Intelligent remote security and protection monitoring system for constructional engineering

Abstract

The invention discloses an intelligent remote security and protection monitoring system for constructional engineering, which relates to the technical field of constructional engineering and comprises a monitoring time period acquisition module, a to-be-analyzed image data packet acquisition module, an image processing module and an abnormity monitoring module, the technical problems that an existing safety monitoring system of the unmanned aerial vehicle is simple in judgment and analysis of the shielding condition and difficult to accurately judge shielding are solved, images shot by the unmanned aerial vehicle can be analyzed in real time through quantitative analysis of the area of the shielding region and calculation of complex coefficients of shielding boundary lines, and the safety of the unmanned aerial vehicle is improved. The blocking area abnormal block and the blocking boundary abnormal block can be rapidly marked, the abnormal change of the blocking area can be accurately identified, the accuracy and sensitivity of abnormal detection can be improved, the potential safety hazard can be timely eliminated, and the construction safety of the construction site can be ensured.

Description

Technical Field

[0001] The invention belongs to the technical field of construction engineering, and in particular is an intelligent remote security monitoring system for construction engineering. Background Art

[0002] Safety monitoring at construction sites is an important part of ensuring construction safety. Traditional security monitoring systems mainly rely on fixed cameras for real-time monitoring. However, due to the complex environment of the construction site and the presence of a large number of obstructions, such as building materials, equipment, and temporary facilities, there are many blind spots in monitoring, making it difficult to fully cover the construction area. In addition, traditional monitoring systems lack the ability to analyze dynamic changes in obstruction areas, and are unable to promptly detect potential safety hazards caused by abnormal accumulation of obstructions or equipment failures.

[0003] In recent years, with the rapid development of drone technology, drones have been widely used in the field of monitoring construction sites. Drones can fly flexibly and capture image data of construction sites from multiple angles and in all directions, effectively making up for the shortcomings of fixed cameras.

[0004] However, although the existing image data taken by drones has a wide coverage, it lacks the ability to quantitatively analyze the occluded area, and cannot accurately determine whether the change of the occluded area is normal. It also lacks accurate calculation and dynamic analysis of the area and boundary changes of the occluded area, making it difficult to accurately determine whether the occlusion is abnormal, and cannot take timely and effective measures to improve the monitoring effect and eliminate safety hazards. Based on this, an intelligent remote security monitoring system for construction projects is proposed. Summary of the invention

[0005] The purpose of the present invention is to provide an intelligent remote security monitoring system for construction projects, which solves the technical problem that the existing safety monitoring system of unmanned aerial vehicles has a relatively simple judgment and analysis of occlusion conditions and is difficult to accurately judge whether the occlusion is abnormal.

[0006] An intelligent remote security monitoring system for construction projects, comprising: A monitoring period acquisition module obtains multiple monitoring periods corresponding to the construction site; The module for acquiring the image data packet to be analyzed acquires multiple regional images taken by the drone of the construction area in multiple monitoring periods in the preset number of days T2, and binds the multiple regional images to the corresponding monitoring periods one by one to form an image data packet calibrated with a time series. Here, T2 is a positive integer, satisfying 360≥T2≥30; An image processing module obtains a reference table of blocked area intervals and a reference table of boundary transformation intervals corresponding to the construction site in each monitoring period; The abnormality monitoring module is used to obtain the real-time monitoring image of the UAV, obtain the real-time occlusion area and real-time complexity coefficient corresponding to each calibrated block in the real-time monitoring image, and mark the occlusion abnormal blocks in the real-time monitoring image according to the occlusion area interval reference table and the boundary transformation interval reference table.

[0007] As a further solution of the present invention: the specific method of obtaining multiple monitoring time periods corresponding to the construction site is: The specified construction period of the construction site is evenly divided into multiple monitoring periods Ch according to the preset time interval T1, where T1 is the preset time interval, h is a different monitoring period, h=1, 2, ..., n, n is the number of monitoring periods, and n is a positive integer, satisfying n≥2.

[0008] As a further solution of the present invention: the image processing module includes a pre-division processing unit, an occlusion area interval reference table acquisition unit and a boundary transformation interval reference table acquisition unit; A pre-division processing unit, used for evenly dividing each area image in the image data packets to be analyzed corresponding to the multiple monitoring time periods into a plurality of calibration blocks; The obstruction area interval reference table acquisition unit is used to analyze each calibration block of each regional image in the image data packet to be analyzed corresponding to each monitoring period, and obtain the obstruction area interval reference table corresponding to each monitoring period of the construction site; The boundary transformation interval reference table acquisition unit is used to analyze each calibration block of each regional image in the image data packet to be analyzed corresponding to each monitoring period, and obtain the boundary transformation interval reference table corresponding to each monitoring period of the construction site.

