CN113657925A - Artificial intelligence-based civil engineering cost management method - Google Patents

Abstract

The invention relates to the technical field of artificial intelligence, in particular to a civil engineering cost management method based on artificial intelligence. The method obtains the construction area through the panoramic top-down image of the construction scene when it is not under construction, and obtains the initial simulated heat map of the construction area; obtains the actual heat map of the actual panoramic thermal image collected in real time; The distance between the first pixel points corresponding to the thermal value determines the construction situation. When it is confirmed that the construction is not normal, the project cost is adjusted according to the efficacy deviation and position deviation. Only based on the simulated heat map and the actual heat map of the construction area, a fast and effective construction abnormality analysis can be carried out, and then the project cost can be accurately adjusted through the obtained abnormal information, which reduces the error of the project cost adjustment.

Description

Artificial intelligence-based civil engineering cost management method

technical field

The invention relates to the technical field of artificial intelligence, in particular to a civil engineering cost management method based on artificial intelligence.

Background technique

The problem of project cost is that it emphasizes design and neglects management. For cost management, the existing technology usually monitors potential abnormal sources in real time during the construction process, so as to trace the cause of the deviation of the project progress, and then adjust the project cost according to the deviation.

In practice, the inventor finds that the above-mentioned prior art has the following defects: multiple monitoring ports are required for real-time monitoring of potential abnormal sources, which will lead to increased costs, and it is difficult to directly perceive some abnormal situations through monitoring methods, and the monitoring ports are relatively restricted. , it is difficult to trace the source of the abnormal situation of personnel in the construction process without using the fixed-point monitoring method, and it is impossible to accurately adjust the cost.

SUMMARY OF THE INVENTION

In order to solve the above-mentioned technical problems, the purpose of the present invention is to provide a kind of artificial intelligence-based civil engineering cost management method, and the adopted technical scheme is as follows:

An embodiment of the present invention provides an artificial intelligence-based civil engineering cost management method, which includes the following specific steps:

Obtain a panoramic top view image of the construction scene when it is not under construction to obtain the construction area; take the construction start position as the center, construct an ellipse area according to the center of gravity and shape direction of the construction area, and divide the part of the ellipse area that includes the construction area As the to-be-treated area, obtain an initial simulated heat map of the to-be-treated area;

Collect the actual panoramic overhead image of the construction site in real time, and obtain the actual thermal map of the actual panoramic overhead image; respectively obtain the initial simulated thermal map and the first pixel point corresponding to the maximum thermal value in the actual thermal map, by The distance between the first pixel points determines the construction situation;

When it is confirmed that the construction is abnormal, the simulated heat map corresponding to the distance equal to the distance threshold is obtained, and the efficacy deviation is obtained according to the construction efficacy of the simulated heat map and the initial simulated heat map; if the efficacy deviation cannot be obtained, Obtain the position deviation according to the abnormal pixel point in the simulated heat map;

The engineering cost is adjusted according to the efficacy deviation and the position deviation.

Preferably, the method for obtaining the initial simulated heat map of the to-be-treated area includes:

Constructing a covariance matrix from the major and minor axes of the ellipse region and the shape direction;

A two-dimensional Gaussian probability density function of the spatial dimension is constructed with the covariance matrix and the center, and the first thermal value of each pixel in the construction area is obtained according to the two-dimensional Gaussian probability density function to obtain a first simulation. heatmap;

A second simulated heat map of the to-be-treated area is obtained according to the first simulated heat map.

Preferably, after the second simulated heat map is obtained, an optimization method for the second simulated heat map includes:

Based on the second simulated heat map, a second pixel point in the to-be-processed area that does not belong to the construction area is acquired, and the pixel point in the construction area is optimized according to the first thermal value of the second pixel point the first thermal value of the pixel point to obtain a third simulated thermal map;

A one-dimensional Gaussian distribution function of the time dimension is constructed from the sampling time of the actual panoramic bird's-eye view image, the construction efficiency, the number of construction workers and the second thermal value of the third simulated thermal map;

The second thermal value of each pixel in the third simulated heat map is updated by using the one-dimensional Gaussian distribution function to obtain an initial fourth simulated heat map.

