Construction progress identification method and device for construction site, equipment and medium

Abstract

The invention discloses a construction progress identification method and device for a construction site, equipment and a medium, and the method comprises the steps: obtaining a first key point set and a panoramic image through obtaining video data uploaded by a panoramic inspection system and carrying out image synthesis processing; according to an identification area marked in the project drawing file in advance, selecting N key points in the identification area from the first key point set to form a second key point set; according to the angle between the target object and each key point in the N key points, reducing the visual angle range of the panoramic image, and generating images of M target angles corresponding to the N key points; performing construction progress identification on the image by using a neural network model to obtain M engineering construction progress labels; and determining a target progress in the M construction progress labels based on a voting mechanism, and sending an identification result of the target progress to the system, thereby realizing automatic identification of the construction site progress, and enabling construction site progress supervision not to depend on experiences of engineers any more.

Description

Technical Field

[0001] The present invention relates to the technical field of engineering construction, and in particular to a construction progress identification method, device, electronic equipment and medium for a construction site. Background Art

[0002] Project management on construction sites is the basis for ensuring that large-scale projects can be carried out effectively and safely. Nowadays, the increase in the types of engineering projects and the expansion of their scale have put higher and higher requirements on construction project management. Project management on construction sites is a complex and critical task, which involves management and coordination in many aspects. For example, construction plan and progress management: This is one of the most core aspects of project management. A detailed construction plan needs to be formulated, including construction progress, construction methods, resource allocation, etc., and timely adjustments should be made according to actual conditions. At the same time, it is necessary to ensure that the construction progress is consistent with the requirements of design drawings, construction contracts, etc. Quality control: Quality control is crucial in construction site project management. A complete quality management system needs to be established, including material testing, construction process control, acceptance standard formulation, etc., to ensure that the construction project meets the design requirements and quality standards. In short, construction site project management requires coordination and management in many aspects to ensure the smooth implementation of the project and the realization of quality, cost, safety and other goals.

[0003] Traditional engineering project management methods rely more on manual analysis and judgment, requiring managers to have rich industry experience, and it is difficult to achieve standardized and process-based engineering progress control. Summary of the invention

[0004] Therefore, in order to solve the problem that the traditional project management method relies more on manual analysis and judgment, requires managers to have rich industry experience, and is difficult to achieve standardized and process-based project progress control, the embodiments of the present invention provide a construction progress identification method, device, electronic device and medium for a construction site, and specifically disclose the following technical solutions:

[0005] In a first aspect, an embodiment of the present invention discloses a method for identifying a construction progress of a construction site, the method comprising:

[0006] The video data uploaded from the panoramic inspection system is obtained, and the video data is the video image data taken on the construction site when the user passes through the target path; the video data is subjected to image synthesis processing to obtain a first key point set and a panoramic image corresponding to each key point, and the first key point set includes at least one key point; according to the identification area pre-marked in the project drawing file, N key points in the identification area are selected from the first key point set to form a second key point set; according to the angle between the target object and each of the N key points, the viewing angle range of the panoramic image is narrowed to generate images of M target angles corresponding to the N key points, M ≥ N; the neural network model is used to identify the construction progress of the images of the M target angles to obtain M engineering construction progress labels; based on the voting mechanism, the target progress is determined from the M construction progress labels, and the identification result of the target progress is sent to the panoramic inspection system.

[0007] The method provided in this aspect performs image synthesis processing on original video data to obtain a panoramic image and key points on the panoramic line, calculates the angle of the target object facing the key points on the image, generates a local image when the key point position is facing the target object, processes the local image with a classification algorithm to obtain a classified second local image, and finally performs a voting algorithm on the second local image to obtain the construction site progress corresponding to the second local image, thereby realizing the identification of the construction progress of the construction site. Compared with manual analysis and judgment, this method saves the analysis time of the construction progress and no longer relies on experienced staff for analysis and judgment.

[0008] In combination with the first aspect, in a possible implementation, N key points within the recognition area are selected from the first key point set to form a second key point set, including: obtaining the coordinates of each key point in the first key point set; obtaining the marked recognition area; based on the marked recognition area, filtering the key points in the first key point set to obtain a filtered second key point set, the second key point set containing N key points, and the coordinates of the N key points are all within the range of the recognition area.

[0009] The method provided in this aspect calculates a second key point set belonging to the target identification area in the first key point set based on the key points of the engineering building and the range of the target identification area. The second key point set contains N key points, and the N key points provide assistance for subsequent processing of the panoramic image of the target identification area.

[0010] In combination with the first aspect, in a possible implementation, according to the angle between the target object and each of the N key points, the viewing angle range of the panoramic image is narrowed, and images of M target angles corresponding to the N key points are generated, M≥N, which includes: obtaining the center point of the identification area and the panoramic first image of the identification area; based on the center point of the identification area and the panoramic first image, generating the angles of N key points in the second key point set facing the center point; based on the angles facing the center point, narrowing the visual range, and obtaining M second images of the N key points facing the center point, the M second images being images of the M target angles corresponding to the N key points.