[0009] As a further solution of the present invention: the specific method of obtaining the reference table of the shielding area intervals corresponding to the construction site at each monitoring period is: S1: Randomly select a monitoring period from each monitoring period as the target period; S2: Randomly select a calibration block from multiple calibration blocks as the target block; Obtain the occluded area Mr corresponding to the target blocks of each regional image in the target period image data packet, where r refers to the different regional images contained in the target period image data packet, and the value is r=1, 2, ..., a, where a refers to the number of regional images contained in the target period image data packet, and a is a positive integer and a>1; S3: Filter Mr according to the preset screening condition A, obtain the value Mb that meets the condition, and determine the standard occlusion area H1 according to the comparison result of the number of Mb and the preset threshold Y1, and generate the occlusion area interval L1 [FB, FA] according to the mean CZ of the absolute value of the difference between the standard occlusion area H1 and the mean Mp of Mr; S4: Obtain the occlusion area intervals Le corresponding to each calibration block of the construction site in the target period, bind the occlusion area intervals Le corresponding to each calibration block in the target period with the target period, and generate a reference table R1 of the occlusion area intervals corresponding to the construction site in the target period, where e refers to different calibration blocks in the regional image, and is the block number corresponding to each calibration block, and the value is e=1, 2, ..., c, where c refers to the number of calibration blocks, c is a positive integer and c>1; S5: Repeat the above steps S1-S4 to generate a reference table Rh of the shielding area intervals corresponding to the construction site in each monitoring period.

[0010] As a further solution of the present invention, a specific method of generating the occlusion area interval L1 is: Obtain the mean CZ of the absolute values ​​of the differences between the areas of each occlusion region Mr and the mean Mp of Mr, and use the sum of the standard occlusion area area H1 and CZ corresponding to the calibrated block of the construction site within the target period as the upper limit FA of the occlusion area interval corresponding to the calibrated block of the construction site within the target period, and use the difference between the standard occlusion area area H1 and CZ corresponding to the calibrated block of the construction site within the target period as the lower limit FB of the occlusion area interval corresponding to the calibrated block of the construction site within the target period, thereby generating the occlusion area interval L1 [FB, FA] corresponding to the calibrated block of the construction site within the target period.

[0011] As a further solution of the present invention: the specific method of determining the standard shielding area H1 is: Get the numerical value Mb of each occlusion area Mr that meets the preset filtering condition A: |Mr-Mp|≥Y2; wherein b is the number of numerical values ​​in Mr that meet the preset filtering condition A, a≥b≥1, when the number b is greater than the preset threshold value Y1, define the mean value Mp of Mr as the standard occlusion area H1 corresponding to the calibrated block of the construction site during the target period, when the number b is less than the preset threshold value Y1, define the mean value of the maximum and minimum values ​​in Mr as the standard occlusion area H1 corresponding to the calibrated block of the construction site during the target period.

[0012] As a further solution of the present invention, a specific method of obtaining a reference table of boundary transformation intervals corresponding to each monitoring period of a construction site is as follows: S01: Select the same monitoring period as step S1 from each monitoring period as the target period; S02: Selecting the same calibrated block as in step S2 from multiple calibrated blocks as a target block; S03: Obtain the complex coefficients Ur of the occlusion boundary lines corresponding to the target blocks of each regional image in the target period image data packet; generate the boundary transformation interval X1 according to the mean value Up of the complex coefficients Ur and the mean value Q of the absolute value of the difference between Ur and Up; S04: Repeat steps S02-S03 to obtain the boundary transformation intervals Xe corresponding to each calibrated block of the construction site in the target period, bind the boundary transformation intervals Xe corresponding to each calibrated block in the target period with the target period, and generate a boundary transformation interval reference table K1 corresponding to the construction site in the target period S04: Repeat the above steps S02-S03 to generate a boundary transformation interval reference table Kh corresponding to each monitoring period of the construction site.

