in Fig. 4 (S1). Here, the spiral frame 22 and the captured image will be described with reference to Figs. 4 and 5.

[0049] 4 is a top view showing a worker capturing an image of a construction site. The image shows a worker 21 capturing an image of a construction site 24 within a field of view 23 while holding an information processing terminal 1 and a spiral frame 22. The field of view 23 is determined by the specifications of the imaging unit 2 provided in the information processing terminal 1.

[0050] For example, in photographs taken to record a construction site24, it is common for the composition to be biased to one side rather than symmetrical. The reasons for this include the fact that a wide area is often recorded starting from a close-up, and that a construction blackboard displaying recording notes is placed in the foreground. Therefore, a spiral frame based on the Fibonacci spiral, which has been proposed as having "harmonious beauty," is used as the scale for the angle of view23 when taking photographs. The beauty of the Fibonacci spiral composition is thought to be functional even in the construction state, but this is not a matter of intuition and has not been logically proven. However, good composition is thought to lead to ease of evaluation.

[0051] The spiral frame 22 is a transparent panel on which a Fibonacci spiral is drawn. A worker 21 can obtain an image of the construction site 24 with the Fibonacci spiral reflected in it by capturing an image of the construction site 24 through the spiral frame 22. The construction site 24 has at least one of the above-mentioned 4Ms (Man, Machine, Material, Method).

[0052] FIG. 5 is a diagram showing an example of a spiral frame 22 and a Fibonacci spiral 22a. A Fibonacci spiral 22a is drawn on the spiral frame 22 as an example of a scale to be imaged together with the construction site 24. The Fibonacci spiral 22a is created with reference to the composition of the Fibonacci spiral (golden ratio 1:1.618), which is said to be effective in high-speed processing of visual information. However, the standard angle of view (1:1.333) of the imaging unit 2 used at the construction site 24 differs from the golden ratio of 1:1.618. For this reason, the Fibonacci spiral 22a is created by applying the Fibonacci spiral to the standard angle of view of the imaging unit 2 used at the construction site 24 using a unique method.

[0053] Based on previous on-site verification, it has been determined that it is best to set production management points at 23%, 50%, and 78% of the process. The three points set in the process are based on empirical knowledge, established based on past construction records of changes in construction speed and the percentage of confirmation inspections during construction, as well as data from flow line analysis tests, and on the fact that change points are concentrated at these points. For this reason, the percentages (23%, 50%, and 78%) of the spiral combining each radius of the Fibonacci spiral 22a are set through statistical analysis.

[0054] The radius of the Fibonacci spiral 22a is associated with 23%, 50%, and 78% of the process. The shape of the Fibonacci spiral 22a, which unfolds from the center outward, is assumed to represent the order in which checkpoints are configured. For example, in the Fibonacci spiral 22a, markers M50, M78, and M100 are set in order, starting with M23 at the center of the spiral.

[0055] When a worker 21 photographs a construction site 24, the center point is aimed at a "procedure" that could be a cause of an error. From this center point, any of the elements of "people, materials, or equipment" are checked and extracted in order along the curve of the Fibonacci spiral 22a. This flow is the same as the flow of process management (inspection) on a time axis, so the process ratio can also be applied as a checkpoint for the Fibonacci spiral 22a.

[0056] The scale used in this embodiment is a modified Fibonacci spiral 22a, with markers set at the centers of gravity of the quarter circles that make up the Fibonacci spiral 22a. The control points are determined by selecting the centers of gravity of the eight quarter circles (R0.6, R3.2, R9.7, R15.1, R21.7, R29.5, R36.6, and R51.5) that make up the Fibonacci spiral 22a, and determining which combination provides the most effective risk prediction model for regression analysis. Figure 5 shows an example in which the centers of gravity of the quarter circles with radii R15.1, R36.6, and R51.5 are extracted as control points.

[0057] The spiral frame 22 shown in the upper part of FIG. 5 includes the Fibonacci spiral 22a described above, as well as a center point within the spiral (marked "center" in the figure) and three marker points (downward-pointing triangular points marked "markers" in the figure). The marker points are located at the centers of gravity of the quarter circles that make up the Fibonacci spiral 22a. Each quarter circle is marked with an R, which indicates its radius. For example, R3.2 indicates a circle with a radius of 3.2 mm. A downward-pointing triangular mark is marked at each of the centers of gravity of the quarter circles. In addition, the Fibonacci spiral 22a has three marker points, M50, M78, and M100, set as control points.

