Monitoring method and system for building engineering construction

Abstract

The invention relates to the technical field of image processing, in particular to a monitoring method and system for building engineering construction. The method comprises the following steps: acquiring a monitoring image and a standard image of each position of a building engineering construction site; marking any position as a target position, and calculating the scene complexity of the target position; feature points are obtained through an ORB algorithm and are matched; calculating an initial similarity degree of the monitoring images of the target position at any two adjacent moments; correcting the initial similarity degree; and determining the condition of the monitoring image according to the size of the corrected similarity so as to realize the monitoring of the construction of the building engineering. According to the method, the similarity degree is corrected through calculation, changes between the images are accurately evaluated, errors are reduced, changes related to construction are accurately recognized, irrelevant environmental interference is ignored, and the monitoring effectiveness is improved; analysis of irrelevant pictures is reduced, playback time is optimized, workload is reduced, and monitoring efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and particularly to a monitoring method and system for construction engineering. Background Art

[0002] Due to frequent occurrence of safety accidents at construction sites of construction projects, such as falls from heights, object strikes, etc., and complex potential hazards, like equipment failures and material stacking problems that are difficult to detect and hide, there are safety hazards. In addition, due to the large site area and complex flow of personnel and equipment, traditional patrol supervision is difficult to meet the requirements. In order to timely discover safety hazards and hold people accountable for subsequent accidents, it is necessary to monitor the construction sites of construction projects to ensure the safety and standardization of the construction process.

[0003] The patent document with the publication number CN113542690B discloses a building construction safety monitoring system and method, which relates to the technical field of building safety monitoring, and includes a monitoring module, a controller, a safety analysis module, a display module, an operation monitoring module, and an alarm module; the monitoring module is used to monitor personnel, equipment, materials, and construction status within a construction site, and divide the corresponding monitoring videos into verification videos and ordinary videos according to the opening and closing of construction equipment within the monitored area; the safety analysis module is used to perform safety analysis on the verification videos to obtain the monitoring values of the verification videos, and send the verification videos with monitoring values ≥ the monitoring threshold to the display module for synchronous display, prompting the administrator to pay key attention to, browse and check this video; the operation monitoring module is used to collect and analyze the operation data of construction equipment, judge whether the construction equipment is operating normally, reduce the losses caused by equipment failures, and play a role in early warning and active defense.

[0004] Currently, the monitoring system usually sets a key point marking (marking points on the video timeline) system, that is, when there is a change in the monitoring screen, the moment of the screen change is marked to facilitate the playback and viewing of the monitoring. However, the above patent document does not solve the problem that when marking the monitoring screen, due to factors such as light changes and wind effects, the monitoring screen will be marked, which interferes with the monitoring screen that really needs to be concerned, increases the subsequent playback and viewing time of the monitoring of construction projects, thereby reducing efficiency and increasing the workload. Summary of the Invention

[0005] In order to solve the problem that when marking the monitoring screen, due to external factors, the monitoring screen that really needs to be concerned is interfered, increasing the subsequent playback and viewing time of the monitoring of construction projects, thereby reducing efficiency and increasing the workload, the present invention provides a monitoring method and system for construction engineering.

[0006] In a first aspect, the present invention provides a monitoring method for construction engineering, adopting the following technical solutions: A monitoring method for construction engineering construction, comprising: obtaining monitoring images of each position at each moment on the construction site of the construction project, and standard images of each position in each time period, and obtaining the information entropy of the standard images by using the gray histogram of the standard images; taking any position as the target position, and calculating the scene complexity of the target position based on the information entropy of all the standard images of the target position; taking the monitoring images of the target position at any two adjacent moments as the target adjacent images, using a matching algorithm to match all the feature points of the two target adjacent images obtained by the ORB algorithm, and calculating the initial similarity degree of the two target adjacent images based on the number of matching feature points of the two target adjacent images, the number of feature points of each target adjacent image, the cosine similarity between the gray feature vectors of the two target adjacent images, and the scene complexity; obtaining the coordinates of the feature points of the two target adjacent images, calculating the distance between the two coordinates of each pair of matching feature points, and calculating the corrected similarity degree of the two target adjacent images based on the initial similarity degree, the variance of the distance, and the variances of the abscissas and ordinates of the unmatched feature points of the two target adjacent images; determining the condition of the monitoring images according to the magnitude of the corrected similarity degree of the two target adjacent images, so as to realize the monitoring of the construction engineering construction.