[0013] As a further solution of the present invention, the specific method of obtaining the complex coefficient Ur of the occlusion boundary line corresponding to the target blocks of each regional image in the target period image data packet is: S031: Select a target block from the target blocks of each regional image without replacement as the analysis block; obtain the occlusion boundary line corresponding to the occlusion area in the analysis block, construct a two-dimensional coordinate system with the center point of the analysis block as the origin, obtain the coordinates corresponding to each inflection point on the occlusion boundary line of the analysis block, and calculate the distance Gg between each inflection point on the occlusion boundary line of the analysis block and the center point of the analysis block according to the Euclidean distance formula; through the formula, , obtain the complex coefficient U1 of the occlusion boundary line of the analysis block, Gi is any one of Gg, where Gp is ​​the mean value of Gg, g refers to different inflection points on the occlusion boundary line of the analysis block, and the value is g=1, 2, ..., d, where d refers to the number of inflection points on the occlusion boundary line of the analysis block, d is a positive integer and d>1; S032: Repeat step S031 to obtain the complex coefficients Ur corresponding to the occlusion boundary lines of the target blocks of each area image in the target time period image data packet, obtain the mean Q of the absolute values ​​of the differences between each complex coefficient Ur and Up, and use the sum of Up and Q as the upper limit value BHA of the boundary transformation interval corresponding to the target block of the construction site in the target time period, and use the sum of Up and Q as the lower limit value BHB of the boundary transformation interval corresponding to the target block of the construction site in the target time period, and then the boundary transformation interval X1 [BA, BA] corresponding to the target block of the construction site in the target time period, where Up is the mean value of Ur.

[0014] As a further solution of the present invention, the specific method of marking the abnormal occlusion block in the real-time monitoring image is: The real-time occlusion area corresponding to each calibrated block in the real-time monitoring image is obtained, and the occlusion area interval reference table of the corresponding time period is matched according to the acquisition time of the real-time monitoring image, which is used as the occlusion area interval comparison reference table, and the real-time occlusion area corresponding to each calibrated block in the real-time monitoring image is compared with the occlusion area interval corresponding to each calibrated block in the occlusion area interval comparison reference table one by one, and the calibrated block whose real-time occlusion area does not belong to the corresponding occlusion area interval is marked as an abnormal occlusion area block, otherwise no processing is performed, and the corresponding occlusion area of ​​the block marked as abnormal occlusion area in the real-time monitoring image is obtained. The real-time complexity coefficient of the corresponding occlusion boundary line is matched according to the boundary transformation interval reference table of the corresponding time period according to the acquisition time of the real-time monitoring image, and it is used as the boundary transformation interval comparison reference table. The real-time complexity coefficient of the occlusion boundary line corresponding to the abnormal occlusion area block in the real-time monitoring image is compared with the boundary transformation interval corresponding to the corresponding calibration block in the boundary transformation interval comparison reference table one by one, and the calibration block whose real-time complexity coefficient does not belong to the corresponding boundary transformation interval is marked as an abnormal occlusion boundary block, and the corresponding calibration block is marked as an abnormal occlusion block for output, otherwise no processing is done.

[0015] Compared with the prior art, the present invention has the following beneficial effects: The present invention can analyze the images taken by the drone in real time through quantitative analysis of the area of ​​the occluded region and calculation of the complex coefficient of the occluded boundary line, and quickly mark the abnormal blocks of the occluded area and the abnormal blocks of the occluded boundary, that is, the abnormal occluded blocks. The system can accurately identify abnormal changes in the occluded area and improve the accuracy and sensitivity of abnormal detection. The image data taken by the drone is obtained in real time, and the abnormal changes in the occluded area are quickly analyzed and judged. Once an abnormality is detected, the system will immediately generate an abnormal report and notify the relevant personnel to ensure that the problem is handled in a timely manner, so that the staff can go to the site in time for investigation and processing, and can eliminate safety hazards in time to ensure the construction safety of the construction site. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a schematic diagram of the system framework structure of the present invention; Figure 2 Schematic diagram of the calibration process for blocking abnormal blocks. DETAILED DESCRIPTION

[0017] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than 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.