[0058] The bottom of Figure 5 shows only the Fibonacci spiral 22a and the downward-pointing triangular marks. Here, the downward-pointing triangular marks are numbered 31 to 37 in order from the center outward to identify them. The numbers assigned to the triangular marks will be used to explain modified examples of the three marker points, which will be described later.

[0059] The worker 21 photographs the positions of at least two of the three elements, Man (person), Machine (machine or equipment), and Material (materials), aligning them with three marker points marked on the spiral frame 22. In the risk management method according to this embodiment, information on the element (person, material, or equipment) at the center point of the Fibonacci spiral 22a and information on the elements captured at the marker positions are input into a risk prediction value calculation table (see FIG. 11, described later) that is used as a risk prediction model, and a risk prediction value is output. The worker is provided with the appropriateness of the position of the Fibonacci spiral 22a and an evaluation of the construction site 24, which are calculated based on this risk prediction value.

[0060] The worker 21 simply overlaps the spiral frame 22 to photograph the construction site 24, and the rules for setting the angle of view are clear. The rules for setting the angle of view can be shared not only with the worker 21 who photographs the construction site 24, but also with the skilled person who manages the construction, which also has a synergistic effect of improving risk communication.

[0061] Returning to the explanation of FIG. After step S1, the region classification unit 11 analyzes the image, classifies the regions, and calculates the occupancy rate of each region in the image (S2). Now, the region classification will be described with reference to FIG. FIG. 6 is a diagram showing an example of an image on which region classification is performed.

[0062] Image (1) shown on the left side of Figure 6 is an example of an image of a construction site 24 with a Fibonacci spiral 22a superimposed on it. Image (1) was taken at an angle to the horizontal, but the image position was not corrected to make it horizontal during area classification.

[0063] Image (2) shown on the right side of Figure 6 is a diagram showing an example of the results of region classification of image (1). Image (1) is divided into 15 rows and 20 columns, and region classification is performed for each section. Each section is color-coded according to the region that represents the element, but in Figure 6, they are represented by different types of hatching. The numbers assigned to each section also represent the element. "0" is an unclassifiable region. "1" is the Man region. "2" is the Machine region (machinery or equipment). "3" is the Material region. Note that image (1) may be further divided into smaller sections, and region classification may be performed for each section.

[0064] Here, a method for setting the Fibonacci spiral 22a will be described. Fibonacci spiral 22a is a figure that applies the Fibonacci sequence. The Fibonacci sequence is a sequence of numbers that starts with 1 and the number of the next number is the sum of the two previous numbers. The Fibonacci sequence continues as 1, 1, 2, 3, 5, 8, 13, 21...

[0065] 7 is a diagram showing the increments of the radius of the Fibonacci spiral 22a, in which the horizontal axis represents the number of increments and the vertical axis represents the value of the radius.

[0066] 7, a graph of the radius increment of the Fibonacci spiral 22a is expressed by an S-shaped curve of the following formula (1): Therefore, the Fibonacci spiral 22a is set by converging to an S-shaped curve.

[0067]

number

[0068] FIG. 8 is a diagram depicting the radii of the Fibonacci spiral. When the Fibonacci spiral is drawn without adjusting its radii, arcs of radius 1, radius 2, radius 3, radius 5, radius 8, and radius 13 are formed from the center. If the Fibonacci spiral is fitted to a screen size with an aspect ratio of 1:1.414, there will be parts that are not included in the screen (shown shaded in the illustration). These parts that are not included in the screen must be adjusted.

[0069] FIG. 9 is a diagram showing a method for adjusting the image ratio. By performing a regression analysis on the S-curve equation (1), the S-curve is converged to a model based on the Fibonacci spiral 22a shown in Fig. 7. Also, the parameter a in equation (1) is set to "246804" and the parameter b is set to "0.00222".

[0070] The No. items in Figure 9 correspond to the item numbers in Figure 7. The progress items in Figure 9 correspond to the No. items in Figure 9, but represent values ​​using formula (1). As progress conditions, 2.3, 5.0, and 7.8, which correspond to 23%, 50%, and 78% of the above-mentioned process, are set, and these values ​​are underlined for emphasis. The cumulative items in Figure 9 are initially set to the cumulative values ​​passing through 2.3, 5.0, and 7.8 of the Fibonacci spiral 22a, using the radius of the Fibonacci spiral 22a as reference. The calculated value is calculated by converging using regression analysis with the initially set values.

[0071] The calculated value item stores the calculated value obtained by substituting the value of the No. item for x in formula (1). The squared difference item stores the squared value of the difference between the cumulative total item and the calculated value item. Since differences include both positive and negative values, the square of the difference is found. The sum item stores the sum of the squared differences calculated for No. items 1 to 9.

