An engineering construction management method and system based on AI

Abstract

The present invention provides an AI-based construction management method and system, wherein the method comprises: obtaining construction images in a construction site; detecting construction risks based on the construction images and a preset AI detection model; and providing construction warnings for the construction site based on the construction risks. The AI-based construction management method and system of the present invention introduces an AI detection model, obtains construction images, and performs construction risk detection based on the AI ​​detection model. There is no need to set up multiple supervisors at the construction site, which reduces labor costs and improves the comprehensiveness and efficiency of construction safety supervision at the construction site.

Description

Technical Field

[0001] The present invention relates to the field of AI processing technology, and in particular to an engineering construction management method and system based on AI. Background Art

[0002] At present, in order to ensure the safe and stable construction of the project, construction safety supervision is required at the construction site. Generally, when conducting construction safety supervision, it is necessary to set up multiple supervisors at the construction site, which causes labor costs. In addition, when supervisors conduct construction safety supervision at the construction site, they mostly adopt the method of patrol inspection, which makes it difficult to conduct comprehensive safety supervision of the construction site and improve supervision efficiency. Therefore, a solution is urgently needed. Summary of the invention

[0003] One of the purposes of the present invention is to provide an AI-based engineering construction management method, introduce an AI detection model, obtain construction images, and perform construction risk detection based on the AI ​​detection model. There is no need to set up multiple supervisors at the construction site, thereby reducing labor costs and improving the comprehensiveness and efficiency of construction safety supervision at the construction site.

[0004] An embodiment of the present invention provides an AI-based engineering construction management method, comprising:

[0005] Acquire construction images within the construction site;

[0006] Detect construction risks based on construction images and preset AI detection models;

[0007] Based on construction risks, construction early warning is carried out on the construction site.

[0008] Preferably, obtaining a construction image in a construction site includes:

[0009] Obtain construction status information within the construction site;

[0010] The construction status information is displayed on a preset three-dimensional map model corresponding to the construction site;

[0011] Determine the area to be photographed that meets the conditions of the area to be photographed from the three-dimensional map model;

[0012] Get the flight position of the drone within the construction site;

[0013] Represent the flight position on a three-dimensional map model;

[0014] Plan the flight route based on the area to be photographed and the flight position on the 3D map model;

[0015] Based on the flight route, the drone is controlled to take images of the area to be photographed, and used as the construction image;

[0016] The conditions of the area to be photographed include:

[0017] The model content type set within the area to be photographed exists in a preset standard content type set library;

[0018] The time of the most recent historical construction risk in the area to be photographed is no more than the preset first time period;

[0019] The last time the drone took photos in the shooting area is no more than the preset second time period.

[0020] Preferably, based on the flight route, controlling the drone to fly and photograph the regional image of the area to be photographed includes:

[0021] Control the drone to start flying along the flight route;

[0022] Determine the first straight-line distance between the drone and the center of the area to be photographed;

[0023] When the first straight-line distance is less than or equal to a preset first distance threshold, determining a first landing position that meets the first landing position condition from the three-dimensional map model;

[0024] Control the drone to the first landing position;

[0025] Controlling the drone to capture a first image of the area to be captured;

[0026] Based on the first image, determine the head position and head orientation of each construction worker in the area to be photographed;

[0027] The head position and head orientation of each construction worker are represented on the three-dimensional map model;

[0028] Determine a second landing position that meets the second landing position condition from the three-dimensional map model;

[0029] Control the drone to the second landing location;

[0030] Controlling the drone to capture a second image of the area to be captured, and using the second image as the area image;

[0031] The first landing position conditions include:

[0032] The second straight-line distance between the first landing position and the area center is greater than or equal to a preset second distance threshold;

[0033] The area to be photographed completely falls within the first photographing range when the drone is located at the first landing position;

[0034] The second landing position conditions include:

[0035] The minimum straight-line distance between the second foot position and the head position is greater than or equal to a preset third distance threshold;

[0036] The vector angle between the first direction vector and the second direction vector falls within a preset angle range; the first direction vector is constructed based on the current position and shooting direction of the drone; the second direction vector is constructed based on the head position and head orientation of the construction worker;

[0037] The area to be photographed completely falls into the second photographing range when the drone is located at the second landing position;

[0038] Among them, the first distance threshold is greater than the second distance threshold and greater than the third distance threshold.