[0011] The method provided in this aspect generates M images of the centers corresponding to N key points by obtaining the center point of the recognition area. The original image of the N key points is a 360° panoramic image. The angle of the panoramic image is reduced by the angle of the center corresponding to the key point, so that the image range of the key point is an image facing the center point, thereby obtaining a 360° panoramic image with the center point as the key point, which provides help for subsequent image recognition operations and classification algorithm processing.

[0012] In combination with the first aspect, in a possible implementation, a neural network model is used to identify the construction progress of images of M target angles to obtain M engineering construction progress labels, including: obtaining a neural network model, the neural network model is obtained by training project progress data, and the neural network model contains the correlation between the project progress image and the engineering construction progress; the images of M target angles are input into the neural network model, the engineering construction progress corresponding to each target angle image is searched according to the correlation, and the corresponding engineering construction progress label is generated to obtain M engineering progress labels.

[0013] The method provided in this aspect obtains a neural network model; inputs images of M target angles into the neural network model; searches for the engineering construction progress corresponding to each target angle image according to the association relationship; generates the corresponding engineering construction progress label; and finally obtains M engineering progress labels. The neural network model can automatically recognize the input image and output the corresponding engineering construction progress label without manual intervention, thereby improving work efficiency. In addition, the neural network model can continuously learn and update, can adapt to the progress recognition needs of different projects and different construction stages, and has strong scalability.

[0014] In combination with the first aspect, in a possible implementation, a target progress is determined among M construction progress labels based on a voting mechanism, including: obtaining M construction progress labels, the M construction progress labels including a first priority of the construction progress; counting multiple votes corresponding to the M construction progress labels, and obtaining a first number of votes from the multiple votes, the first number of votes being the number of labels corresponding to the construction progress of the same priority; if the construction progress priority to which the first number of votes belongs is the first priority, determining the target progress as the construction progress corresponding to the first priority.

[0015] The method provided in this aspect obtains M construction progress labels and the first priority of the construction progress, then counts multiple votes corresponding to the M construction progress labels to obtain the first vote, and finally determines the construction progress priority to which the first vote belongs to determine the target progress. During the voting process, the priorities of different construction progresses are taken into consideration, which can reflect the situations of different construction progresses, making the determination of the target progress more reasonable. The voting mechanism is an objective decision-making method, which can avoid the influence of subjective factors on the determination of the target progress and improve the objectivity and fairness of the decision.

[0016] In combination with the first aspect, in a possible implementation, the M construction progress labels also include a second priority of the construction progress, and the second priority is less than the first priority. The method also includes: calculating the ratio of the second number of votes to the first number of votes, the second number of votes being the number of labels corresponding to the construction progress of the second priority counted; if the ratio is greater than or equal to a threshold, determining the target progress to be the construction progress corresponding to the second priority; if the ratio is less than the threshold, determining the target progress to be the construction progress corresponding to the first priority.

[0017] The method provided in this aspect determines the progress of the target by including the first priority and the second priority of the construction progress in the M construction progress tags, and then calculating the ratio of the second vote to the first vote and the size relationship of the threshold. By comprehensively considering multiple factors, including the number of votes, ratio, priority, etc., the accuracy of determining the target progress can be improved, the decision risk can be reduced, and the decision quality can be improved.

[0018] On the other hand, an embodiment of the present invention discloses a method for identifying the construction progress of a construction site. The method can be applied to a processor CPU or a processing chip. The method includes: receiving video image data shot on the construction site when a user passes through a target path; generating video data based on the video image data; uploading the video data to a server so that the server can identify the progress of the target area of ​​the construction site based on the video data.

[0019] The method provided by the present invention generates video data based on the video image data taken by the user, and uploads the video data to the server so that the server can identify the progress of the target area of ​​the construction site based on the video data. The user can upload the video image data to the server and let other users or team members share the data for collaborative work, communication and decision-making. At the same time, by uploading the video data to the server, the user can remotely monitor and manage the construction progress of the construction site and discover and solve problems in a timely manner.

[0020] In a second aspect, an embodiment of the present invention discloses a construction progress identification device for a construction site, the device comprising:

[0021] An acquisition module is used to acquire video data uploaded from a panoramic inspection system, where the video data is video image data shot on a construction site when a user passes through a target path;

[0022] A synthesis module, used to perform image synthesis processing on the video data to obtain a first key point set and a panoramic image corresponding to each key point, wherein the first key point set includes at least one key point;

[0023] A calculation module, used for selecting N key points in the identification area from the first key point set according to the identification area pre-marked in the project drawing file to form a second key point set;

[0024] A processing module, used to reduce the viewing angle range of the panoramic image according to the angle between the target object and each of the N key points, and generate images of M target angles corresponding to the N key points, M ≥ N;

[0025] A recognition module is used to use a neural network model to identify the construction progress of images at M target angles to obtain M project construction progress labels;

[0026] The voting module is used to determine the target progress among M construction progress tags based on the voting mechanism, and send the identification result of the target progress to the panoramic inspection system.