Preferably, the method for using the one-dimensional Gaussian distribution function to update the second thermal value of each pixel in the third simulated heat map to obtain an initial fourth simulated heat map includes:

Based on the initial sampling time of the actual panoramic image and each time point between the sampling time, the one-dimensional Gaussian distribution function is used to update the second thermal value of each pixel, so that each time point can be obtained The corresponding update simulation heat map;

Based on the updated simulated heat map, thermal superposition is performed on the updated second thermal values of the pixels corresponding to each time point to obtain the initial fourth simulated heat map corresponding to the sampling time.

Preferably, the modification method of the covariance matrix includes:

The construction trajectory type of the construction area is acquired, and the covariance matrix is modified according to the construction trajectory type.

Preferably, the method for judging the construction situation by the distance between the first pixel points includes:

The Euclidean distance between the first pixel points is acquired, and when the Euclidean distance is less than or equal to the distance threshold, it is determined that the construction is normal; otherwise, it is determined that the construction is abnormal.

Preferably, when both the efficacy deviation and the position deviation cannot be obtained, it is determined that both the efficacy deviation and the position deviation exist in the construction area.

Preferably, the method for obtaining the center of gravity of the construction area includes:

The first-order moment of the construction area is acquired, and the first-order moment is used as the center of gravity.

Preferably, the method for obtaining the shape direction of the construction area includes:

Obtain the second-order moment of the construction area, and use the second-order moment as the shape direction.

The embodiments of the present invention have at least the following beneficial effects: fast and effective construction abnormality analysis can be carried out only according to the simulated heat map and the actual heat map of the construction area, and then the project cost can be accurately adjusted through the acquired abnormal information, which reduces the adjustment of the project cost error.

Description of drawings

In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or in the prior art, the following briefly introduces the accompanying drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description The drawings are only some embodiments of the present invention, and for those of ordinary skill in the art, other drawings can also be obtained based on these drawings without any creative effort.

Fig. 1 is the step flow chart of a kind of artificial intelligence-based civil engineering cost management method provided by an embodiment of the present invention;

Detailed ways

In order to further illustrate the technical means and effects adopted by the present invention to achieve the predetermined purpose of the invention, the following describes an artificial intelligence-based civil engineering cost management method proposed by the present invention with reference to the accompanying drawings and preferred embodiments, and its specific implementation The mode, structure, features and their functions are described in detail as follows. In the following description, different "one embodiment" or "another embodiment" are not necessarily referring to the same embodiment. Furthermore, the particular features, structures, or characteristics in one or more embodiments may be combined in any suitable form.

Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

The specific scheme of an artificial intelligence-based civil engineering cost management method provided by the present invention will be specifically described below with reference to the accompanying drawings.

The specific scenarios targeted by the embodiments of the present invention are: the cost progress management in the civil engineering scenario, mainly analyzes the construction process management in the whole process cost management, and the actual scenario takes the single-layer scenario as the analysis object, and the single-layer scenario is as follows Single-storey floor construction, road construction, etc. can be approximated as scenes of plane construction operations; in plane scenes, RGB cameras are deployed for image acquisition. The RGB cameras are arranged at high places and the pose is fixed. Multiple RGB cameras can be arranged to pass the image. The panoramic image is obtained by stitching, and the panoramic top-down image of the flat scene can be obtained by default in subsequent processing.

Please refer to FIG. 1, which shows a flow chart of steps of an artificial intelligence-based civil engineering cost management method provided by an embodiment of the present invention, and the method specifically includes the following specific steps:

Step S001, obtaining a panoramic top view image of a construction scene when it is not under construction to obtain a construction area; taking the construction start position as the center, constructing an elliptical area according to the center of gravity and shape direction of the construction area, and taking the part of the elliptical area including the construction area as a to-be-processed area to obtain a simulated heatmap of the area to be treated.