[0072] The image ratio is adjusted based on the vertical size calculated from the calculated value items. The vertical size is "66.1172", which is the sum of "29.5145" for No. 7 and "36.6028" for No. 8. The final calculated value of No. 9, "51.5", is then adjusted to match the camera's aspect ratio (3:4), which is a ratio of 1.333. Specifically, the horizontal size of the image ratio is "88.1028", which is the product of "66.1172" multiplied by the ratio of "1.333". After adjusting the calculated value of No. 9, the length of the circle corresponding to No. 9 is slightly larger than one-quarter, but the Fibonacci spiral and the subject are still included within the angle of view of the Fibonacci frame.

[0073] <Element Extraction> Next, an example of the structure of the analysis format 40 will be described. 10 is a diagram showing an example of the analysis format 40. The calculation of the numerical values ​​to be stored in the analysis format 40 and the process of storing the numerical values ​​are performed by the analysis unit 12.

[0074] 3, the analysis unit 12 extracts elements corresponding to the markers to be analyzed from the image based on the results of the area classification by the area classification unit 11, and calculates a risk assessment value (S3). The elements extracted by the analysis unit 12 from the image are basic elements (people, materials, or equipment) corresponding to the center point of the Fibonacci spiral 22a and the control points of the three marker points.

[0075] The analysis unit 12 inputs 15 element variables (binary data of either 0 or 1) into the point and marker cells of the analysis format 40 based on the region classified by the region classification unit 11 and the position of the marker. An element variable of "0" indicates that no element is shown at the point or marker position, and "1" indicates that an element is shown. The cells in which the element variables are stored correspond to the 15 cells surrounded by thick black frames in the analysis format 40. The analysis format 40 has major items such as element information, severity (element), and evaluation. The numerical values ​​stored in each item will be explained below with reference to image (2) in Figure 6.

[0076] In the element information item, the photographing date, lot, point (P00) M23 items are stored as medium items. The photographing date field stores the date on which the construction site was photographed. The lot item stores a lot number for identifying the captured image.

[0077] The Point (P00) M23 field stores information indicating what object is shown at the marker M23, which corresponds to the center of the Fibonacci spiral 22a. In the example of Fig. 10, a machine is shown at the position of the marker (M23), so "1" is stored in the "2" field, which represents Machine (machine or equipment).

[0078] The severity (element) item stores the marker (M100), marker (M50), and marker (M78) items as medium items. Severity is an index used for risk estimation in the Ministry of Land, Infrastructure, Transport and Tourism's risk assessment evaluation criteria. In the marker (M100), marker (M50), and marker (M78) items, "0.78" is stored in "1" representing Man, "0.23" is stored in "2" representing Machine (machine or equipment), and "0.5" is stored in "3" representing Material. As mentioned above, these values ​​are coefficients set in order of risk in the event of a disaster: Man (0.78), Machine (0.5), and Material (0.23).

[0079] As explained with reference to Fig. 6, each point or marker is assigned an item: "0" representing unclassifiable, "1" representing Man, "2" representing Machine (machine or equipment), and "3" representing Material. If something corresponding to each element is captured at the marker's position on the image, a "1" is stored; if not, a "0" is stored.

[0080] In the analysis format 40 shown in FIG. 10, a "1" is stored in the item "2" representing the Machine (machine or equipment) of point (P00) M23, so a Machine (machine or equipment) is shown at the position of marker M23. Also, a "1" is stored in the item "1" representing the Man (person) of marker (M100), so a Man (person) is shown at the position of marker M100. Similarly, a "1" is stored in the item "3" representing the Material (material) of marker (M100), so a Material (material) is shown at the position of marker M100. A "1" is stored in the item "2" representing the Machine (machine or equipment) of marker (M78), so a Machine (machine or equipment) is shown at the position of marker M78.

[0081] For the severity (element) markers (M100), (M50), and (M78), the coefficient and numerical value are multiplied to calculate the severity (element). However, only 0 or 1 can be stored in the items marker (M100), marker (M50), and marker (M78). Therefore, the severity (element) of marker (M100) is calculated as 0.78, the severity (element) of marker (M50) is calculated as 0.5, and the severity (element) of marker (M78) is calculated as 0.23.