[0027] In a third aspect, an embodiment of the present invention further discloses an electronic device, comprising a processor and a memory, wherein the memory is coupled to the processor; computer-readable program instructions are stored on the memory, and when the instructions are executed by the processor, a method for identifying the construction progress of a construction site according to the first aspect or any implementation method of the first aspect is implemented.

[0028] In a fourth aspect, an embodiment of the present invention further discloses a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, a method for identifying the construction progress of a construction site as in the first aspect or any embodiment of the first aspect is implemented. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0030] Figure 1 It is a flow chart of a construction progress identification method of a construction site provided by an embodiment of the present invention;

[0031] Figure 2 It is a schematic diagram of key points of a construction progress identification method for a construction site provided by an embodiment of the present invention;

[0032] Figure 3 It is a schematic diagram of an identification area of ​​a construction progress identification method of a construction site provided by an embodiment of the present invention;

[0033] Figure 4 is a flow chart of another construction progress identification method for a construction site provided by an embodiment of the present invention;

[0034] Figure 5 It is a schematic diagram of an identification area of ​​a construction progress identification method of a construction site provided by an embodiment of the present invention;

[0035] Figure 6 is a flow chart of another construction progress identification method for a construction site provided by an embodiment of the present invention;

[0036] Figure 7 It is a flowchart of another construction progress identification method of a construction site provided by an embodiment of the present invention;

[0037] Figure 8 It is a flowchart of another construction progress identification method of a construction site provided by an embodiment of the present invention;

[0038] Fig. 9 It is a result schematic diagram of a construction progress identification method for a construction site provided by an embodiment of the present invention;

[0039] Fig.10 It is a panoramic image schematic diagram of a construction progress identification method of a construction site provided by an embodiment of the present invention;

[0040] Fig.11 It is a structural block diagram of a construction progress identification and processing device for a construction site provided by an embodiment of the present invention;

[0041] Fig.12It is a schematic diagram of the hardware structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0042] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0043] The embodiment of the present invention proposes a technical solution for a construction progress identification method of a construction site, which is used to solve the problem that traditional engineering project management methods rely more on manual analysis and judgment, require managers to have rich industry experience, and are difficult to achieve standardized and process-based engineering progress control.

[0044] According to an embodiment of the present invention, an embodiment of a method for identifying the construction progress of a construction site is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0045] In this embodiment, a construction progress identification method for a construction site is provided, which can be used for the above-mentioned mobile terminal, such as a PC, a tablet computer, etc. Figure 1 is a flow chart of a construction progress identification method for a construction site according to an embodiment of the present invention. Figure 1 As shown, the process includes the following steps:

[0046] Step 101: Acquire video data uploaded from a panoramic inspection system, where the video data is video image data shot on a construction site when a user passes through a target path.

[0047] The method provided in this embodiment obtains the original video data of the target building, and the original video data is a panoramic video. The staff wearing a panoramic camera helmet needs to shoot according to the preset planned path and upload the shooting content to the cloud system platform.

[0048] Step 102: performing image synthesis processing on the video data to obtain a first key point set and a panoramic image corresponding to each key point, wherein the first key point set includes at least one key point.

[0049] A panoramic image is a fixed observation point that can provide free browsing of 360° horizontal azimuth and 180° vertically. Panoramic images use wide-angle expression methods and forms such as paintings, photos, videos, and three-dimensional models to show as much of the surrounding environment as possible. It is also called a 360° panorama, which means that by using a professional camera to capture the image information of the entire scene or using modeling software to render the picture, using software to stitch the pictures, and playing them with a special player, a flat photo or computer modeling picture is turned into a 360° panorama for virtual reality browsing, simulating a two-dimensional plane map into a real three-dimensional space and presenting it to the viewer.

[0050] Key points are also called interest points. Key points on a panoramic image refer to areas or points with obvious features in the image. These key points can provide geometric and texture information of the image. In the process of panoramic image stitching, these key points are usually used to find similar areas between images and perform matching and stitching. Figure 2 As shown in Figure 1, it is a stable and distinctive point set on a 2D image, 3D point cloud or surface model that can be obtained through detection criteria. Technically, the number of key points is much smaller than the amount of data in the original point cloud or image, and it is combined with the local feature descriptor to form a key point descriptor. It is often used to form a compact representation of the original data, which is representative and descriptive, thereby speeding up the subsequent recognition, tracking and other data processing speeds.

[0051] Specifically, image synthesis processing is performed on the original video data to generate a panoramic image and a set of key points on the panoramic image.

[0052] Step 103: According to the identification area pre-marked in the project drawing file, N key points in the identification area are selected from the first key point set to form a second key point set.

[0053] The identification area refers to the area identified by the staff using the circle marking algorithm. The inspection trajectory points contained in the marked area are extracted, calculated, and the target objects are screenshotted. Then, the key points of the path contained in the algorithm calibration area are extracted.