[0018] Example 1: Please refer to Figure 1 - Figure 2 ,This application provides an intelligent remote security monitoring system for construction projects, including; A monitoring period acquisition module sets multiple monitoring periods according to the specified construction period of the construction site, and then obtains multiple monitoring periods corresponding to the construction site; It should be noted that the prescribed construction period is the prescribed construction time of the construction site, which can be manually obtained and entered, and the time intervals between multiple monitoring periods can be equal time intervals or unequal time intervals. Here, the time intervals between multiple monitoring periods are equal time intervals; according to the preset time interval T1, the prescribed construction period of the construction site is evenly divided into multiple monitoring periods Ch, wherein T1 is the preset time interval, and the specific value is formulated by relevant personnel according to needs, h is different monitoring periods, h=1, 2, ..., n, n is the number corresponding to the monitoring period, and n is a positive integer, satisfying n≥2; The module for acquiring the image data packet to be analyzed acquires multiple regional images of the construction area taken by the drone in the preset days T2 in multiple monitoring periods, sorts the multiple monitoring periods of the construction site in chronological order, and matches and binds the multiple regional images of the construction area taken by the drone in each monitoring period in the preset days T2 to the corresponding monitoring period one by one, thereby forming an image data packet calibrated with the time series, and then obtaining the image data packets to be analyzed corresponding to the multiple monitoring periods of the construction site, wherein the specific value of the preset days T2 is formulated by relevant personnel according to needs, and here, T2 is a positive integer, satisfying 360≥T2≥30; Among them, the image data taken by the drone in the construction area during multiple monitoring periods in the preset number of days T2 are obtained to ensure that the image data taken by the drone can fully cover these periods. The preset number of days T2 is not less than 30 days to obtain more representative data; It should be noted that the use of drones to take multi-angle and all-round images of construction sites is an existing and mature technology, so no further explanation will be given here; The image processing module analyzes the image data in the image data packets to be analyzed corresponding to the multiple monitoring periods of the construction site, and then obtains the reference table of the occlusion area interval and the reference table of the boundary transformation interval corresponding to the construction site in each monitoring period, specifically in the following manner: A pre-division processing unit performs the same pre-division processing on each area image in the image data packet to be analyzed corresponding to each of the multiple monitoring periods of the construction site, so that each area image has multiple calibration blocks; The specific method of the pre-division processing is: the regional image is evenly divided into a plurality of calibration blocks. It should be noted that the pre-division processing methods of each regional image in the image data packets to be analyzed corresponding to the plurality of monitoring time periods are the same, and thus the number and shape of the calibration blocks in each regional image are the same; It should be noted that the images taken by the default drone have a fixed number of pixels and a fixed actual length, and there are no special cases such as irregular image size; The obstruction area interval reference table acquisition unit analyzes each calibration block of each regional image in the image data packet to be analyzed corresponding to each monitoring period, and then obtains the obstruction area interval reference table corresponding to each monitoring period of the construction site, specifically in the following manner: S1: Randomly select a monitoring period from each monitoring period without replacement as the target period; S2: Randomly select a calibration block from multiple calibration blocks without replacement as the target block; S3: Obtain and output the occluded area Mr corresponding to the target blocks of each regional image in the target period image data packet, where r refers to the different regional images contained in the target period image data packet, and the value is r=1, 2, ..., a, where a refers to the number of regional images contained in the target period image data packet, and a is a positive integer and a>1; It should be noted that for construction site images taken by drones, image analysis software (such as Adobe Photoshop, ImageJ, etc.) can be used. After importing the image into the software, the software's measurement tool is used to set the conversion relationship between image pixels and actual length according to a certain ratio (for example, it is known that an object with an actual length of 10 meters in the image corresponds to a pixel length, so as to determine the conversion ratio), and then manually or using the software's automatic recognition function to outline the occluded area. The software can calculate the area of ​​the corresponding outline (in pixels, and then the actual area is obtained through the conversion ratio) and the occluded area. The above are all existing and mature technologies, so no further elaboration is made here.