Specifically, in the case of no construction, the RGB camera is used to collect the panoramic overhead image of the construction scene, and the obtained panoramic overhead image is divided into the construction area. Since the camera pose is fixed, it is only necessary to know the construction cost design in advance. The area can be manually divided, and the divided construction area is saved in the form of a binary image, that is, the pixel value of the pixel in the construction area is set to 1, and the pixel value of other pixels is set to 0.

Obtain cost design information, including construction number, construction efficacy, and construction starting position, where construction efficacy is the amount of work that construction personnel can complete in one day. The binary image of the construction area is processed, and the first-order moment of the construction area is obtained, and the first-order moment is taken as the center of gravity; the second-order moment of the construction area is obtained, and the second-order moment is taken as the shape direction θ. In order to ensure that the construction area can be covered as much as possible, an elliptical area is _constructed based on the construction area. y _s ) constructs an elliptical region for the center, combining the major and minor axes. Since the straight line where the short axis is located equally divides the ellipse area, the part of the ellipse area including the construction area is taken as the area to be treated.

以协方差矩阵和中心构建空间维度的二维高斯概率密度函数

根据二维高斯概率密度函数得到施工区域中每个像素点的第一热力值以得到第一模拟热力图；保留第一模拟热力图中待处理区域部分作为第二模拟热力图，将其他区域的像素点的热力值置为0。Construct a covariance matrix from the major and minor axes of the ellipse region and the shape direction, which is

Construct 2D Gaussian probability density function of spatial dimension with covariance matrix and center

Obtain the first thermal value of each pixel in the construction area according to the two-dimensional Gaussian probability density function to obtain the first simulated heat map; retain the part of the to-be-processed area in the first simulated heat map as the second simulated heat map, The thermal value of the pixel is set to 0.

Since the to-be-treated area has spatial redundancy relative to the construction area, and the position corresponding to the redundancy usually does not have construction personnel staying during the construction process, the second simulation heat map is optimized to improve the second simulation The characterization ability and accuracy of the heat map, the optimization method of the second simulated heat map is:

1) Based on the second simulated heat map, obtain the second pixel point in the area to be processed that does not belong to the construction area, and optimize the first thermal value of the pixel point in the construction area according to the first thermal value of the second pixel point to obtain the first thermal value of the pixel point in the construction area. Three simulated heatmaps.

As an example, start traversing column by column along the shape direction from the short axis included in the area to be processed, taking any column of pixels as an example, obtain all the second pixels in the column that belong to the area to be processed but not the construction area. For pixel points, the sum of the first thermal values of these pixel points is equally divided among the pixels belonging to the construction area in this column, and the first thermal value of these second pixel points is set to 0; after traversing the area to be processed, obtain The largest thermal value in the construction area and the thermal value is scaled to 1, and then the scaling ratio can be obtained. Using the scaling ratio, the thermal values of other pixels in the construction area are scaled by the same size, and then the third thermal value of the area to be processed can be obtained. Simulate a heatmap.

It should be noted that if there is any column of pixels that belong to the second pixel, the first thermal value of the pixel of this column is directly set to 0.

2) A one-dimensional Gaussian distribution function in the time dimension is constructed from the collection time of the panoramic top-down image, the construction efficiency, the number of construction workers, and the second thermal value of the third simulated thermal map.

Specifically, the formula of the one-dimensional Gaussian distribution function is:

m为施工区域的像素点数量。Among them, ε is the expected construction period based on the project cost; μ is the mean value, which is set to 0; t is the adjusted sampling time, obtained from t=t′-α(1- _hi ), where h ₁ is the third The second thermal value of the ith pixel in the simulated thermal map; t' is the sampling time of the actual panoramic bird's-eye view image; α is the adjustment coefficient obtained based on the construction efficiency S and the number of construction workers n, namely

m is the number of pixels in the construction area.