[0039] Preferably, the AI-based engineering construction management method further includes:

[0040] Update and train AI detection models.

[0041] Preferably, the AI-based engineering construction management method further includes:

[0042] Visualize and output 3D map model.

[0043] An embodiment of the present invention provides an AI-based engineering construction management system, including:

[0044] A construction image acquisition module, used to acquire construction images within the construction site;

[0045] Construction risk detection module, used to detect construction risks based on construction images and preset AI detection models;

[0046] The construction early warning module is used to provide early warning for construction sites based on construction risks.

[0047] Preferably, the construction image acquisition module acquires the construction image in the construction site, including:

[0048] Obtain construction status information within the construction site;

[0049] The construction status information is displayed on a preset three-dimensional map model corresponding to the construction site;

[0050] Determine the area to be photographed that meets the conditions of the area to be photographed from the three-dimensional map model;

[0051] Get the flight position of the drone within the construction site;

[0052] Represent the flight position on a three-dimensional map model;

[0053] Plan the flight route based on the area to be photographed and the flight position on the 3D map model;

[0054] Based on the flight route, the drone is controlled to take images of the area to be photographed, and used as the construction image;

[0055] The conditions of the area to be photographed include:

[0056] The model content type set within the area to be photographed exists in a preset standard content type set library;

[0057] The time of the most recent historical construction risk in the area to be photographed is no more than the preset first time period;

[0058] The last time the drone took photos in the shooting area is no more than the preset second time period.

[0059] Preferably, the construction image acquisition module controls the drone to fly and capture regional images of the area to be captured based on the flight route, including:

[0060] Control the drone to start flying along the flight route;

[0061] Determine the first straight-line distance between the drone and the center of the area to be photographed;

[0062] When the first straight-line distance is less than or equal to a preset first distance threshold, determining a first landing position that meets the first landing position condition from the three-dimensional map model;

[0063] Control the drone to the first landing position;

[0064] Controlling the drone to capture a first image of the area to be captured;

[0065] Based on the first image, determine the head position and head orientation of each construction worker in the area to be photographed;

[0066] The head position and head orientation of each construction worker are represented on the three-dimensional map model;

[0067] Determine a second landing position that meets the second landing position condition from the three-dimensional map model;

[0068] Control the drone to the second landing location;

[0069] Controlling the drone to capture a second image of the area to be captured, and using the second image as the area image;

[0070] The first landing position conditions include:

[0071] The second straight-line distance between the first landing position and the area center is greater than or equal to a preset second distance threshold;

[0072] The area to be photographed completely falls within the first photographing range when the drone is located at the first landing position;

[0073] The second landing position conditions include:

[0074] The minimum straight-line distance between the second foot position and the head position is greater than or equal to a preset third distance threshold;

[0075] The vector angle between the first direction vector and the second direction vector falls within a preset angle range; the first direction vector is constructed based on the current position and shooting direction of the drone; the second direction vector is constructed based on the head position and head orientation of the construction worker;

[0076] The area to be photographed completely falls into the second photographing range when the drone is located at the second landing position;

[0077] Among them, the first distance threshold is greater than the second distance threshold and greater than the third distance threshold.

[0078] Preferably, the AI-based engineering construction management system further includes:

[0079] Update training module, used to update and train AI detection models.

[0080] Preferably, the AI-based engineering construction management system further includes:

[0081] The visualization module is used to visualize and output the three-dimensional map model.

[0082] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings.