[0019] Get the value Mb of each occlusion area Mr that meets the preset filtering condition A, where b is the number of values ​​in Mr that meet the preset filtering condition A, a≥b≥1, and compare the number b with the preset threshold value Y1. When the number b is greater than the preset threshold value Y1, it means that the number of values ​​in Mr that meet the preset filtering condition A is large, and then the mean of Mr is representative, and then the mean Mp of Mr is defined as the standard occlusion area H1 corresponding to the calibration block of the construction site during the target period. When the number b is less than the preset threshold value Y1, it means that the number of values ​​in Mi that meet the preset filtering condition A is small, and the mean of Mr is not representative, and then the mean of the maximum and minimum values ​​in Mr is defined as the standard occlusion area H1 corresponding to the calibration block of the construction site during the target period, that is, H1=(Mmin+Mmax) / 2, where Mmax and Mmin are the maximum and minimum values ​​in Mr, respectively. Here, the preset screening condition A is specifically: |Mr-Mp|≥Y2, where Mp is the mean value of Mr, Y2 is the preset threshold, and the specific values ​​of the preset thresholds Y1 and Y2 are formulated by relevant personnel according to needs; Obtain the mean value CZ of the absolute value of the difference between each occlusion area Mr and the mean value Mp of Mr, take the sum of the standard occlusion area H1 and CZ corresponding to the calibration block of the construction site in the target period as the upper limit FA of the occlusion area interval corresponding to the calibration block of the construction site in the target period, take the difference between the standard occlusion area H1 and CZ corresponding to the calibration block of the construction site in the target period as the lower limit FB of the occlusion area interval corresponding to the calibration block of the construction site in the target period, and then generate the occlusion area interval L1 [FB, FA] corresponding to the calibration block of the construction site in the target period; S4: Repeat the above steps S2-S3 to obtain the occlusion area intervals Le corresponding to each calibrated block of the construction site in the target period, bind the occlusion area intervals Le corresponding to each calibrated block in the target period with the target period, and generate a reference table R1 of occlusion area intervals corresponding to the construction site in the target period; Where e refers to different calibration blocks in the regional image, and is the block number corresponding to each calibration block, and the value is e=1, 2, ..., c, where c refers to the number of calibration blocks, c is a positive integer and c>1; S5: Repeat the above steps S1-S4 to obtain the reference table Rh of the shielding area intervals corresponding to the construction site in each monitoring period.

[0020] The boundary transformation interval reference table acquisition unit analyzes each calibration block of each regional image in the image data packet to be analyzed corresponding to each monitoring period, and then obtains the boundary transformation interval reference table corresponding to each monitoring period of the construction site, specifically in the following manner: S01: Select the same monitoring period as step S1 from each monitoring period as the target period; S02: Selecting the same calibrated block as in step S2 from multiple calibrated blocks as a target block; S03: obtaining the center points corresponding to the target blocks of each regional image in the target period image data packet and the occlusion boundary lines corresponding to the occlusion areas in each target block, analyzing them to obtain the complex coefficients corresponding to the occlusion boundary lines of the target blocks of each regional image in the target period image data packet, analyzing the complex coefficients corresponding to the occlusion boundary lines of the target blocks of each regional image in the target period image data packet to obtain the boundary transformation interval reference table X1 corresponding to the target blocks of the construction site in the target period; S031: selecting a target block from the target blocks of each regional image as an analysis block without replacement; Obtain the occlusion boundary line corresponding to the occlusion area in the analysis block, construct a two-dimensional coordinate system with the center point of the analysis block as the origin, obtain the coordinates corresponding to each inflection point on the occlusion boundary line of the analysis block, and calculate the distance Gg between each inflection point on the occlusion boundary line of the analysis block and the center point of the analysis block according to the Euclidean distance formula; It should be noted that the edge detection algorithm of image processing software (such as OpenCV) can be used to extract the coordinates of the inflection points on the occlusion boundary line, and the center point of the analysis block is taken as the origin (0,0) to construct a two-dimensional rectangular coordinate system. The unit length of the x-axis and the y-axis can be the conversion ratio between the image pixels and the actual length. The above are all existing and mature calculation methods, so they will not be elaborated here.

[0021] Where g refers to different inflection points on the occlusion boundary line of the analysis block, and the value is g=1, 2, ..., d, where d refers to the number of inflection points on the occlusion boundary line of the analysis block, d is a positive integer and d>1; By formula, , obtain the complex coefficient U1 of the occlusion boundary line of the analysis block, Gi is any one of Gg, where Gp is ​​the mean value of Gg, i≥d≥1; S032: Repeat step S031, and then obtain the complex coefficients Ur corresponding to the occlusion boundary lines of the target blocks of each area image in the target period image data packet, and obtain the average value Q of the absolute value of the difference between each complex coefficient Ur and Up, wherein Up is the average value of Ur, and take the sum of Up and Q as the upper limit value BHA of the boundary transformation interval corresponding to the target block of the construction site in the target period, and take the sum of Up and Q as the lower limit value BHB of the boundary transformation interval corresponding to the target block of the construction site in the target period, and then the boundary transformation interval X1 [BA, BA] corresponding to the target block of the construction site in the target period; S04: repeat the above steps S02-S03, and then obtain the boundary transformation intervals Xe corresponding to each calibrated block of the construction site in the target period, bind the boundary transformation intervals Xe corresponding to each calibrated block in the target period with the target period, and generate a boundary transformation interval reference table K1 corresponding to the construction site in the target period; S05: Repeat the above steps S01-S04, and then obtain the boundary transformation interval reference table Kh corresponding to each monitoring period of the construction site; The abnormal monitoring module acquires the real-time monitoring image of the UAV, and inputs it into the image processing module to pre-divide the real-time monitoring image, and obtains the real-time occlusion area corresponding to each calibrated block in the real-time monitoring image. The occlusion area interval reference table of the corresponding time period is matched according to the acquisition time of the real-time monitoring image, and is used as the occlusion area interval comparison reference table. The real-time occlusion area corresponding to each calibrated block in the real-time monitoring image is compared with the occlusion area interval corresponding to each calibrated block in the occlusion area interval comparison reference table one by one, and the abnormal occlusion area block is marked. The specific method is as follows: The calibrated blocks whose real-time occlusion area does not belong to the corresponding occlusion area interval are marked as abnormal occlusion area blocks, otherwise no processing is performed.