[0054] The method provided in this embodiment sequentially calculates the key points in the first key point set within the recognition area.

[0055] For example, in a construction site, the northern part of the construction site is the identification area, that is, the construction status of the northern part of the construction site needs to be known. The first key point set is all the key points of the entire construction site. The second key point set is the key points of the northern part of the construction site. The first key point set is screened by the identification area, and the key points needed to be used in the subsequent steps are selected, which is the second key point set.

[0056] Step 104: narrowing the viewing angle of the panoramic image according to the angle between the target object and each of the N key points, and generating images of M target angles corresponding to the N key points, M≥N.

[0057] The method provided in this embodiment generates a panoramic image corresponding to a set of key points through multiple angles of the target object in the panoramic image.

[0058] For example, Figure 3 As shown, the three circles represent three recognition ranges, and the point in the middle of the circle represents the center point of the recognition range. It is necessary to first know the target object in the center of the circle, take the target object as the center, and use the distance of the surrounding key points as the radius to find the key points on the edges of multiple circles. Through the key points on the edges of the circles, a panoramic image of the target object is generated.

[0059] Step 105: Use a neural network model to identify the construction progress of the images of the M target angles to obtain M engineering construction progress labels.

[0060] The method provided in this embodiment obtains a neural network model; inputs images of N target angles into the neural network model; searches for the engineering construction progress corresponding to each target angle image based on the association relationship; generates the corresponding engineering construction progress label; and finally obtains N engineering progress labels, and the neural network model can automatically recognize the input image.

[0061] Step 106: Determine the target progress among the M construction progress tags based on a voting mechanism, and send the identification result of the target progress to the panoramic inspection system.

[0062] The method provided in this embodiment determines the target progress by obtaining N construction progress labels and a voting algorithm. In the voting process, the priorities of different construction progress are taken into account, which can reflect the conditions of different construction progress, making the determination of the target progress more reasonable. The voting mechanism is an objective decision-making method, which can avoid the influence of subjective factors on the determination of the target progress and improve the objectivity and fairness of the decision.

[0063] Optionally, in another embodiment, see Figure 4 The above step 103 specifically includes:

[0064] Step 401: Obtain the coordinates of each key point in the first key point set.

[0065] The method provided in this embodiment obtains drawings of a target construction site, wherein the drawings are architectural CAD (Computer Aided Design, CAD) drawings, such as Figure 5A CAD drawing of a target construction site is shown. The coordinates of all key points in the drawing are obtained.

[0066] CAD drawings refer to CAD construction drawings, which are drawings made using AutoCAD software to combine the overall layout of the project, the building's exterior shape, interior layout, structural construction, interior and exterior decoration, material methods, equipment, and construction. CAD construction drawings are characterized by complete drawings, accurate expressions, and specific requirements. They are the basis for engineering construction, preparation of construction drawing budgets, and construction organization design, and are also important technical documents for technical management.

[0067] Step 402, obtaining the marked identification area.

[0068] The method provided in this embodiment plans a path for taking a panoramic image of the target construction site based on the drawings of the construction site. Figure 5 As shown in the figure, the gray circle in the middle is the identification area, which is also the target building area for which the project progress needs to be obtained. For example, if a user wants to obtain the project progress of an area in a drawing, he needs to circle the target building area based on the CAD drawing of the construction site and mark it with a special color. The identification area can be one or more, such as Figure 3 As shown, Figure 3 There are three circles in it, representing three different identification areas.

[0069] Step 403, based on the marked recognition area, filter the key points in the first key point set to obtain a filtered second key point set, wherein the second key point set includes N key points, and the coordinates of the N key points are all within the range of the recognition area.

[0070] In this embodiment, based on the key points of the engineering building and the range of the target recognition area, a second key point set belonging to the target recognition area in the first key point set is calculated, and the second key point set includes N key points, which provide help for the subsequent processing of the panoramic image of the target recognition area. Figure 5 As shown, for example, the first key point set is composed of all the key points on the entire drawing. The user confirms that the target recognition area is the circle in the middle. According to the range of the circle, some key points within the range are obtained, which are the second key point set.

[0071] Key point screening refers to selecting points with important information or features for extraction and analysis when processing image or video data. In the field of computer vision and image processing, key point screening is an important task that can help us effectively extract important features from images or videos, thereby improving the accuracy and efficiency of tasks such as target detection, tracking, and recognition.

[0072] Optionally, in another embodiment, see Figure 6 The above step 104 specifically includes:

[0073] Step 601: Acquire the center point of the recognition area and a first panoramic image of the recognition area.

[0074] In this implementation, based on the key points of the engineering building and the range of the target identification area, the center point of the identification area is obtained, such as Figure 3 As shown in the figure, the point at the center of the circle is the center point of the recognition area. The image is recognized, and then the panoramic image is generated by extracting the image key points using the adapted SLAM model.