[0007] The beneficial effects are as follows: By obtaining the information entropy of the standard images and then calculating the scene complexity, it can effectively identify and exclude external interferences, and improve the monitoring accuracy of real construction activities; Using the Oriented FAST and Rotated BRIEF (ORB) algorithm to extract feature points and perform matching, and then by calculating the corrected similarity degree, accurately evaluate the changes between adjacent images, reduce errors, and then accurately identify the changes related to the construction, ignoring irrelevant environmental interferences, and improving the effectiveness of monitoring; By reducing the analysis of irrelevant pictures, optimizing the playback time, reducing the workload, and improving the monitoring efficiency; Dynamically adapting to different construction environments and stages, and flexibly adjusting the monitoring and analysis methods; Automated analysis reduces manual intervention and improves the intelligent level of monitoring.

[0008] Further, the standard image is an image without picture changes in the monitoring images of the corresponding position in each time period.

[0009] Further, the scene complexity satisfies the following relational expression: ; where is the scene complexity of position , is the number of all standard images of position , is position at the The information entropy of the standard images in a time period is a linear normalization function.

[0010] The beneficial effects are as follows: By calculating the mean value of the information entropy of the standard images at the same position in each time period and performing normalization processing, the scene complexity can be evaluated more accurately, so as to better identify important monitoring areas; Through the evaluation of the scene complexity, the monitoring images that need to be focused on can be effectively screened out, reducing irrelevant interference and improving the efficiency of monitoring playback and data processing.

[0011] Further, the matching algorithm adopts the nearest neighbor matching method.

[0012] Further, the acquisition method of the grayscale feature vector is as follows: The number of pixel points corresponding to each gray level in the grayscale histogram of the target adjacent images is used as the elements of the grayscale feature vector to form the grayscale feature vector of the target adjacent images.

[0013] Further, the initial similarity degree satisfies the following relational expression: ; In the formula, is the position in the adjacent th and th monitoring images at the initial similarity degree, is the number of matching feature points of the position in the adjacent th and th monitoring images, and are respectively the total number of feature points of the position in the adjacent th and th monitoring images, is the cosine similarity between the grayscale feature vectors of the position in the adjacent th and th monitoring images, is the scene complexity of the position is a preset parameter for adjusting the correction ability.

[0014] The beneficial effects are as follows: By comprehensively considering the number of feature point matches, the total number of feature points, and the cosine similarity of the grayscale feature vectors, the similarity between surveillance images can be measured more precisely, thereby improving the accuracy of similarity evaluation; introducing scene complexity and adjustment parameters enables the similarity evaluation to be dynamically adjusted according to the complexity of different scenes, improving adaptability in different environments. For example, in complex or dynamic environments, the system can make more detailed adjustments to the similarity, reducing misjudgments; calculating the similarity by integrating multiple factors avoids relying on a single indicator, thereby reducing the influence of external interference and improving robustness in noisy or dynamically changing environments; by more precisely evaluating the similarity between images, the system can more effectively identify important surveillance moments and areas, thereby enhancing the processing and analysis efficiency of surveillance data and reducing the interference of irrelevant information.

[0015] Further, the distance is the Euclidean distance.

[0016] Further, the corrected similarity degree of the two target adjacent images satisfies the following relational expression: ; where and are respectively the corrected similarity degree and the initial similarity degree of the surveillance images at positions at the adjacent th and th moments, is the variance of the distances between the two coordinates of all the matching feature points of the surveillance images at positions at the adjacent th and th moments, and are respectively the variances of the abscissas of all the unmatched feature points of the surveillance images at positions at the adjacent th and th moments, and are respectively the variances of the ordinates of all the unmatched feature points of the surveillance images at positions at the adjacent th and th moments, and are respectively the width and height of the surveillance images at position , is the natural exponential function, is the absolute value symbol.

[0017] The beneficial effects are as follows: By comprehensively correcting the initial similarity with multiple parameters, the similarity between images can be evaluated more precisely, reducing the risk of false matching and improving the matching accuracy; By adjusting the corrected similarity, a high matching reliability can be maintained even under imperfect image matching conditions; By introducing image size normalization (width-to-height ratio), the influence of image size on similarity calculation is eliminated, enabling fair comparison between images of different sizes and adapting to image data from different sources.