[0082] The evaluation items include vulnerability (occupancy rate) and threat (marker) items as medium items. Vulnerability is a value calculated by multiplying the occupancy rate by the weight (severity). The analysis unit 12 calculates the occupancy rate of each element in the image based on the results of the area classification unit 11 performing area classification into basic elements (people, materials, or equipment) based on the image data. For example, the occupancy rate of "0," which represents unclassifiable, is 0.423, and the occupancy rate of "1," which represents Man, is 0.117. The occupancy rate of "2," which represents Machine (machine or equipment), is 0.290, and the occupancy rate of "3," which represents Material, is 0.170. The sum of each occupancy rate is "1."

[0083] Looking at the column for "1," which represents Man, the coefficient is 0.78 and the occupancy rate is 0.117, so vulnerability is calculated as 0.091, calculated as coefficient x occupancy rate. Vulnerability is similarly calculated as coefficient x occupancy rate for "2," which represents Machine (machine or equipment), and "3," which represents Material. The sum of the coefficient x occupancy rate values ​​calculated for each element is then calculated as 0.2430.

[0084] In the threat (marker) item, the coefficients and numerical values ​​of the severity (element) markers (M100), (M50), and (M78) calculated earlier are multiplied and transferred to the items M0.23, M0.5, and M0.78.Then, the values ​​of the top two severity levels are multiplied and stored as the total.For example, 0.78 multiplied by 0.5 is 0.39, and this is stored as the threat (marker) value.

[0085] On the right side of the analysis format 40, there are provided items for risk evaluation value, evaluation value 1 (expert), and evaluation value 2 (expert). The item of risk evaluation value stores an evaluation value evaluated by the analysis unit 12 using the analysis format 40. In the example of Fig. 6, the evaluation value is 0.09475.

[0086] The evaluation value 1 (expert) item stores an evaluation value 1 calculated from evaluation value 2, which will be described later, judged by an experienced worker (referred to as an "expert") after viewing the image. Evaluation value 1 is a value based on evaluation value 2 entered by the expert. If evaluation value 2 is within the range of 1 to 5, "0" is entered in evaluation value 1 as "caution." If evaluation value 2 is within the range of 5.5 to 10, "1" is entered in evaluation value 1 as "safe."

[0087] The evaluation value 2 (expert) field stores the evaluation value 2 that an expert judges after viewing the image. The evaluation value 2 is a value that the expert evaluates based on the results of the angle-of-view evaluation and rule-based evaluation. For each field, one of the values ​​"0.0", "0.5", or "1.0" is entered, and the sum of the values ​​entered for the 10 fields is calculated as the evaluation value 2.

[0088] The angle of view evaluation includes, for example, the following items. It looks safe and feels secure. · The point is clear. ·No waste. -You can predict what will happen next. · Areas for improvement can be predicted.

[0089] Examples of rule-based evaluation include the following items: -There are two or more of the 4M. Fibonacci spiral 22a is within the frame. -The 22a angle of the Fibonacci spiral is good. -The marker position is acceptable. - The location of the points is satisfactory.

[0090] 6, the evaluation value 2 is 6.5. In this way, the evaluation unit 13 evaluates whether the Fibonacci spiral 22a is properly captured in the image based on the results of the angle-of-view evaluation and the rule-based evaluation, and outputs the risk evaluation value and the evaluation value 1.

[0091] For example, the evaluation unit 13 determines that the risk assessment value is 0.09475, and the evaluation value 1 by the expert is 1.0, so there is a discrepancy between the risk assessment value and the evaluation value 1. Therefore, the evaluation unit 13 indicates to the worker as an evaluation result that caution is required at the construction site 24.

[0092] <Calculation of risk prediction value> Next, in the process shown in Fig. 3, the analysis unit 12 analyzes each element in the image and outputs a risk prediction value. At this time, the analysis unit 12 performs a regression analysis to determine whether the Fibonacci spiral 22a matches the image (S4). Here, the process by which the analysis unit 12 calculates the risk prediction value will be described with reference to Fig. 11.

[0093] 11 is a diagram showing an example of a risk prediction value calculation table 60. The analysis unit 12 uses the risk prediction value calculation table 60 as an example of a risk prediction model for predicting an evaluation value, and outputs the results of analyzing each element in the image as a risk prediction value.

[0094] The risk prediction model used as the risk prediction value calculation table 60 is set as an optimal discriminant using a linear function consisting of 11 coefficients obtained from the results of regression analysis using analytical values ​​from images of current and past construction sites and learning data from multiple experts. The analysis unit 12 inputs the binary information of each marker of severity (element) shown in Figure 10 into the risk prediction value calculation table 60 to calculate the risk prediction value.