其中，

为遗忘系数，通常设置为0.95；Z′为当前时刻叠加后的热力值；Z为前一时刻的热力值；z为当前时刻的热力值。3) Based on each time point between the initial sampling time and the sampling time of the actual panoramic bird's-eye view image, the second thermal value of each pixel is updated by a one-dimensional Gaussian distribution function, and then the update simulation corresponding to each time point can be obtained. The heat map, based on the updated simulated heat map at each time point, obtains the initial fourth simulated heat map corresponding to the sampling time by thermally superimposing the updated second heat value of the pixel corresponding to each time point, The formula for thermal superposition is:

in,

is the forgetting coefficient, usually set to 0.95; Z′ is the thermal value after superposition at the current moment; Z is the thermal value at the previous moment; z is the thermal value at the current moment.

It should be noted that the initial fourth simulated heat map can represent the staying probability of construction workers at each moment.

Step S002, collect the actual panoramic top view image of the construction site in real time, and obtain the actual heat map of the actual panoramic top view image; respectively obtain the first pixel point corresponding to the maximum thermal value in the initial simulated heat map and the actual heat map, and the difference between the first pixel points is obtained. The distance between them judges the construction situation.

Specifically, the RGB camera is used to collect the actual panoramic overhead image of the construction scene in real time, and the actual panoramic overhead image is passed through the key point prediction network to obtain the corresponding actual heat map.

Preferably, in the embodiment of the present invention, the key points of the construction personnel's footsteps are used as labels.

进而利用变更后的协方差矩阵来获取初始第四模拟热力图。Since the initial fourth simulated heat map can only represent the staying probability of each construction worker when there is no actual construction track, the embodiment of the present invention corrects the initial fourth simulated heat map based on the type of the construction track corresponding to the actual heat map, then the correction method It is: in the embodiment of the present invention, the construction track types of the construction area are a priori horizontal S-shaped operation and vertical S-shaped operation, wherein the direction parallel to the shape is horizontal, and the direction perpendicular to the shape is vertical. If the construction trajectory type is horizontal, the fourth simulated heat map does not need to be corrected; if the construction trajectory type is vertical, the covariance matrix in the above step S001 is changed to

Further, the changed covariance matrix is used to obtain the initial fourth simulated heat map.

It should be noted that the type of construction trajectory can be judged by the changes of the real-time heat map.

Further, extract the first pixel points corresponding to the maximum thermal value in the initial fourth simulated heat map and the actual heat map respectively, and calculate the Euclidean distance between the first pixel points. When the Euclidean distance is less than or equal to the distance threshold, it is determined that the construction is normal. ; Otherwise, confirm that the construction is not normal.

Step S003, when it is confirmed that the construction is not normal, obtain the corresponding simulated heat map when the distance is equal to the distance threshold, and obtain the efficiency deviation according to the construction efficiency of the simulated heat map and the initial simulated heat map; if the efficiency deviation cannot be obtained, according to the simulated heat map The abnormal pixel points of the get position deviation.

Specifically, when it is confirmed that the construction is abnormal, the traceability of the one-dimensional Gaussian distribution function in the time dimension is firstly carried out to analyze the change of construction efficiency, that is, to obtain the changed construction efficiency and the corresponding change start time, then the analysis of the construction efficiency change The method is: based on the known sampling time of the actual heat map and the construction effect before the change, obtain the unknown construction effect after the change and the fourth simulated heat map corresponding to the change start time, and make the fourth simulated heat map and the actual heat map. The Euclidean distance between the first pixels corresponding to the maximum thermal value in the figure is equal to the distance threshold; since the sampling time of the known actual thermal map is fixed, the construction effect and the change start time may be the same after different changes. Therefore, two frames of actual heatmaps at different sampling times are needed to trace the source, that is, according to the known sampling times corresponding to the two different actual heatmaps and the construction efficiency before the change, the unknown after the change is obtained. The construction effect and the two fourth simulated heatmaps generated at the beginning of the change, so that the Euclidean distance between the two fourth simulated heatmaps and the first pixel corresponding to the maximum thermal value in the actual heatmap is equal to the distance threshold as a limit The condition searches for the two unknown parameters of the changed construction effect and the change start time; if the search is successful, confirm the change of the construction effect, and then use the obtained changed construction effect and the construction effect of the initial fourth simulated heat map Get the power bias.