[0075] The technology for performing image synthesis processing on the raw video data is a panoramic synthesis technology based on visual SLAM (Simultaneous Localization and Mapping, SLAM). SLAM refers to simultaneous positioning and map construction, which is a technology for robots to autonomously explore the environment and build maps. SLAM technology can help robots obtain information about the surrounding environment through sensors in unknown environments, so as to gradually build a map of the environment during the movement of the robot, and improve the autonomy of the robot through understanding and perception of the environment. SLAM technology can be applied to various fields, such as unmanned driving, drones, service robots, etc. Through SLAM technology, robots can self-navigate, avoid obstacles, plan paths, etc. in unknown environments, thereby realizing functions such as autonomous exploration, target tracking, and environmental monitoring. At the same time, SLAM technology can also help robots better adapt to different environments and tasks, and improve the intelligence and autonomy of robots. To realize SLAM, multiple sensors and algorithms are needed, such as sensors such as laser radars and cameras, as well as algorithms such as probability estimation, machine learning, and computer vision. SLAM technology can be divided into 2D and 3D. 2D SLAM is mainly used in planar environments, while 3D SLAM can be applied to more complex environments, such as outdoor scenes, high-rise buildings, etc.

[0076] Step 602: Based on the center point of the recognition area and the panoramic first image, generate angles of N key points in the second key point set facing the center point.

[0077] The method provided in this embodiment obtains the angle of the second key point facing the center point according to the center point of the recognition area and the panoramic first image. Figure 3 As shown, multiple key points around the circle of the identification area.

[0078] For example, Figure 3As shown, the circular area is the recognition area, the black dot is the second key point, the center point of the recognition area is the target object, the white dot is the path, and the angle is half of the angle formed by the two black dots and the center point.

[0079] Step 603, based on the angles of the N key points facing the center point, narrow the visual range to obtain M second images of the N key points facing the center point, where the M second images are images of M target angles corresponding to the N key points.

[0080] The method provided in this embodiment first determines a center point, and then adjusts the visual range based on the center point to narrow the visual range to N key points, which are some points on the edge of the recognition area. For each key point, the viewing angle of the image is adjusted according to the angle at which they face the center point, thereby obtaining M second images.

[0081] For example, Figure 3 As shown, one of the key points around the circle faces the angle formed by the center of the circle and takes a screenshot perpendicular to the angle of the center of the circle. The key points are respectively rotated 15° clockwise and 15° counterclockwise along the recognition area for screenshots, thereby obtaining images of three target angles.

[0082] Optionally, in another embodiment, see Figure 7 The above step 105 specifically includes:

[0083] Step 701, obtaining a neural network model, wherein the neural network model is obtained by training project progress data, and the neural network model includes a correlation relationship between a project progress image and a construction progress.

[0084] The method provided in this embodiment obtains a neural network model through training, so that it can learn and simulate the correlation between the project progress image and the engineering construction progress.

[0085] Neural network models can be used for various machine learning tasks, such as classification, regression, clustering, etc., and have shown good performance in processing complex data, pattern recognition, natural language processing, etc. Common neural network models include multi-layer perceptron, convolutional neural network, recurrent neural network, etc.

[0086] Step 702: input the images of the M target angles into the neural network model, search for the engineering construction progress corresponding to each target angle image according to the association relationship, and generate the corresponding engineering construction progress label to obtain the M engineering progress labels.

[0087] The method provided in this embodiment inputs images of M target angles into a neural network model based on ResNet, and then searches for the engineering construction progress corresponding to each target angle image according to the association relationship stored in the neural network model. This association relationship may be learned in advance through training data, or it may be manually set according to actual project progress data. Finally, for each target angle image, a corresponding engineering construction progress label is generated according to its corresponding engineering construction progress. These labels can be simple text labels or more complex numerical labels, depending on the needs of the project and the design of the neural network model. In this way, the user obtains M engineering progress labels, each label corresponding to an image of a target angle. These labels can be used for further analysis, evaluation or decision-making, such as evaluating the progress of engineering construction, predicting future engineering progress, etc.

[0088] ResNet (ImageNet Large Scale Visual Recognition Challenge, ResNet) refers to a deep neural network proposed by He Kaiming and others in 2015, and won the first place in the classification task in the ImageNet competition that year. The main feature of ResNet is the ultra-deep network structure. By introducing residual blocks, the gradient vanishing problem in deep neural networks is solved. When traditional neural networks are deepened, the optimization effect will gradually decrease, and even the problems of gradient explosion and gradient vanishing may occur. In order to solve this problem, ResNet introduced residual blocks, which skip connections in certain layers and weaken the strong connection between each layer, thereby solving the gradient vanishing problem.

[0089] Optionally, in another embodiment, see Figure 8 The above step 106 specifically includes:

[0090] Step 801, obtaining the M construction progress tags, wherein the M construction progress tags include a first priority of the construction progress.

[0091] The method provided in this embodiment obtains M construction progress labels through ResNet, and then determines the first priority of the construction progress, where the first priority is the latest construction progress.