[0095] For example, the following linear function, equation (2), is set as a method for predicting one objective variable from multiple quantitative explanatory variables. In equation (2), x, which ranges from 1 to 11, is multiplied by 11 coefficients, a1 to a11, and then an intercept b and a risk assessment value c are added. The value of y in equation (2) used in regression analysis is the value of the perspective assessment by an expert, and takes a value between 0 and 5. y = a1*1 + a2*2 + b + c × risk assessment value …(2)

[0096] The risk prediction value calculation table 60 has the following items: major item, minor item, fixed value, and minor item x fixed value. The major items store values ​​to be substituted into the risk prediction model. The intercept corresponds to intercept b in equation (2). The M100_0 to M100_3 items are transcribed with values ​​stored in "0" to "3" in the marker (M100) of the analysis format 40. The M50_0 to M50_3 items are transcribed with values ​​stored in "0" to "3" in the marker (M50) of the analysis format 40. The M78_0 to M78_3 items are transcribed with values ​​stored in "0" to "3" in the marker (M78) of the analysis format 40. The risk assessment value item is transcribed with values ​​stored in the risk assessment value of the analysis format 40.

[0097] In the minor items, "1" is stored in the items that are the subject of the calculation in formula (2), and "1" is stored in the items that are not the subject of the calculation. However, in the risk assessment value item, the value stored in the risk assessment value of the analysis format 40 is transcribed. In the example of FIG. 11, "0.09475" shown in FIG. 10 is stored as the risk assessment value. If the image to be assessed is different, the risk assessment value will also be different.

[0098] In the fixed value item, the intercept b, coefficients a1 to a11, and fixed value c are stored as fixed values ​​used in equation (2). The subitem x fixed value item stores a value obtained by multiplying the value stored in the subitem by the value stored in the fixed value item for each row. A risk prediction value item is provided below the risk prediction value calculation table 60. In the risk prediction value item, the total sum of the subitems x the fixed value is stored as the risk prediction value.

[0099] <Evaluation result output> 3, the evaluation unit 13 displays the evaluation result on the information processing terminal 1 held by the worker based on the risk prediction value, and also displays a scenario for recognizing the error (S5). Here, the evaluation result displayed on the information processing terminal 1 will be described with reference to FIG. FIG. 12 is a diagram showing an example of the evaluation result display unit 70. As shown in FIG.

[0100] The evaluation results are composed of items such as evaluation comments, thresholds, appropriateness judgment results, risk prediction value, evaluation, and impact. The evaluation unit 13 outputs the risk prediction value output from the risk prediction model, which indicates the stability of the safety or security evaluation (e.g., the ease of evaluation due to the presence of all elements), on a 10-point scale, and the word "safe" or "cautious" as a guide to the risk prediction value, as a visualization of the evaluation. The evaluation unit 13 also displays the marker position and element, with the element most related to the center point as the impact.

[0101] In the evaluation result (1) in Fig. 12, "5" is displayed in the threshold value, and "OK_match" and an upward arrow are displayed in the appropriateness determination result. "OK_match" indicates that the setting of the Fibonacci spiral 22a for the construction site 24 is appropriate.

[0102] As described above, the field of view evaluation and rule evaluation store a numerical value of 0, 0.5, or 1.0 for each item. If the total value is within the range of 0 to 10, it is determined as "OK_match," and the worker is notified that there is no problem with the field of view, etc. of the image. If the total value is not within the range of 0 to 10, it is determined as "UN_match," and the worker is notified that there is a problem with the field of view, etc. of the image. In this case, the worker must re-image the construction site 24.

[0103] In addition, the risk prediction value in the evaluation result (1) is "3.24", which is below the threshold of "5". Therefore, the evaluation item is displayed as "Caution". Also, it is shown that the Man (person) of marker M100 displayed in the impact degree is the element most related to the Machine (machine or equipment) located at the center point.

[0104] On the other hand, if the evaluation value deviates from the range width (1 to 10), the evaluation unit 13 determines that there is a problem with the setting of the Fibonacci spiral 22a when extracting elements. For example, if the arrangement direction of the Fibonacci spiral 22a is misaligned with respect to the angle of view, the evaluation value is more likely to deviate from the range width. Note that the range represents the range of data, and in the case of quantitative information, it is calculated using the formula (maximum value - minimum value). In the case of qualitative information, the range is either 1 or 0.

[0105] In the evaluation result (2) of Figure 12, "5" is displayed in the threshold value, and "UN_match" and a downward arrow are displayed in the appropriateness judgment result. "UN_match" indicates that the setting of the Fibonacci spiral 22a for the construction site 24 is inappropriate. In addition, the risk prediction value is "-4.03", which is outside the range width (1 to 10). Therefore, the evaluation is displayed as "Caution". In addition, it is shown that the Material of the marker M78 displayed in the influence degree is most related to the element located at the center point.