[0018] Furthermore, the monitoring for realizing construction engineering construction includes: in response to the corrected similarity degree of two target adjacent images being less than the average value of the corrected similarity degrees of the standard images at the target position in all adjacent two time periods, marking the two target adjacent images to complete the monitoring of construction engineering construction; The calculation method of the corrected similarity degree of the standard images at the target position in all adjacent two time periods is the same as that of the corrected similarity degree of the two target adjacent images.

[0019] The beneficial effects are as follows: The corrected similarity reduces the influence caused by image differences, improves the monitoring accuracy, and at the same time, helps to accurately identify the target image and reduce false marking; It reduces manual intervention, improves the monitoring efficiency, and saves manpower.

[0020] In a second aspect, the present invention provides a monitoring system for construction engineering construction, adopting the following technical solution: A monitoring system for construction engineering construction includes: a processor and a memory, and the memory stores computer program instructions, which implement the above-mentioned monitoring method for construction engineering construction when the computer program instructions are executed by the processor.

[0021] By adopting the above technical solution, the above-mentioned monitoring method for construction engineering construction is generated into a computer program and stored in the memory to be loaded and executed by the processor, so as to manufacture a terminal device according to the memory and the processor, which is convenient to use.

[0022] The present invention has the following technical effects: By precisely quantifying the complexity of the scene using the image grayscale histogram and information entropy, complex scenes can be distinguished, avoiding mis-triggering of markings by external interference factors and reducing unnecessary markings on surveillance images; according to the feature point matching results between two adjacent target images, the initial similarity degree is calculated. By combining the local features and global features of the images, it is possible to more accurately determine whether two adjacent target images are similar; based on the ORB algorithm and the matching algorithm, according to the coordinates of all feature points of two adjacent target images, the initial similarity degree is corrected, and it is possible to more accurately judge whether there is an actual object change between two adjacent target images, rather than a visual change caused by environmental factors; accurately judge the status of surveillance images, greatly shorten the playback viewing time, improve the surveillance efficiency, reduce the burden on staff, and ensure the effective implementation of construction project surveillance. Brief Description of the Drawings

[0023] By referring to the following detailed description with reference to the accompanying drawings, the above and other objects, features, and advantages of the exemplary embodiments of the present invention will become readily understood. In the drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts.

[0024] Figure 1 It is a flowchart of a method in an embodiment of a surveillance method for construction project construction of the present invention. Detailed Embodiments

[0025] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0026] It should be understood that when the claims, specifications, and drawings of the present invention use terms such as "first" and "second", they are only used to distinguish different objects, rather than to describe a specific order. The terms "including" and "comprising" used in the specifications and claims of the present invention indicate the presence of the described features, wholes, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their combinations.

[0027] An embodiment of the present invention discloses a surveillance method for construction project construction. Referring to Figure 1 , it includes steps S1 - S6: S1: Obtain surveillance images and standard images of each position of the construction project construction site.

[0028] Obtain the monitoring images of each location at each moment on the construction site of the building project, as well as the standard images of each location in each time period, and obtain the information entropy of the standard images by using the gray histogram of the standard images.

[0029] Specifically, the standard image is an image without picture changes in the monitoring images of the corresponding location in each time period.

[0030] According to the actual situation of the construction site of the building project, arrange the monitoring equipment. For example, arrange the monitoring equipment at appropriate positions such as high-risk operation areas, material and equipment storage areas, etc., and collect the monitoring images of each moment in the monitoring scenarios (each location) and the standard images of each time period (the specific duration of the time period is not limited, it can be a time period under different lighting conditions in a day, or it can be a time period demarcated by morning, noon, and evening). The standard image is an image without picture changes among multiple monitoring pictures in each time period of the monitoring scenario. For example, the monitoring pictures of the equipment storage area include the monitoring pictures with only equipment and the monitoring pictures with operators taking equipment. At this time, select the former as the standard image.

[0031] S2: Denote any location as the target location and calculate the scene complexity of the target location.

[0032] It should be noted that the monitoring pictures collected in different monitoring scenarios have different characteristics, and the feature points of the ORB algorithm are obtained through the local features of the images. In some monitoring scenarios, the changes in the monitoring pictures are easy to be recognized, while in some monitoring scenarios, the changes in the monitoring pictures are not easy to be recognized. This phenomenon will interfere with the recognition of the changes in the monitoring pictures. Therefore, the present invention calculates the scene complexity of the monitoring scenario through the monitoring images.