[0092] Step 802, counting multiple votes corresponding to the M construction progress labels, and obtaining a first vote from the multiple votes, where the first vote is the number of labels corresponding to the construction progress of the same priority.

[0093] In the method provided in this embodiment, for example, if project progress A obtains the most votes and is the latest progress of the project and has the first priority, the progress of the target area is project progress A.

[0094] Step 803: If the construction progress priority to which the first number of votes belongs is the first priority, determining the target progress to be the construction progress corresponding to the first priority.

[0095] In the method provided in this embodiment, for example, if project progress A obtains the most votes and project progress A is not the latest progress of the project and is the second priority, the progress of the target area needs to be judged based on parameters and thresholds.

[0096] Specifically include:

[0097] The ratio of the second number of votes to the first number of votes is calculated, where the second number of votes is the number of labels corresponding to the construction progress of the second priority.

[0098] If the ratio is greater than or equal to a threshold, the target progress is determined to be the construction progress corresponding to the second priority.

[0099] If the ratio is less than the threshold, the target progress is determined to be the construction progress corresponding to the first priority.

[0100] The method provided in this embodiment needs to calculate the second number of votes, and the ratio of the picture with the second number of votes to the first number of votes. If the ratio is greater than or equal to the threshold, the target progress is determined to be the construction progress corresponding to the second priority. If the ratio is less than the threshold, the target progress is determined to be the construction progress corresponding to the first priority. For example, the threshold is set to 0.6, the first number of votes is 100 votes, and the first number of votes is the second priority, and the progress of the first number of votes is not the latest progress. If the second number of votes is 70 votes, the parameter is 0.7, and the parameter is greater than the threshold value of 0.6, then the progress of the target identification area is determined to be the progress of the second number of votes. On the contrary, if the second number of votes is 50 votes, the parameter is 0.5, which is less than the threshold value of 0.6, then the progress of the target identification area is determined to be the progress of the first number of votes.

[0101] The logic of voting is to first get the highest and second highest votes based on the known project progress. For example, taking the construction sequence of a building as an example, the project progress is: foundation pit excavation, basement construction, and main body construction. If the highest vote progress is the latest progress, the final progress result is the highest vote progress. If the second highest vote progress is the latest progress, calculate the second highest vote divided by the highest vote. If the second highest vote divided by the highest vote is greater than 0.5, the final progress result is the second highest vote progress, otherwise it is the highest vote progress.

[0102] Specific as Fig. 9 The following is a display of the construction progress identification results of a construction site, where the construction progress of the current target building is displayed under the progress node. For example, the first row of progress is the basement construction, and the progress value is 100%. The second row is the indoor and outdoor decoration, and the progress value is also 100%.

[0103] Optionally, in another embodiment, a construction progress identification method for a construction site is characterized in that the method comprises:

[0104] Receive video image data captured on the construction site when the user passes through the target path.

[0105] The method provided in this embodiment requires the staff to wear a panoramic camera helmet and shoot according to the planned travel path. For example, an engineer wears a panoramic camera helmet, turns on the recording function of the panoramic camera, and then the engineer patrols around the planned target area to obtain the original video data of the target area. A panoramic camera helmet refers to a device that integrates a panoramic camera on a helmet. This helmet usually has a high-definition panoramic shooting function and can take 360-degree photos and videos. The main advantage of a panoramic camera helmet is that it is easy to carry and operate, and can provide a high-definition panoramic view, which is suitable for occasions where wide-angle scenes need to be shot, such as construction sites, road traffic, tourist attractions, etc.

[0106] Video data is generated according to the video image data.

[0107] The video data is uploaded to a server so that the server can identify the progress of a target area of ​​the construction site based on the video data.

[0108] The method provided in this embodiment, such as Fig.10 As shown, it is a 360° panoramic image. The raw video data obtained through the steps is uploaded to the corresponding system platform through the cloud. For example, after the engineer completes the recording of the raw video data, the engineer uploads the raw video data to the system platform. The engineer also needs to upload the project CAD drawings and planned CAD drawings to the system platform.

[0109] The technical method for identifying the construction progress of a construction site provided in this embodiment has the following beneficial effects on construction project management:

[0110] 1. Realize automatic identification of construction progress at construction sites. Compared with manual analysis and judgment, this method saves construction progress analysis time.

[0111] 2. This method makes the identification of the construction progress of the building no longer rely on the analysis and judgment of experienced staff, thus saving the company's manpower costs.

[0112] In this embodiment, a device for identifying the construction progress of a construction site is also provided, and the device is used to implement the construction progress identification method of the construction site in the above embodiment, which has been described and will not be repeated. As used below, the term "unit" or "module" can be a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceived.

[0113] This embodiment provides a device for identifying the construction progress of a construction site. Fig.11 As shown, the device includes: an acquisition module 1011, a synthesis module 1012, a calculation module 1013, a processing module 1014, an identification module 1015, and a voting module 1016. In addition, the device also includes other more or fewer units / modules, such as a storage module, etc., which is not limited in this embodiment.