[0106] When the evaluation unit 13 issues an alert that the evaluation is "alert," the worker adjusts the center point and direction of the Fibonacci spiral 22a and again captures an image of the construction site 24. By repeatedly adjusting the Fibonacci spiral 22a and capturing images, "peace of mind" is displayed in the evaluation item as the evaluation value based on the optimal angle of view.

[0107] When the evaluation result shown in Fig. 12 is alerted as "alert," the worker may not know how to adjust the Fibonacci spiral 22a to take a photograph. Therefore, to make the worker aware of this, a scenario display unit 71 shown in Fig. 13 may be provided.

[0108] 13 is a diagram showing a display example of the scenario display section 71. The scenario display section 71 is displayed together with the evaluation result display section .

[0109] In order to promote awareness in decision-making during construction, the management points output by the evaluation unit 13 are displayed in a scenario format. In the evaluation results shown in Fig. 13, the symbols (a) to (d) are assigned in the order of "awareness scenarios" in the figure.

[0110] A person is shown at the center point (a) of Fibonacci spiral 22a. At marker M78, a position with a high degree of influence on the center point (b), equipment is shown. To prevent any disruption to the influence on the center point, workers need to pay close attention to the correlation between the person shown at the center point and the related factors (c) and (d). This makes it possible to simulate countermeasures that include the person shown at marker M100, represented by symbol (c), and the material shown at marker M50, represented by symbol (d).

[0111] For example, if contact between the worker at the position indicated by symbol (a) and the vehicle at the position indicated by symbol (b) is to be prevented, a scenario is issued in which measures are taken to strengthen monitoring by the worker at the position indicated by symbol (c) or to adjust the amount of material discharged at the position indicated by symbol (d).The "awareness scenario" item in the scenario display section 71 displays a scenario in which, for example, the error check by the person indicated by symbol (a) should focus on the relationship between the material indicated by symbol (d) and the person indicated by symbol (c) when using the equipment indicated by symbol (b).

[0112] The method of checking each element is to imagine the occurrence of a defect by assuming the influence of each element in a spiral from a central point in order.By correlating and checking the elements that make up the frame, such as the posture of the person and the use of protective equipment (code (a)), the quality of the material (code (d)), the operating sound of the vehicle (code (b)), and the position of the person (code (c)), the whole can be checked in a correlated manner.

[0113] Note that images may not include human elements. For example, an image may only contain equipment (vehicles, scaffolding, and cranes) and material (steel and steel plates) elements. In this case, by creating a scenario based on the relationship between the vehicle and vehicles at adjacent construction sites, the system will prompt workers to take into consideration human elements that have not been entered as related elements. For example, if only people and equipment are included in the field of view, an alert will be displayed to take materials into consideration, and information that takes into consideration elements that have not been entered can be displayed, encouraging workers to become aware of the issue.

[0114] Returning to the explanation of FIG. The worker checks the evaluation results displayed on the information processing terminal 1 and makes a decision to eliminate the error from the construction site 24. The evaluation results are also recorded in the recording unit 14 and shared with other workers (S6). This ensures the safety of the construction site 24 where the error occurred.

[0115] Here, examples of quality of images taken of the actual construction site 24 will be described with reference to FIGS. 14 and 15. FIG. FIG. 14 is a diagram showing an example of an image on which a Fibonacci spiral 22a is superimposed. Image (1) in Figure 14 is an example of a landscape image, showing workers and a construction site. Image (2) in Figure 14 is an example of a vertically long image, showing workers and a construction site.

[0116] Both images (1) and (2) show a thick Fibonacci spiral 22a superimposed on the image. The Fibonacci spiral 22a in image (1) is represented by lines spreading clockwise from the center. Image (1) was taken with people, materials, and equipment well-positioned.

[0117] The Fibonacci spiral 22a in image (2) is represented by lines spreading clockwise from the center. Image (2) does not show people, but does show machinery and the ground. Image (2) was taken in good condition, allowing the ground to be seen.

[0118] In this way, the Fibonacci spiral 22a can be photographed superimposed on the construction site 24 and used for evaluation, whether it is in the horizontal direction as in image (1) or the vertical direction as in image (2). The orientation of the Fibonacci spiral 22a may be either clockwise from the center or counterclockwise from the center.