[0083] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0084] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0085] Figure 1 Schematic diagram of an AI-based engineering construction management method in an embodiment of the present invention;

[0086] Figure 2 Schematic diagram of an AI-based engineering construction management system in an embodiment of the present invention. DETAILED DESCRIPTION

[0087] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0088] The embodiment of the present invention provides an engineering construction management method based on AI, such as Figure 1 As shown, including:

[0089] Step S1: Acquire a construction image in the construction site;

[0090] Step S2: Detect construction risks based on construction images and a preset AI detection model;

[0091] Step S3: Based on the construction risks, a construction early warning is issued to the construction site.

[0092] In the above technical solution, the AI ​​detection model is an artificial intelligence model that trains the neural network model to convergence based on a large number of construction risk images [such as images of construction workers not wearing safety helmets or not wearing safety ropes, etc.]. The AI ​​detection model can determine whether there is a construction risk based on the construction image; obtain construction images on the construction site, and detect construction risks based on the construction images based on the AI ​​detection model; when construction risks are detected, the construction site is given a construction warning. This application introduces an AI detection model to obtain construction images, and performs construction risk detection based on the AI ​​detection model. There is no need to set up multiple supervisors on the construction site, which reduces labor costs and improves the comprehensiveness and efficiency of construction safety supervision on the construction site.

[0093] In one embodiment, obtaining a construction image in a construction site includes:

[0094] Obtain construction status information on the construction site; construction status information includes: the location of each construction worker on the construction site, the operation status of each operation area [such as: operation today, no operation today, etc.], the location of each construction vehicle, the location of construction equipment, etc.;

[0095] The construction status information is displayed on a preset three-dimensional map model corresponding to the construction site;

[0096] Determine the area to be photographed that meets the conditions of the area to be photographed from the three-dimensional map model; the area to be photographed is the area where the drone needs to fly to for photography;

[0097] Get the flight position of the drone within the construction site;

[0098] Represent the flight position on a three-dimensional map model;

[0099] Plan a flight route based on the area to be photographed and the flight position on the three-dimensional map model; the flight route is the shortest route from the flight position to the photographing area;

[0100] Based on the flight route, the drone is controlled to take images of the area to be photographed, and used as the construction image;

[0101] The conditions of the area to be photographed include:

[0102] The model content type set in the area to be photographed exists in the preset standard content type set library; when the model content type set exists in the standard content type set library, it means that the area to be photographed needs a drone to go there for flight photography, for example: the model content type set includes the construction personnel location [need a drone to go there for photography to realize personnel monitoring], the operation status is today's operation [the area will be operated, and a drone is needed to go there for photography], etc.;

[0103] The time of the most recent historical construction risk in the area to be photographed is no more than a preset first time period; the first time period can be, for example, 3 hours; by setting this condition of the area to be photographed, the drone can track and capture construction images of the area where the historical construction risk has recently occurred;

[0104] The last time the drone took photos of the area to be photographed is no more than a preset second time period. The second time period can be, for example, 20 minutes. By setting this condition of the area to be photographed, the drone can track and capture the area where the construction image has been captured recently.

[0105] In one embodiment, based on the flight route, controlling the drone to fly and capture the regional image of the area to be captured includes:

[0106] Control the drone to start flying along the flight route;

[0107] Determine the first straight-line distance between the drone and the center of the area to be photographed;

[0108] When the first straight-line distance is less than or equal to a preset first distance threshold, a first landing position that meets the first landing position condition is determined from the three-dimensional map model; the first distance threshold may be, for example, 150 meters;

[0109] Control the drone to the first landing position;

[0110] Controlling the drone to capture a first image of the area to be captured;

[0111] Based on the first image, the head position and head orientation of each construction worker in the area to be photographed are determined; when the first straight-line distance is less than or equal to the first distance threshold, it means that the drone is about to approach the area to be photographed; when the drone is at the first landing position, the construction workers in the area to be photographed will not hear the working sound of the drone and will not follow the sound to find the drone; at this time, the first image is captured to determine the head position and head orientation of the construction workers;

[0112] The head position and head orientation of each construction worker are represented on the three-dimensional map model;