[0022] The abnormal monitoring module obtains the real-time monitoring image of the drone, performs pre-division processing on it, and obtains the real-time occlusion area and real-time complexity coefficient of each calibration block. Then, according to the acquisition time of the real-time monitoring image, the occlusion area interval reference table and the boundary transformation interval reference table of the corresponding time period are matched as comparison references. The real-time occlusion area is compared with the interval in the occlusion area interval comparison reference table, and the real-time complexity coefficient is compared with the interval in the boundary transformation interval comparison reference table. If the real-time occlusion area is not within the corresponding occlusion area interval, it is marked as an abnormal occlusion area block; for the block marked as abnormal occlusion area, its real-time complexity coefficient is further compared. If it is not within the corresponding boundary transformation interval, it is marked as an abnormal occlusion boundary block, that is, an abnormal occlusion block.

[0023] Embodiment 2: As the embodiment 2 of the present invention, when the present application is specifically implemented, compared with embodiment 1, the difference between the technical scheme of this embodiment and embodiment 1 is that in this embodiment, the block marked as abnormal occlusion area is further analyzed, the real-time monitoring image is input into the boundary transformation interval reference table acquisition unit, and the complex coefficient of the occlusion boundary line corresponding to the block marked as abnormal occlusion area in the real-time monitoring image is analyzed, so as to obtain the real-time complex coefficient of the occlusion boundary line corresponding to the block marked as abnormal occlusion area in the real-time monitoring image; the boundary transformation interval reference table of the corresponding time period is matched according to the acquisition time of the real-time monitoring image and used as the boundary transformation interval comparison reference table, the real-time complex coefficient of the occlusion boundary line corresponding to the block marked as abnormal occlusion area in the real-time monitoring image is compared with the boundary transformation intervals corresponding to the corresponding calibration blocks in the boundary transformation interval comparison reference table one by one, the calibration block whose real-time complex coefficient does not belong to the corresponding boundary transformation interval is marked as an abnormal occlusion boundary block, and the corresponding calibration block is marked as an abnormal occlusion block and output, otherwise no processing is performed; By accurately calculating and analyzing the area, boundaries, and normal fluctuation range of the blocked area, the accuracy and sensitivity of abnormal judgment can be improved, and abnormal changes in the blocked area caused by abnormal accumulation of objects, sudden equipment failure, etc. can be discovered in time; once an abnormality is determined, staff can be immediately arranged to go to the blocked area for on-site inspection to check whether the abnormal change in the blocked area is caused by temporary stacking of objects, equipment failure, or other factors that may affect safety, and eliminate safety hazards in time to ensure that the monitoring area can function normally and ensure the construction safety of the construction site. It can also better grasp the changes in the monitoring blocked area of ​​the construction site, promptly discover the safety hazards hidden by abnormal changes, and prepare response measures in advance.

[0024] Embodiment 3: As the embodiment 3 of the present invention, when the present application is specifically implemented, compared with the embodiments 1 and 2, the technical solution of this embodiment is to combine the solutions of the above-mentioned embodiments 1 and 2 for implementation.

[0025] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and thresholds in the formula are set by technicians in this field according to actual conditions. At the same time, the contents not described in detail in this specification belong to the existing technology known to those skilled in the art.

[0026] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. An intelligent remote security monitoring system for construction projects, characterized in that: include: A monitoring period acquisition module obtains multiple monitoring periods corresponding to the construction site; The module for acquiring the image data packet to be analyzed acquires multiple regional images taken by the drone of the construction area in multiple monitoring periods in the preset number of days T2, and binds the multiple regional images to the corresponding monitoring periods one by one to form an image data packet calibrated with a time series. Here, T2 is a positive integer, satisfying 360≥T2≥30; An image processing module obtains a reference table of blocked area intervals and a reference table of boundary transformation intervals corresponding to the construction site in each monitoring period; The abnormality monitoring module is used to obtain the real-time monitoring image of the UAV, obtain the real-time occlusion area and real-time complexity coefficient corresponding to each calibrated block in the real-time monitoring image, and mark the occlusion abnormal blocks in the real-time monitoring image according to the occlusion area interval reference table and the boundary transformation interval reference table.