[0033] Calculate the scene complexity of the target location based on the information entropy of all the standard images of the target location.

[0034] Specifically, the scene complexity satisfies the following relational expression: ; In the formula, is the scene complexity of location , is the number of all standard images of location , is the information entropy of the standard image of location in the th time period, is the linear normalization function.

[0035] Among them, The larger it is, the more complex the monitoring picture under this location is, and the greater the complexity of this location is. The smaller it is, the simpler the monitoring image at this position is, and the lower the complexity of this position is.

[0036] S3: Obtain feature points through the ORB algorithm and perform matching.

[0037] Denote the monitoring images of the target position at any two adjacent moments as target adjacent images, and use the matching algorithm to match all the feature points of the two target adjacent images obtained through the ORB algorithm.

[0038] Specifically, the matching algorithm adopts the nearest neighbor matching method.

[0039] S4: Calculate the initial similarity degree of the monitoring images of the target position at any two adjacent moments.

[0040] It should be noted that since the feature points generated by the ORB algorithm can well describe the local characteristics of the image, the initial similarity degree between two images can be judged by the result of feature point matching to determine whether there are substantial changes in the picture content. Compared with the full-image pixel-level comparison, it has stronger anti-noise performance. Therefore, the present invention calculates the initial similarity degree between two monitoring images according to the feature point matching result between two monitoring images at adjacent moments at the same position.

[0041] Based on the number of matching feature points of the two target adjacent images, the number of feature points of each target adjacent image, the cosine similarity between the gray feature vectors of the two target adjacent images, and the scene complexity, calculate the initial similarity degree of the two target adjacent images.

[0042] Specifically, the acquisition method of the gray feature vector is as follows: Take the number of pixel points corresponding to each gray level in the gray histogram of the target adjacent image as the elements of the gray feature vector to form the gray feature vector of the target adjacent image. For example, in a certain gray histogram, there are 10 pixel points with a gray value of 0, 11 pixel points with a gray value of 1,..., and 110 pixel points with a gray value of 255, then the gray feature vector is (10, 11,..., 110).

[0043] Specifically, the initial similarity degree satisfies the following relational expression: ; In the formula, is the initial similarity degree of the monitoring images at the position at the th and th adjacent moments, is the initial similarity degree of the monitoring images at the position at the th and The number of matching feature points in the monitoring images at each moment, and are respectively the total number of feature points in the monitoring images at the adjacent -th and -th moments, is the cosine similarity between the gray feature vectors of the monitoring images at the adjacent -th and -th moments for the position, is the scene complexity of the position , is the preset parameter for adjusting the correction ability.

[0044] Implementers can set the value of according to the specific implementation situation, for example, 2.

[0045] Among them, when two adjacent target images are more similar, the gray histograms of the two images are more likely to have similar distribution patterns, so the cosine similarity between the gray feature vectors is greater. When the difference between the two images is greater, the gray histograms of the two images are more likely to have different distribution patterns, so the cosine similarity between the gray feature vectors is smaller. Therefore the larger it is, the greater the initial similarity between two adjacent target images, the smaller it is, the smaller the initial similarity between two adjacent target images. For the convenience of calculation, here is normalized by the method of adding 1 and dividing by 2. At the same time, when two images are more similar, there will be more feature points successfully matched in the two images. When the two images are exactly the same, all feature points in the two images will be successfully matched. When the two images are different, there will be fewer feature points successfully matched in the two images. Therefore the larger it is, the greater the initial similarity between the two images, the smaller it is, the smaller the initial similarity between the two images. The same principle is not elaborated here. In positions with different complexities, different judgments are made on whether the images are similar. In a simple monitoring scene (position), if the discrimination of the images is too strict, a large number of images in the monitoring screen may be marked. In a complex monitoring scene, if the discrimination is too loose, important changes in the monitoring screen may be ignored. Therefore, through corrects . The larger it is, then the larger it is, then is more likely to be greater than 1. According to the gamma transformation principle, at this time Pair plays a downward correction role, making the judgment of whether the two images at this position are similar more stringent. The smaller it is, the more likely it is to be less than 1. At this time, Pair plays an upward correction role, making the judgment of whether the two images at this position are similar more lenient.