[0113] Determine a second landing position that meets the second landing position condition from the three-dimensional map model;

[0114] Control the drone to the second landing location;

[0115] The drone is controlled to capture a second image of the area to be captured and used as the area image; when the drone is at the second landing position, the drone is closer to the area to be captured than the first landing position. Similarly, construction workers in the area to be captured will not hear the working sound of the drone and will not follow the sound to find the drone. What is different is that the workers will not see the drone either;

[0116] The first landing position conditions include:

[0117] The second straight-line distance between the first landing position and the center of the area is greater than or equal to a preset second distance threshold; the second distance threshold may be, for example, 100 meters; the second distance threshold must be set so that when the drone is at the first landing position, construction workers in the area to be photographed will not hear the working sound of the drone and will not follow the sound to find the drone, and the drone sound test can be performed in advance to complete the setting;

[0118] The area to be photographed falls completely within the first shooting range when the drone is at the first landing position; it is necessary to ensure that the drone can photograph the entire area to be photographed, so as to facilitate the determination of the head position and head orientation of each construction worker;

[0119] The second landing position conditions include:

[0120] The minimum straight-line distance between the second landing position and the head position is greater than or equal to a preset third distance threshold; the third distance threshold may be, for example, 60 meters; the setting of the third distance threshold must satisfy that when the drone is at the second landing position, construction workers in the area to be photographed will not hear the working sound of the drone and will not follow the sound to find the drone, and the drone sound test can be performed in advance to complete the setting;

[0121] The vector angle between the first direction vector and the second direction vector falls within a preset angle range; the first direction vector is constructed based on the current position and shooting direction of the drone; the second direction vector is constructed based on the head position and head orientation of the construction worker; this second landing position condition can ensure that when the drone is in the second landing position, the worker will not see the drone; the angle range can be, for example: 0 to 30 degrees;

[0122] The area to be photographed falls completely within the second shooting range when the drone is located at the second landing position; it is necessary to ensure that the drone can photograph the entire area to be photographed, so as to facilitate the collection of construction images;

[0123] The first distance threshold is greater than the second distance threshold which is greater than the third distance threshold. When the drone flies to shoot the construction image, it is a process of slowly approaching. Therefore, the first distance threshold is set to be greater than the second distance threshold which is greater than the third distance threshold.

[0124] Generally, construction sites involve some construction work that construction workers cannot be distracted from, such as aerial work and operation of large construction vehicles. In order to improve the comprehensiveness and convenience of construction image acquisition, drones can be used for flight photography. However, drones are emerging equipment and may attract the attention of construction workers during flight, causing distraction and even causing work safety accidents in serious cases. Therefore, in order to overcome this problem, the embodiments of the present invention design the above technical solution, which greatly improves the applicability of the system and is more intelligent.

[0125] In one embodiment, the AI-based engineering construction management method further includes:

[0126] Update and train AI detection models.

[0127] In one embodiment, the AI-based engineering construction management method further includes:

[0128] Visualize and output 3D map model.

[0129] The embodiment of the present invention provides an engineering construction management system based on AI, such as Figure 2 As shown, including:

[0130] The construction image acquisition module 1 is used to acquire the construction image in the construction site;

[0131] Construction risk detection module 2, used to detect construction risks based on construction images and preset AI detection models;

[0132] The construction early warning module 3 is used to provide early warning of construction on the construction site based on construction risks.

[0133] The construction image acquisition module 1 acquires the construction image in the construction site, including:

[0134] Obtain construction status information within the construction site;

[0135] The construction status information is displayed on a preset three-dimensional map model corresponding to the construction site;

[0136] Determine the area to be photographed that meets the conditions of the area to be photographed from the three-dimensional map model;

[0137] Get the flight position of the drone within the construction site;

[0138] Represent the flight position on a three-dimensional map model;

[0139] Plan the flight route based on the area to be photographed and the flight position on the 3D map model;

[0140] Based on the flight route, the drone is controlled to take images of the area to be photographed, and used as the construction image;

[0141] The conditions of the area to be photographed include:

[0142] The model content type set within the area to be photographed exists in a preset standard content type set library;

[0143] The time of the most recent historical construction risk in the area to be photographed is no more than the preset first time period;

[0144] The last time the drone took photos in the shooting area is no more than the preset second time period.