[0114] The acquisition module 1011 is used to acquire video data uploaded from the panoramic inspection system, wherein the video data is video image data shot on the construction site when the user passes through the target path.

[0115] The synthesis module 1012 is used to perform image synthesis processing on the video data to obtain a first key point set and a panoramic image corresponding to each key point, wherein the first key point set includes at least one key point.

[0116] The calculation module 1013 is used to select N key points in the identification area from the first key point set according to the identification area pre-marked in the project drawing file to form a second key point set.

[0117] The processing module 1014 is used to reduce the viewing angle range of the panoramic image according to the angle between the target object and each of the N key points, and generate images of M target angles corresponding to the N key points, M≥N.

[0118] The recognition module 1015 is used to use a neural network model to perform construction progress recognition on the images of the M target angles to obtain M engineering construction progress labels.

[0119] The voting module 1016 is used to determine the target progress among the M construction progress tags based on a voting mechanism, and send the identification result of the target progress to the panoramic inspection system.

[0120] In some optional embodiments, the acquisition module 1011 is specifically used to obtain the coordinates of each key point in the first key point set; obtain the marked identification area; based on the marked identification area, filter the key points in the first key point set to obtain a filtered second key point set, wherein the second key point set contains N key points, and the coordinates of the N key points are all within the range of the identification area.

[0121] In other optional embodiments, the synthesis module 1012 is specifically used to obtain the center point of the recognition area and the panoramic first image of the recognition area; based on the center point of the recognition area and the panoramic first image, generate the angles of N key points in the second key point set facing the center point; based on the angles of the N key points facing the center point, narrow the visual range to obtain M second images of the N key points facing the center point, and the M second images are images of M target angles corresponding to the N key points.

[0122] In other optional embodiments, the identification module 1015 is specifically used to obtain a neural network model, which is obtained by training project progress data, and the neural network model contains the correlation between the project progress image and the engineering construction progress; the images of the M target angles are input into the neural network model, the engineering construction progress corresponding to each target angle image is searched according to the correlation, and the corresponding engineering construction progress label is generated to obtain the M engineering progress labels.

[0123] In some other optional implementations, the voting module 1016 is specifically used to obtain the M construction progress labels, wherein the M construction progress labels include the first priority of the construction progress; count the multiple votes corresponding to the M construction progress labels, and obtain the first vote from the multiple votes, wherein the first vote is the number of labels corresponding to the construction progress of the same priority; if the construction progress priority to which the first vote belongs is the first priority, then determine that the target progress is the construction progress corresponding to the first priority. The embodiment of the present invention also provides a computer device having the above Fig.10 A construction progress identification device for a construction site is shown.

[0124] In other optional embodiments, the voting module 1016 is specifically used to calculate the ratio of the second number of votes to the first number of votes, where the second number of votes is the number of labels corresponding to the construction progress of the second priority. If the ratio is greater than or equal to a threshold, the target progress is determined to be the construction progress corresponding to the second priority. If the ratio is less than the threshold, the target progress is determined to be the construction progress corresponding to the first priority.

[0125] See also Fig.12 , Fig.12 is a schematic diagram of the structure of a computer device provided by an optional embodiment of the present invention, such as Fig.12 As shown, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including a high-speed interface and a low-speed interface. Each component utilizes different buses to communicate with each other, and can be installed on a common mainboard or installed in other ways as required. The processor can process the instructions executed in the computer device, including instructions stored in the memory or on the memory to display the graphical information of the GUI on an external input / output device. In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Equally, multiple computer devices can be connected, and each device provides part of the necessary operation. Fig.12 A processor 10 is taken as an example.

[0126] The processor 10 may be a central processing unit, a network processor or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic or any combination thereof.

[0127] The memory 20 stores instructions executable by at least one processor 10, so that the at least one processor 10 executes the method shown in the above embodiment.

[0128] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created by the use of a computer device based on the presentation of a small program landing page, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely arranged relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0129] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid state drive; the memory 20 may also include a combination of the above types of memory.

[0130] The computer device also includes an input device 30 and an output device 40. The processor 10, the memory 20, the input device 30 and the output device 40 may be connected via a bus or other means. Fig.12 The example of connecting through bus is taken in the following.

[0131] The input device 30 can receive input digital or character information, and generate key signal input related to the user settings and function control of the computer device, such as a touch screen, a keypad, a mouse, a track pad, a touch pad, an indicator bar, one or more mouse buttons, a trackball, a joystick, etc. The output device 40 may include a display device, an auxiliary lighting device, a tactile feedback device, etc. The above-mentioned display device includes but is not limited to a liquid crystal display, a light emitting diode, a display and a plasma display. In some optional embodiments, the display device can be a touch screen.

[0132] The embodiment of the present invention also provides a computer-readable storage medium. The method according to the embodiment of the present invention can be implemented in hardware, firmware, or can be implemented as a computer code that can be recorded in a storage medium, or can be implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and will be stored in a local storage medium through a network download, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state hard disk, etc.; further, the storage medium can also include a combination of the above types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor, or hardware, the method shown in the above embodiment is implemented.