2. The intelligent remote security monitoring system for construction projects according to claim 1 is characterized in that: The specific method of obtaining multiple monitoring periods corresponding to the construction site is: The specified construction period of the construction site is evenly divided into multiple monitoring periods Ch according to the preset time interval T1, where T1 is the preset time interval, h is a different monitoring period, h=1, 2, ..., n, n is the number of monitoring periods, and n is a positive integer, satisfying n≥2.

3. The intelligent remote security monitoring system for construction projects according to claim 2 is characterized in that: The image processing module includes a pre-division processing unit, an occlusion area interval reference table acquisition unit and a boundary transformation interval reference table acquisition unit; A pre-division processing unit, used for evenly dividing each area image in the image data packets to be analyzed corresponding to the multiple monitoring time periods into a plurality of calibration blocks; The obstruction area interval reference table acquisition unit is used to analyze each calibration block of each regional image in the image data packet to be analyzed corresponding to each monitoring period, and obtain the obstruction area interval reference table corresponding to each monitoring period of the construction site; The boundary transformation interval reference table acquisition unit is used to analyze each calibration block of each regional image in the image data packet to be analyzed corresponding to each monitoring period, and obtain the boundary transformation interval reference table corresponding to each monitoring period of the construction site.

4. The intelligent remote security monitoring system for construction engineering according to claim 3 is characterized in that: The specific method of obtaining the reference table of the shielding area intervals corresponding to the construction site at each monitoring period is as follows: S1: Randomly select a monitoring period from each monitoring period as the target period; S2: Randomly select a calibration block from multiple calibration blocks as the target block; Obtain the occluded area Mr corresponding to the target blocks of each regional image in the target period image data packet, where r refers to the different regional images contained in the target period image data packet, and the value is r=1, 2, ..., a, where a refers to the number of regional images contained in the target period image data packet, and a is a positive integer and a>1; S3: Filter Mr according to the preset screening condition A, obtain the value Mb that meets the condition, and determine the standard occlusion area H1 according to the comparison result of the number of Mb and the preset threshold Y1, and generate the occlusion area interval L1 [FB, FA] according to the mean CZ of the absolute value of the difference between the standard occlusion area H1 and the mean Mp of Mr; S4: Obtain the occlusion area intervals Le corresponding to each calibration block of the construction site in the target period, bind the occlusion area intervals Le corresponding to each calibration block in the target period with the target period, and generate a reference table R1 of the occlusion area intervals corresponding to the construction site in the target period, where e refers to different calibration blocks in the regional image, and is the block number corresponding to each calibration block, and the value is e=1, 2, ..., c, where c refers to the number of calibration blocks, c is a positive integer and c>1; S5: Repeat the above steps S1-S4 to generate a reference table Rh of the shielding area intervals corresponding to the construction site in each monitoring period.

5. The intelligent remote security monitoring system for construction projects according to claim 4 is characterized in that: The specific method of generating the occlusion area interval L1 is: Obtain the mean CZ of the absolute values ​​of the differences between the areas of each occlusion region Mr and the mean Mp of Mr, and use the sum of the standard occlusion area area H1 and CZ corresponding to the calibrated block of the construction site within the target period as the upper limit FA of the occlusion area interval corresponding to the calibrated block of the construction site within the target period, and use the difference between the standard occlusion area area H1 and CZ corresponding to the calibrated block of the construction site within the target period as the lower limit FB of the occlusion area interval corresponding to the calibrated block of the construction site within the target period, thereby generating the occlusion area interval L1 [FB, FA] corresponding to the calibrated block of the construction site within the target period.

6. The intelligent remote security monitoring system for construction engineering according to claim 4 is characterized in that: The specific method for determining the standard occlusion area H1 is: Get the numerical value Mb of each occlusion area Mr that meets the preset filtering condition A: |Mr-Mp|≥Y2; wherein b is the number of numerical values ​​in Mr that meet the preset filtering condition A, a≥b≥1, when the number b is greater than the preset threshold value Y1, define the mean value Mp of Mr as the standard occlusion area H1 corresponding to the calibrated block of the construction site during the target period, when the number b is less than the preset threshold value Y1, define the mean value of the maximum and minimum values ​​in Mr as the standard occlusion area H1 corresponding to the calibrated block of the construction site during the target period.