[0145] The construction image acquisition module 1 controls the UAV to fly and take regional images of the area to be photographed based on the flight route, including:

[0146] Control the drone to start flying along the flight route;

[0147] Determine the first straight-line distance between the drone and the center of the area to be photographed;

[0148] When the first straight-line distance is less than or equal to a preset first distance threshold, determining a first landing position that meets the first landing position condition from the three-dimensional map model;

[0149] Control the drone to the first landing position;

[0150] Controlling the drone to capture a first image of the area to be captured;

[0151] Based on the first image, determine the head position and head orientation of each construction worker in the area to be photographed;

[0152] The head position and head orientation of each construction worker are represented on the three-dimensional map model;

[0153] Determine a second landing position that meets the second landing position condition from the three-dimensional map model;

[0154] Control the drone to the second landing location;

[0155] Controlling the drone to capture a second image of the area to be captured, and using the second image as the area image;

[0156] The first landing position conditions include:

[0157] The second straight-line distance between the first landing position and the area center is greater than or equal to a preset second distance threshold;

[0158] The area to be photographed completely falls within the first photographing range when the drone is located at the first landing position;

[0159] The second landing position conditions include:

[0160] The minimum straight-line distance between the second foot position and the head position is greater than or equal to a preset third distance threshold;

[0161] The vector angle between the first direction vector and the second direction vector falls within a preset angle range; the first direction vector is constructed based on the current position and shooting direction of the drone; the second direction vector is constructed based on the head position and head orientation of the construction worker;

[0162] The area to be photographed completely falls into the second photographing range when the drone is located at the second landing position;

[0163] Among them, the first distance threshold is greater than the second distance threshold and greater than the third distance threshold.

[0164] The AI-based construction management system also includes:

[0165] Update training module, used to update and train AI detection models.

[0166] The AI-based construction management system also includes:

[0167] The visualization module is used to visualize and output the three-dimensional map model.

[0168] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. An AI-based engineering construction management method, characterized in that: include: Acquire construction images within the construction site; Detect construction risks based on the construction images and a preset AI detection model; Based on the construction risks, early warning of construction is provided to the construction site; The obtaining of the construction image in the construction site includes: Obtain construction status information within the construction site; The construction status information is displayed on a preset three-dimensional map model corresponding to the construction site; Determine an area to be photographed that meets the conditions of the area to be photographed from the three-dimensional map model; Get the flight position of the drone within the construction site; Representing the flight position on the three-dimensional map model; Planning a flight route based on the area to be photographed and the flight position on the three-dimensional map model; Based on the flight route, controlling the drone to fly and photograph the regional image of the area to be photographed, and using the regional image as the construction image; Based on the flight route, controlling the drone to fly and capture the regional image of the area to be captured includes: Controlling the UAV to start flying along the flight route; Determine a first straight-line distance between the drone and the center of the area to be photographed; When the first straight-line distance is less than or equal to a preset first distance threshold, determining a first landing position that meets a first landing position condition from the three-dimensional map model; Control the drone to go to the first landing position; Controlling the drone to capture a first image of the area to be captured; Based on the first image, determining the head position and head orientation of each construction worker in the area to be photographed; The head position and head orientation of each construction worker are indicated on the three-dimensional map model; Determine a second landing position that meets the second landing position condition from the three-dimensional map model; Control the drone to go to the second landing position; Controlling the drone to capture a second image of the area to be captured, and using the second image as the area image; Wherein, the first landing position condition includes: A second straight-line distance between the first landing position and the center of the area is greater than or equal to a preset second distance threshold; The area to be photographed completely falls within the first photographing range when the drone is located at the first landing position; Wherein, the second landing position condition includes: The minimum straight-line distance between the second foot position and the head position is greater than or equal to a preset third distance threshold; The vector angle between the first direction vector and the second direction vector falls within a preset angle range; the first direction vector is constructed based on the current position and shooting direction of the drone; the second direction vector is constructed based on the head position and head orientation of the construction worker; The area to be photographed completely falls within the second photographing range when the drone is located at the second landing position; Among them, the first distance threshold is greater than the second distance threshold and greater than the third distance threshold.