When the efficacy deviation cannot be obtained, trace the source of the two-dimensional Gaussian probability density function of the spatial dimension to analyze the construction abnormal points, that is, to obtain the changed pixel points and the degree of change of the pixel points, the analysis method of the construction abnormal points is: Based on The sampling time of the known actual heat map and the unknown changed pixel position and degree of change are obtained to obtain the corresponding fourth simulated heat map, so that the fourth simulated heat map is equal to the first pixel corresponding to the maximum heat value in the actual heat map. The Euclidean distance between the two is equal to the distance threshold; since the sampling time of the actual heat map is fixed, the position and degree of change of the pixel points with different changes may generate the same fourth simulated heat map, so two frames of actual heat at different sampling times are required. Trace the source of the image, that is, generate two fourth simulated heatmaps according to the sampling time corresponding to the two different actual heatmaps and the changed pixel position and degree of change, so that the two fourth simulated heatmaps are the same as the actual heatmap maximum heatmap. The Euclidean distance between the first pixel points corresponding to the value is equal to the distance threshold as a constraint to search for the two unknown parameters of the changed pixel point position and the degree of change; if the search is successful, confirm the occurrence of abnormal construction points, and then get The changed pixel position and degree of change are taken as the position deviation.

It should be noted that the above method for judging efficacy deviation and position deviation is only used in the case of one efficacy change and one construction abnormal point. If the efficacy deviation or position deviation cannot be obtained, confirm that both efficacy deviation and position deviation exist in the construction area. .

Step S004, the engineering cost is adjusted according to the efficacy deviation and the position deviation.

Specifically, the obtained deviation information is directly transmitted to the cost terminal, and the cost terminal performs reasonable project cost management according to the deviation information, wherein the deviation information includes efficacy deviation and position deviation.

If it is confirmed that there are both efficacy deviation and position deviation in the construction area, it is necessary to dispatch a project coster to the construction site to adjust the project cost according to the actual situation.

It should be noted that, in the embodiment of the present invention, the construction cost is adjusted according to the abnormal situation of the construction efficiency of the construction personnel during the construction process, and the abnormal situation caused by other factors such as changes in the price of materials and environmental factors can still be used by the implementer. Existing cost management methods are adjusted.

To sum up, the embodiments of the present invention provide an artificial intelligence-based civil engineering cost management method. The method obtains a construction area through a panoramic top-down image of a construction scene before construction, and obtains an initial simulated heat map of the construction area; The actual heat map of the actual panoramic thermal image collected in real time; the construction situation is judged by the distance between the initial simulated heat map and the first pixel point corresponding to the maximum thermal value in the actual heat map. and position deviation to adjust the project cost. Only based on the simulated heat map and the actual heat map of the construction area, a fast and effective construction abnormality analysis can be carried out, and then the project cost can be accurately adjusted through the obtained abnormal information, which reduces the error of the project cost adjustment.

It should be noted that: the above-mentioned order of the embodiments of the present invention is only for description, and does not represent the advantages and disadvantages of the embodiments. And the foregoing describes specific embodiments of the present specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in an order different from that in the embodiments and still achieve desirable results. Additionally, the processes depicted in the figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Each embodiment in this specification is described in a progressive manner, and the same and similar parts between the various embodiments may be referred to each other, and each embodiment focuses on the differences from other embodiments.

The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included in the protection of the present invention. within the range.