[0119] FIG. 15 is a diagram showing an example in which a captured image has a defect. Image (3) is an example of a landscape image, with scaffolding visible across the entire image. As a result, it is unclear what to focus attention on as a whole. Image (4) is an example of a landscape image, showing a construction site and a person. However, both the construction site and the person are out of focus, making it difficult to accurately assess the risk.

[0120] According to the site management system 100 of the embodiment described above, even an inexperienced person can capture an image of the construction site 24 using the Fibonacci spiral 22a drawn on the spiral frame 22 as a guide. Furthermore, the nudge server 10 outputs the results of analyzing images containing at least one of the 4Ms to the information processing terminal 1. This allows not only inexperienced people but also all workers to share the results and become aware of risks at the construction site through the display of alerts, etc. As a result, workers who gain insights from the site management system 100 can make decisions, such as issuing optimal instructions for dealing with risks. For example, by setting rules for evaluating error factors in the site management system 100, workers can become more aware of the risks from images of the construction site 24. In situations where an error may occur, workers can improve the construction site 24 to ensure a stable and secure work environment.

[0121] When a worker takes an image of the construction site 24, he or she can use the spiral frame 22 to consciously manipulate the 4M elements so that they coincide with the points of the Fibonacci spiral 22a. In addition, the results of the image analysis are quantified, making it easy for a third party to evaluate the image.

[0122] The site management system 100 quantifies the degree to which safety and security assessments can be ensured based on images of the construction site 24. This allows for the maintenance of regularity in decision-making through standardization of inspection methods and assessments. In other words, rather than determining whether a site is safe or unsafe based on images, the site management system 100 displays on the screen whether a quantitative angle of view that is uniformly easy to assess has been ensured, thereby enabling the standardization of assessments among different workers. Furthermore, each worker can share regular information, and by standardizing decision-making, human skills can be continuously maintained. As a result, the benefits of "techniques" for utilizing new technology and "learning" through information sharing can be achieved. Furthermore, workers can become aware of risks, even for errors that are not solely due to new technology.

[0123] Markers are attached to the spiral frame 22, so even an inexperienced person can easily capture an image so that the 4M elements are reflected on the markers. This makes it easier for an inexperienced person to clearly understand the key points when capturing an image. It also improves the accuracy with which an expert who evaluates the captured image recognizes the image. Furthermore, both inexperienced and experienced people can have a common understanding of the risks at the construction site 24, which makes it possible to improve risk recognition skills. This makes it possible for many people, including inexperienced and experienced people as well as foreigners and women who are expected to be employed at construction sites in the future, to make risk-related decisions at a similar level.

[0124] Furthermore, the site management system 100 can ensure effectiveness not only in safety management at work sites, but also in improving quality, processes, costs, and skills, which are elements of a construction production system. For example, if an alert is displayed at a construction site 24, improvements will be made, improving the quality of the construction site. Furthermore, because the causes of errors are eliminated in advance, extra processes and costs associated with the occurrence of an error are avoided. Furthermore, by continuously using the site management system 100, workers will improve their skills in maintaining a safe construction site, and will be able to build safe construction sites even when they go to other construction sites.

[0125] In the above-described embodiment, the Fibonacci spiral 22a is used, but various other shapes that fall within the golden ratio may also be used for imaging.

[0126] [Variation 1] In the above-described embodiment, the image is divided into 15 rows and 20 columns to perform region classification. However, the number of regions does not need to be limited in order to perform more detailed region classification. FIG. 16 is a diagram showing an example in which the region classification unit 11 automatically determines the type of element and performs region classification.

[0127] The region classification unit 11 automatically classifies elements appearing in an image, for example, using a semantic segmentation technique. The region classification unit 11 then performs color coding or other processing according to the type of automatically determined element. This processing makes it possible to accurately classify elements of complex shapes, not limited to sections divided into 15 rows and 20 columns. It also makes it possible to accurately classify how multiple elements are included in the marker points of the Fibonacci spiral 22a. Furthermore, by having an operator label elements classified using the semantic segmentation technique as correct or incorrect, the region classification unit 11 can learn from the labeled elements and improve classification accuracy.

[0128] Even if the elements are classified using the semantic segmentation method, the occupancy rate of each element relative to the entire image can be easily calculated. Therefore, the analysis by the analysis unit 12 using the analysis format 40 can be performed in the same manner as in the above-described embodiment.

[0129] [Variation 2] Marker positions set on the Fibonacci spiral 22a are also conceivable other than those shown in Fig. 5. Here, modified examples of marker positions will be described with reference to Fig. 17. FIG. 17 is a diagram showing a modified example of the positions of markers set on the Fibonacci spiral 22a.