[0133] Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A construction progress identification method for a construction site, It is characterized in that The method comprises: Acquire video data uploaded from a panoramic inspection system, where the video data is video image data shot on a construction site when a user passes through a target path; Performing image synthesis processing on the video data to obtain a first key point set and a panoramic image corresponding to each key point, wherein the first key point set includes at least one key point; According to the identification area pre-marked in the project drawing file, N key points in the identification area are selected from the first key point set to form a second key point set; According to the angle between the target object and each of the N key points, the viewing angle range of the panoramic image is reduced to generate images of M target angles corresponding to the N key points, M ≥ N; Using a neural network model to identify the construction progress of the images of the M target angles, and obtaining M engineering construction progress labels; The target progress is determined among the M construction progress tags based on a voting mechanism, and the identification result of the target progress is sent to the panoramic inspection system.

2. The method according to claim 1, It is characterized in that Selecting N key points in the identification area from the first key point set to form a second key point set includes: Obtaining the coordinates of each key point in the first key point set; Acquire the identified area of ​​the mark; Based on the marked recognition area, key points in the first key point set are filtered to obtain a filtered second key point set, wherein the second key point set includes N key points, and the coordinates of the N key points are all within the range of the recognition area.

3. The method according to claim 1, It is characterized in that The method narrows the viewing angle of the panoramic image according to the angle between the target object and each of the N key points, and generates images of M target angles corresponding to the N key points, M≥N, which includes: Acquire a center point of the identified area and a first panoramic image of the identified area; Based on the center point of the recognition area and the panoramic first image, generating angles of N key points in the second key point set facing the center point; Based on the angles of the N key points facing the center point, the visual range is narrowed to obtain M second images of the N key points facing the center point, and the M second images are images of M target angles corresponding to the N key points.

4. The method according to claim 1, It is characterized in that The neural network model is used to identify the construction progress of the images of the M target angles to obtain M engineering construction progress labels, including: Obtaining a neural network model, wherein the neural network model is obtained by training project progress data, and the neural network model contains a correlation relationship between the project progress image and the engineering construction progress; The images of the M target angles are input into the neural network model, the engineering construction progress corresponding to each target angle image is searched according to the association relationship, and the corresponding engineering construction progress labels are generated to obtain the M engineering progress labels.

5. The method according to claim 1, It is characterized in that The determining the target progress among the M construction progress tags based on the voting mechanism includes: Obtain the M construction progress tags, wherein the M construction progress tags include a first priority of the construction progress; Counting a plurality of votes corresponding to the M construction progress labels, and obtaining a first vote from the plurality of votes, where the first vote is the number of labels corresponding to the construction progress of the same priority; If the construction progress priority to which the first number of votes belongs is the first priority, the target progress is determined to be the construction progress corresponding to the first priority.

6. The method according to claim 5, It is characterized in that The M construction progress tags also include a second priority of the construction progress, the second priority is lower than the first priority, and the method further includes: Calculate the ratio of the second number of votes to the first number of votes, where the second number of votes is the number of labels corresponding to the construction progress of the second priority. If the ratio is greater than or equal to a threshold, determining the target progress to be the construction progress corresponding to the second priority; If the ratio is less than the threshold, the target progress is determined to be the construction progress corresponding to the first priority.

7. A construction progress identification device for a construction site, It is characterized in that The device comprises: An acquisition module, used to acquire video data uploaded from a panoramic inspection system, wherein the video data is video image data shot on a construction site when a user passes through a target path; A synthesis module, configured to perform image synthesis processing on the video data to obtain a first key point set and a panoramic image corresponding to each key point, wherein the first key point set includes at least one key point; A calculation module, configured to select N key points in the identification area from the first key point set according to the identification area pre-marked in the project drawing file, to form a second key point set; A processing module, used to reduce the viewing angle range of the panoramic image according to the angle between the target object and each of the N key points, and generate images of M target angles corresponding to the N key points, M≥N; A recognition module, used to use a neural network model to perform construction progress recognition on the images of the M target angles to obtain M engineering construction progress labels; A voting module is used to determine the target progress among the M construction progress tags based on a voting mechanism, and send the identification result of the target progress to the panoramic inspection system.

8. A construction progress identification system for a construction site, It is characterized in that Said includes: panoramic inspection system and server, The panoramic inspection system is used to receive video image data taken by a user on a construction site when the user passes through a target path; generate video data based on the video image data, and upload the video data to a server; The server is used to execute the construction progress identification method for a construction site as described in any one of claims 1 to 6.

9. An electronic device, It is characterized in that comprising a processor and a memory, the memory being coupled to the processor; The memory stores computer-readable program instructions, and when the instructions are executed by the processor, the construction progress identification method for a construction site as described in any one of claims 1 to 6 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, It is characterized in that When the computer program is executed by a processor, the construction progress identification method for a construction site according to any one of claims 1 to 6 is implemented.

Images