Claims (9)

1. a civil engineering cost management method based on artificial intelligence, is characterized in that, this method may further comprise the steps: Obtain a panoramic top view image of the construction scene when it is not under construction to obtain the construction area; take the construction start position as the center, construct an ellipse area according to the center of gravity and shape direction of the construction area, and divide the part of the ellipse area that includes the construction area As the to-be-treated area, obtain the initial simulated heat map of the to-be-treated area; Collect the actual panoramic overhead image of the construction site in real time, and obtain the actual thermal map of the actual panoramic overhead image; respectively obtain the initial simulated thermal map and the first pixel point corresponding to the maximum thermal value in the actual thermal map, by The distance between the first pixel points determines the construction situation; When it is confirmed that the construction is abnormal, the simulated heat map corresponding to the distance equal to the distance threshold is obtained, and the efficacy deviation is obtained according to the construction efficacy of the simulated heat map and the initial simulated heat map; if the efficacy deviation cannot be obtained, Obtain the position deviation according to the abnormal pixel point in the simulated heat map; The engineering cost is adjusted according to the efficacy deviation and the position deviation. 2. The method according to claim 1, wherein the method for obtaining an initial simulated heat map of the region to be treated comprises: constructing the covariance matrix from the major and minor axes of the elliptical region and the shape direction; A two-dimensional Gaussian probability density function of the spatial dimension is constructed with the covariance matrix and the center, and the first thermal value of each pixel in the construction area is obtained according to the two-dimensional Gaussian probability density function to obtain a first simulation. heatmap; A second simulated heat map of the to-be-treated area is obtained according to the first simulated heat map. 3. The method according to claim 2, wherein after obtaining the second simulated heat map, an optimization method for the second simulated heat map comprises: Based on the second simulated heat map, a second pixel point in the to-be-processed area that does not belong to the construction area is acquired, and the pixel point in the construction area is optimized according to the first thermal value of the second pixel point the first thermal value of the pixel point to obtain a third simulated thermal map; A one-dimensional Gaussian distribution function of the time dimension is constructed from the sampling time of the actual panoramic bird's-eye view image, the construction efficiency, the number of construction workers and the second thermal value of the third simulated thermal map; The second thermal value of each pixel in the third simulated heat map is updated by using the one-dimensional Gaussian distribution function to obtain an initial fourth simulated heat map. 4. The method according to claim 3, wherein the second thermal value of each pixel point in the third simulated thermal map is updated by using the one-dimensional Gaussian distribution function to obtain an initial fourth thermal value. Methods of simulating heatmaps, including: Based on the initial sampling time of the actual panoramic image and each time point between the sampling time, the one-dimensional Gaussian distribution function is used to update the second thermal value of each pixel, so that each time point can be obtained The corresponding update simulation heat map; Based on the updated simulated heat map, thermal superposition is performed on the updated second thermal values of the pixels corresponding to each time point to obtain the initial fourth simulated heat map corresponding to the sampling time. 5. method according to claim 2, is characterized in that, the modification method of described covariance matrix, comprises: The construction trajectory type of the construction area is acquired, and the covariance matrix is modified according to the construction trajectory type. 6. The method according to claim 1, wherein the method for judging the construction situation by the distance between the first pixel points comprises: The Euclidean distance between the first pixel points is acquired, and when the Euclidean distance is less than or equal to the distance threshold, it is determined that the construction is normal; otherwise, it is determined that the construction is abnormal. 7 . The method according to claim 1 , wherein when both the efficacy deviation and the position deviation cannot be obtained, it is determined that both the efficacy deviation and the position deviation exist in the construction area. 8 . 8. The method according to claim 1, wherein the method for obtaining the center of gravity of the construction area comprises: Obtain the first-order moment of the construction area, and use the first-order moment as the center of gravity. 9. The method according to claim 1, wherein the method for obtaining the shape direction of the construction area comprises: Obtain the second-order moment of the construction area, and use the second-order moment as the shape direction.

Images