[0046] In another embodiment, the initial similarity degree satisfies the following relational expression: ; In the formula, is the initial similarity degree of the monitoring images at position at the adjacent th and th moments, is the number of matching feature points of the monitoring images at position at the adjacent th and th moments, and are respectively the total number of feature points of the monitoring images at position at the adjacent th and th moments, is the cosine similarity between the gray feature vectors of the monitoring images at position at the adjacent th and th moments, is the scene complexity of position , is a preset parameter for adjusting the correction ability, is the maximum value function.

[0047] S5: Correct the initial similarity degree.

[0048] It should be noted that due to the influence of light changes and wind force, the monitoring screen may be marked when there is no substantial change, which causes certain interference to the review of the real changes in the screen that need to be concerned. To weaken the influence of this phenomenon on the marking of the monitoring screen, the present invention corrects the initial similarity degree through the coordinates of the feature points of two adjacent target images.

[0049] Obtain the coordinates of the feature points of two adjacent target images, calculate the distance between the two coordinates of each pair of matching feature points, and calculate the corrected similarity degree of the two adjacent target images based on the initial similarity degree, the variance of the distance, and the variances of the abscissas and ordinates of the unmatched feature points of the two adjacent target images.

[0050] Specifically, the distance is the Euclidean distance.

[0051] Specifically, the corrected similarity between the two adjacent target images satisfies the following relationship: ; In the formula, and The location In the adjacent and The corrected similarity and initial similarity of the surveillance images at each moment, For location In the adjacent and The variance of the distance between the two coordinates of all matching feature points in the monitoring image at a certain moment, and The location In the adjacent and The variance of the horizontal coordinates of all unmatched feature points in the monitoring image at the moment, and The positions are In the adjacent and The variance of the ordinates of all unmatched feature points in the monitoring image at the moment, and Respective locations The width and height of the monitoring image, is the natural exponential function, is the absolute value symbol.

[0052] in, The smaller the value, the more consistent the overall position changes of the feature points in the two images are, and the more likely the phenomenon is caused by changes in lighting or wind. The larger the value is, the more inconsistent the overall position changes of the feature points in the two images are. This phenomenon is more likely to be caused by substantial changes in the objects in the monitoring screen. In order to reduce the impact of lighting changes or wind effects on the initial similarity of the two images, right Make an upward correction. The smaller the time, the right The greater the degree of correction, the greater the degree of correction similarity between the images of the monitoring screen changes caused by changes in lighting or wind. When the value is larger, substantial changes may occur in the monitoring screen, and it is less necessary to Make corrections, then The smaller the degree of correction is The larger it is, the greater the distribution difference of the coordinates of the unmatched feature points in the two images, and the more likely this phenomenon is caused by the movement of the object in the monitored video. Then, the smaller the correction similarity degree between the two images is; The smaller it is, the smaller the distribution difference of the coordinates of the unmatched feature points in the two images, and the more likely there is no substantial change in the monitored video. Then, the greater the correction similarity degree between the two images. For the convenience of calculation, the is normalized; Similarly, it will not be elaborated here

[0053] S6: Determine the status of the monitored image according to the magnitude of the correction similarity degree to implement the monitoring of the construction project

[0054] Determine the status of the monitored image according to the magnitude of the correction similarity degree between two adjacent target images to implement the monitoring of the construction project

[0055] Specifically, the implementation of the monitoring of the construction project includes: In response to the correction similarity degree between two adjacent target images being less than the average value of the correction similarity degrees of the standard images at all adjacent two time periods of the target position, mark the two adjacent target images to complete the monitoring of the construction project; The calculation method of the correction similarity degree of the standard images at all adjacent two time periods of the target position is the same as that of the correction similarity degree between the two adjacent target images. Only adjust the monitored images at two adjacent moments in the calculation process to the standard images at two adjacent time periods, and it will not be elaborated here

[0056] The embodiment of the present invention also discloses a monitoring system for construction project, including a processor and a memory. The memory stores computer program instructions, and when the computer program instructions are executed by the processor, a monitoring method for construction project according to the present invention is implemented

[0057] The above system also includes other components well-known to those skilled in the art such as a communication bus and a communication interface. Their settings and functions are known in the art, so they will not be elaborated here

[0058] In the present invention, the aforementioned memory may be any tangible medium that contains or stores a program, which can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, a computer-readable storage medium may be any suitable magnetic storage medium or magneto-optical storage medium, or any other medium that can be used to store the required information and can be accessed by an application program, module, or both. Any such computer storage medium may be part of the device or accessible or connectable to the device.