[0130] In images (1) to (4) of Figure 17, marker points M50, M78, and M100 are set at the centers of gravity of different quarter circles, respectively. Of the reference symbols 31 to 37 used to identify the downward-pointing triangular marks shown in Figure 5, only the parts corresponding to the marker points are labeled.

[0131] In the Fibonacci spiral 22a shown in image (1), M50 is set to the symbol 35, M78 is set to the symbol 36, and M100 is set to the symbol 37. In the Fibonacci spiral 22a shown in image (2), M50 is set to the symbol 34, M78 is set to the symbol 36, and M100 is set to the symbol 37. In the Fibonacci spiral 22a shown in image (3), M50 is set to the symbol 34, M78 is set to the symbol 35, and M100 is set to the symbol 36. In the Fibonacci spiral 22a shown in image (4), M50 is set to the symbol 34, M78 is set to the symbol 35, and M100 is set to the symbol 33.

[0132] In either case, verification showed that this was sufficient for workers to check for errors at the construction site. In this way, it was sufficient to take a photo so that the target corresponding to 4M overlaps the position of at least three markers.

[0133] The present invention is not limited to the above-described embodiment, and it goes without saying that various other applications and modifications are possible without departing from the gist of the present invention as set forth in the claims. For example, the above-described embodiment has described the system configuration in detail and specifically to clearly explain the present invention, and is not necessarily limited to a system including all of the described configurations. Furthermore, it is also possible to add, delete, or replace part of the configuration of this embodiment with other configurations. In addition, the control lines and information lines shown are those that are considered necessary for the explanation, and do not necessarily show all the control lines and information lines in the product. In reality, it can be assumed that almost all components are interconnected. [Explanation of symbols]

[0134] 1...information processing terminal, 2...imaging unit, 3...information display unit, 10...nudge server, 11...area classification unit, 12...analysis unit, 13...evaluation unit, 14...recording unit, 21...worker, 22...spiral frame, 22a...Fibonacci spiral, 24...construction site, 40...analysis format, 60...risk prediction value calculation table, 70...evaluation result display unit, 71...scenario display unit, 100...site management system

Claims

1. an area classification unit that captures an image of the construction site together with a scale and classifies the image, in which at least one element of people, machinery or equipment, and materials is captured, into an area for each of the elements; an analysis unit that extracts the points at which the elements are located in the image based on the correspondence between the points set on the scale and the area, and analyzes the risk of the construction site based on the extracted points; an evaluation unit that evaluates the analysis result and outputs the evaluation result; Site management system.

2. The area classification unit classifies the elements that appear in the image as people, equipment, or materials, and classifies parts of the image that do not include the elements as unclassifiable. The site management system according to claim 1 .

3. The analysis unit extracts the elements that appear at the positions of the markers attached to the scale, calculates the occupancy rate of the elements in the image for each element, and calculates a risk assessment value by multiplying the sum of values ​​obtained by multiplying the occupancy rate by a weight by the highest severity value calculated for each marker on which the element appears. The site management system according to claim 2 .

4. The analysis unit calculates a risk prediction value by substituting the element reflected at the position of the marker attached to the scale into a risk prediction model. The site management system according to claim 3 .

5. The evaluation unit compares the risk prediction value with a threshold value, and outputs information indicating that the risk of the construction site is low when the risk prediction value is equal to or greater than the threshold value, and outputs information indicating that the risk of the construction site is high when the risk prediction value is less than the threshold value. The site management system according to claim 4.

6. The evaluation unit outputs information indicating a scenario for having a worker at the construction site check the construction site based on a relationship between the element shown at the start position of the scale and other elements that have a high degree of influence on the element. The site management system according to claim 5 .

7. The evaluation unit evaluates whether the scale is appropriately reflected in the image based on the results of the angle-of-view evaluation and the rule-based evaluation, and outputs the evaluation information. The site management system according to claim 6.

8. The scale is a processed Fibonacci spiral, and the markers are set at the centers of gravity of the quarter circles that make up the Fibonacci spiral. The site management system according to any one of claims 1 to 7.

9. an imaging unit that captures the image together with the scale at the construction site; and an information display unit that displays the information output from the evaluation unit. The site management system according to claim 8.

10. A step of classifying an image of the construction site together with the scale into regions for each element; extracting the points at which the elements are located in the image based on the correspondence between the points set on the scale and the areas, and analyzing the risk of the construction site based on the extracted points; and outputting an evaluation result obtained by evaluating the analysis result. On-site management methods.

Images