2. The AI-based engineering construction management method according to claim 1, characterized in that: The conditions of the area to be photographed include: The model content type set within the area to be photographed exists in a preset standard content type set library; The time from the most recent historical construction risk in the area to be photographed to the present day does not exceed a preset first time period; The time when the drone last took photos of the area to be photographed is no more than a preset second time period.

3. The AI-based engineering construction management method according to claim 1, characterized in that: Also includes: Update and train the AI ​​detection model.

4. The AI-based engineering construction management method according to claim 1, characterized in that: Also includes: The three-dimensional map model is visually output.

5. An AI-based engineering construction management system, characterized in that: include: A construction image acquisition module, used to acquire construction images within the construction site; A construction risk detection module, used to detect construction risks based on the construction image and a preset AI detection model; A construction early warning module is used to provide a construction early warning to the construction site based on the construction risks; The construction image acquisition module acquires the construction image in the construction site, including: Obtain construction status information within the construction site; The construction status information is displayed on a preset three-dimensional map model corresponding to the construction site; Determine an area to be photographed that meets the conditions of the area to be photographed from the three-dimensional map model; Get the flight position of the drone within the construction site; Representing the flight position on the three-dimensional map model; Planning a flight route based on the area to be photographed and the flight position on the three-dimensional map model; Based on the flight route, controlling the drone to fly and photograph the regional image of the area to be photographed, and using the regional image as the construction image; The construction image acquisition module controls the drone to fly and capture the regional image of the area to be captured based on the flight route, including: Controlling the UAV to start flying along the flight route; Determine a first straight-line distance between the drone and the center of the area to be photographed; When the first straight-line distance is less than or equal to a preset first distance threshold, determining a first landing position that meets a first landing position condition from the three-dimensional map model; Control the drone to go to the first landing position; Controlling the drone to capture a first image of the area to be captured; Based on the first image, determining the head position and head orientation of each construction worker in the area to be photographed; The head position and head orientation of each construction worker are indicated on the three-dimensional map model; Determine a second landing position that meets the second landing position condition from the three-dimensional map model; Control the drone to go to the second landing position; Controlling the drone to capture a second image of the area to be captured, and using the second image as the area image; Wherein, the first landing position condition includes: A second straight-line distance between the first landing position and the center of the area is greater than or equal to a preset second distance threshold; The area to be photographed completely falls within the first photographing range when the drone is located at the first landing position; Wherein, the second landing position condition includes: The minimum straight-line distance between the second foot position and the head position is greater than or equal to a preset third distance threshold; The vector angle between the first direction vector and the second direction vector falls within a preset angle range; the first direction vector is constructed based on the current position and shooting direction of the drone; the second direction vector is constructed based on the head position and head orientation of the construction worker; The area to be photographed completely falls within the second photographing range when the drone is located at the second landing position; Among them, the first distance threshold is greater than the second distance threshold and greater than the third distance threshold.

6. The AI-based engineering construction management system according to claim 5, characterized in that: The conditions of the area to be photographed include: The model content type set within the area to be photographed exists in a preset standard content type set library; The time from the most recent historical construction risk in the area to be photographed to the present day does not exceed a preset first time period; The time when the drone last took photos of the area to be photographed is no more than a preset second time period.

7. The AI-based engineering construction management system according to claim 5, characterized in that: Also includes: The update training module is used to update and train the AI ​​detection model.

8. The AI-based engineering construction management system according to claim 5, characterized in that: Also includes: A visualization module is used to visualize and output the three-dimensional map model.

Images