7. The intelligent remote security monitoring system for construction projects according to claim 4 is characterized in that: The specific method of obtaining the boundary transformation interval reference table corresponding to each monitoring period of the construction site is as follows: S01: Select the same monitoring period as step S1 from each monitoring period as the target period; S02: Selecting the same calibrated block as step S2 from multiple calibrated blocks as the target block; S03: Obtain the complex coefficients Ur of the occlusion boundary lines corresponding to the target blocks of each regional image in the target period image data packet; generate the boundary transformation interval X1 according to the mean value Up of the complex coefficients Ur and the mean value Q of the absolute value of the difference between Ur and Up; S04: Repeat steps S02-S03 to obtain the boundary transformation intervals Xe corresponding to each calibrated block of the construction site in the target period, bind the boundary transformation intervals Xe corresponding to each calibrated block in the target period with the target period, and generate a boundary transformation interval reference table K1 corresponding to the construction site in the target period S04: Repeat the above steps S02-S03 to generate a boundary transformation interval reference table Kh corresponding to each monitoring period of the construction site.

8. The intelligent remote security monitoring system for construction projects according to claim 7 is characterized in that: The specific method of obtaining the complex coefficient Ur of the occlusion boundary line corresponding to the target blocks of each regional image in the target period image data packet is: S031: Select a target block from the target blocks of each regional image without replacement as the analysis block; obtain the occlusion boundary line corresponding to the occlusion area in the analysis block, construct a two-dimensional coordinate system with the center point of the analysis block as the origin, obtain the coordinates corresponding to each inflection point on the occlusion boundary line of the analysis block, and calculate the distance Gg between each inflection point on the occlusion boundary line of the analysis block and the center point of the analysis block according to the Euclidean distance formula; through the formula, , obtain the complex coefficient U1 of the occlusion boundary line of the analysis block, Gi is any one of Gg, where Gp is ​​the mean value of Gg, g refers to different inflection points on the occlusion boundary line of the analysis block, and the value is g=1, 2, ..., d, where d refers to the number of inflection points on the occlusion boundary line of the analysis block, d is a positive integer and d>1; S032: Repeat step S031 to obtain the complex coefficients Ur corresponding to the occlusion boundary lines of the target blocks of each area image in the target time period image data packet, obtain the mean Q of the absolute values ​​of the differences between each complex coefficient Ur and Up, and use the sum of Up and Q as the upper limit value BHA of the boundary transformation interval corresponding to the target block of the construction site in the target time period, and use the sum of Up and Q as the lower limit value BHB of the boundary transformation interval corresponding to the target block of the construction site in the target time period, and then the boundary transformation interval X1 [BA, BA] corresponding to the target block of the construction site in the target time period, where Up is the mean value of Ur.

9. The intelligent remote security monitoring system for construction engineering according to claim 8, characterized in that: The specific method of marking the abnormal occlusion blocks in the real-time monitoring image is as follows: The real-time occlusion area corresponding to each calibrated block in the real-time monitoring image is obtained, and the occlusion area interval reference table of the corresponding time period is matched according to the acquisition time of the real-time monitoring image, which is used as the occlusion area interval comparison reference table, and the real-time occlusion area corresponding to each calibrated block in the real-time monitoring image is compared with the occlusion area interval corresponding to each calibrated block in the occlusion area interval comparison reference table one by one, and the calibrated block whose real-time occlusion area does not belong to the corresponding occlusion area interval is marked as an abnormal occlusion area block, otherwise no processing is performed, and the corresponding occlusion area of ​​the block marked as abnormal occlusion area in the real-time monitoring image is obtained. The real-time complexity coefficient of the corresponding occlusion boundary line is matched according to the boundary transformation interval reference table of the corresponding time period according to the acquisition time of the real-time monitoring image, and it is used as the boundary transformation interval comparison reference table. The real-time complexity coefficient of the occlusion boundary line corresponding to the abnormal occlusion area block in the real-time monitoring image is compared with the boundary transformation interval corresponding to the corresponding calibration block in the boundary transformation interval comparison reference table one by one, and the calibration block whose real-time complexity coefficient does not belong to the corresponding boundary transformation interval is marked as an abnormal occlusion boundary block, and the corresponding calibration block is marked as an abnormal occlusion block for output, otherwise no processing is done.

Images