[0059] Although this specification has shown and described several embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Many changes, alterations, and alternative forms will occur to those skilled in the art without departing from the spirit and scope of the present invention. It should be understood that various alternatives to the embodiments of the present invention described herein may be employed in practicing the present invention.

[0060] The above are all preferred embodiments of the present invention, and the protection scope of the present invention is not limited thereby. Therefore, all equivalent changes made according to the structure, shape, and principle of the present invention should be covered within the protection scope of the present invention.

Claims

1. A monitoring method for construction engineering, characterized in that: include: Obtain monitoring images of each location at each time of the construction site, as well as standard images of each location at each time period, and use the grayscale histogram of the standard image to obtain the information entropy of the standard image; Any position is recorded as the target position, and the scene complexity of the target position is calculated based on the information entropy of all standard images of the target position; the monitoring images of the target position at any two adjacent moments are recorded as target adjacent images, and all feature points of the two target adjacent images obtained by the ORB algorithm are matched using a matching algorithm, and the initial similarity of the two target adjacent images is calculated based on the number of matching feature points of the two target adjacent images, the number of feature points of each target adjacent image, the cosine similarity between the grayscale feature vectors of the two target adjacent images, and the scene complexity; Obtaining coordinates of feature points of two adjacent target images, calculating the distance between two coordinates of each pair of matching feature points, and calculating a corrected similarity between the two adjacent target images based on the initial similarity, the variance of the distance, and the variance of the horizontal coordinates and the variance of the vertical coordinates of the unmatched feature points of the two adjacent target images; According to the degree of corrected similarity between two adjacent target images, the monitoring image status is determined to realize the monitoring of the construction project.

2. A monitoring method for construction engineering according to claim 1, characterized in that: The standard image is an image without picture changes in the surveillance images of the corresponding location in each time period.

3. A monitoring method for construction engineering according to claim 1, characterized in that: The scene complexity satisfies the following relationship: ； In the formula, For location The complexity of the scene, For location The number of all standard images, For location In the The information entropy of the standard image in time period is is a linear normalization function.

4. A monitoring method for construction engineering according to claim 1, characterized in that: The matching algorithm adopts the nearest neighbor matching method.

5. A monitoring method for construction engineering according to claim 1, characterized in that: The grayscale feature vector is obtained as follows: The number of pixels corresponding to each gray level in the grayscale histogram of the target adjacent image is used as the element of the grayscale feature vector to form the grayscale feature vector of the target adjacent image.

6. A monitoring method for construction engineering according to claim 1, characterized in that: The initial similarity satisfies the following relationship: ； In the formula, For location In the adjacent and The initial similarity of the surveillance images at the moment, For location In the adjacent and The number of matching feature points of the surveillance image at a certain moment, and The positions are In the adjacent and The total number of feature points of the monitoring image at each moment, For location In the adjacent and The cosine similarity between the grayscale feature vectors of the monitoring images at each moment, For location The complexity of the scene, For adjustment Modify the preset parameters of the ability.

7. A monitoring method for construction engineering according to claim 1, characterized in that: The distance is the Euclidean distance.

8. A monitoring method for construction engineering according to claim 1, characterized in that: The corrected similarity of the two adjacent target images satisfies the following relationship: ； In the formula, and The positions are In the adjacent and The corrected similarity and initial similarity of the surveillance images at each moment, For location In the adjacent and The variance of the distance between the two coordinates of all matching feature points in the monitoring image at a certain moment, and The positions are In the adjacent and The variance of the horizontal coordinates of all unmatched feature points in the monitoring image at the moment, and The positions are In the adjacent and The variance of the ordinates of all unmatched feature points in the monitoring image at the moment, and Respective locations The width and height of the monitoring image, is the natural exponential function, is the absolute value symbol.

9. A monitoring method for construction engineering according to claim 8, characterized in that: The monitoring of construction works includes: In response to the fact that the corrected similarity of the two adjacent target images is less than the average of the corrected similarities of the standard images of the target position in all two adjacent time periods, the two adjacent target images are marked to complete the monitoring of the construction of the building project; The calculation method of the corrected similarity degree of the standard images of the target position in all two adjacent time periods is the same as the calculation method of the corrected similarity degree of the two adjacent target images.

10. A monitoring system for construction engineering, characterized in that: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a monitoring method for construction engineering according to any one of claims 1 to 9 is implemented.

Images