Construction Behavior Recognition Method and System

Abstract

The present invention discloses a construction behavior recognition method and system, which periodically captures real-time remote sensing images of a work point through a camera device, determines the real-time construction progress of the work point based on the real-time remote sensing images, and then obtains the construction time of the current work point after a series of calculations. Finally, the calculated construction time is compared with the standard time of the work point in the work ticket information management system to obtain the construction behavior recognition result of the work point according to the comparison result, and to identify whether there is an abnormal construction behavior of working without a ticket and without a plan. The present invention improves the real-time monitoring of all work points, reduces the work burden of supervision personnel, and improves the efficiency and accuracy of abnormal construction behavior recognition.

Description

Technical Field

[0001] The present invention relates to the technical field of big data analysis, and in particular, to a method and system for identifying construction behaviors. Background Art

[0002] With the continuous advancement of high-intensity and large-scale power grid construction, safety issues have increased accordingly, and the challenges of safety supervision and early warning work have also increased. Among them, during the basic construction stage of line projects, unlicensed and unplanned operations are relatively frequent. Currently, the violations of unlicensed and unplanned operations are mainly monitored passively, and there are two common monitoring methods as follows: manual inspection, where a manual inspection team regularly inspects each construction site; sensor monitoring, where vibration sensors are installed at infrastructure construction sites to monitor the construction sites in real time. For manual inspection, due to a large number of engineering projects and scattered risk points, the manual inspection team may not be able to achieve full coverage and real-time monitoring of the project due to insufficient personnel, and the quality of the inspection personnel varies, which may lead to unstable inspection work quality and affect the accuracy of the inspection results. For sensor monitoring, vibration sensors may be relatively sensitive to environmental noise and may also generate false alarms for vibrations caused by non-violation operations. Moreover, widely deploying vibration sensors in power grid infrastructure construction may require high installation and maintenance costs. Therefore, in the face of the increasing demand for safety supervision at operation points, the existing technologies cannot comprehensively and effectively inspect all operation points in a timely manner, resulting in the inability to timely and accurately identify abnormal construction behaviors, and further leading to a series of safety risk problems caused by unlicensed and unplanned private operations. Summary of the Invention

[0003] The purpose of the embodiments of the present invention is to provide a method and system for identifying construction behaviors, which can comprehensively and effectively inspect operation points and improve the efficiency and accuracy of identifying abnormal construction behaviors.

[0004] To achieve the above purpose, the embodiments of the present invention provide a method for identifying construction behaviors, including:

[0005] Obtaining a real-time remote sensing image of an operation point by a camera device in the current detection period;

[0006] Determining the real-time construction progress of the operation point according to the real-time remote sensing image;

[0007] Calculating the construction progress difference between the real-time construction progress and the reference construction progress; wherein, the reference construction progress is the construction progress calculated in the previous detection period;

[0008] Calculating the construction duration according to the construction progress difference, a preset error coefficient, and a construction progress coefficient;

[0009] Determine the construction time of the current working point according to the construction duration and the reference time; wherein, the reference time is the shooting time of the remote sensing image in the previous detection period;

[0010] Compare the construction time with the standard time of the working point in the work ticket information management system, and obtain the construction behavior recognition result of the working point according to the comparison result.

[0011] As an improvement of the above solution, the determining the real-time construction progress of the working point according to the real-time remote sensing image includes:

[0012] Perform target recognition on the real-time remote sensing image;

[0013] Determine the real-time construction progress of the working point according to the pixel points of the recognized construction target.

[0014] As an improvement of the above solution, the calculating the construction duration according to the construction progress difference, the preset error coefficient and the construction progress coefficient includes:

[0015] Calculate the ratio of the construction progress difference to the construction progress coefficient;

[0016] Add the ratio to the preset error coefficient to obtain the construction duration.

[0017] As an improvement of the above solution, the determining method of the construction progress coefficient includes:

[0018] Obtain at least one construction influencing factor of the working point, and determine the construction influence degree value of each construction influencing factor and its corresponding influence factor coefficient;

[0019] Calculate the construction progress coefficient according to the preset basic construction duration, the construction influence degree value and the corresponding influence factor coefficient.

[0020] As an improvement of the above solution, the determining method of the error coefficient includes:

[0021] Determine the average time difference according to the historical construction duration and the historical standard duration; wherein, the historical construction duration is the construction duration calculated in the historical detection period, and the historical standard duration is the standard construction duration recorded in the work ticket information management system;

[0022] Superimpose the average time difference on the error coefficient calculated in the previous detection period to obtain the updated error coefficient.

[0023] As an improvement of the above solution, the determining the average time difference according to the historical construction duration and the historical standard duration includes:

[0024] Taking the historical construction duration and its corresponding historical standard duration as a set of time collections;

[0025] Calculating the differences between the historical construction durations and the historical standard durations in k time collections to obtain k time differences, where k≥1 and k is an integer;

[0026] Calculating the average time difference of the k time differences.

[0027] As an improvement to the above solution, obtaining the construction behavior recognition result of the operation point according to the comparison result includes:

[0028] When the comparison result is that the construction time matches the standard time, determining that the construction behavior is a normal construction behavior;

[0029] When the comparison result is that the construction time does not match the standard time, determining that the construction behavior is an abnormal construction behavior.

[0030] As an improvement to the above solution, the method further includes:

[0031] When an abnormal construction behavior is detected, sending a prompt message.

[0032] To achieve the above object, an embodiment of the present invention further provides a construction behavior recognition system, including:

[0033] A real-time remote sensing image acquisition module for acquiring a real-time remote sensing image of an operation point by a camera device in the current detection period;

[0034] A real-time construction progress determination module for determining the real-time construction progress of the operation point according to the real-time remote sensing image;

[0035] A construction progress difference calculation module for calculating the construction progress difference between the real-time construction progress and the reference construction progress; wherein, the reference construction progress is the construction progress calculated in the previous detection period;

[0036] A construction duration calculation module for calculating the construction duration according to the construction progress difference, a preset error coefficient, and a construction progress coefficient;

[0037] A construction time determination module for determining the construction time of the current operation point according to the construction duration and the reference time; wherein, the reference time is the shooting time of the remote sensing image in the previous detection period;

[0038] A construction behavior recognition module for comparing the construction time with the standard time of the operation point in the work ticket information management system and obtaining the construction behavior recognition result of the operation point according to the comparison result.

[0039] As an improvement to the above solution, the system further includes:

[0040] A prompt module for sending a prompt message when detecting an abnormal construction behavior.

[0041] Compared with the prior art, the construction behavior recognition method and system disclosed by the present invention periodically capture real-time remote sensing images of a working point through a camera device, determine the real-time construction progress of the working point based on the real-time remote sensing images, then obtain the construction time of the current working point after a series of calculations, and finally compare the calculated construction time with the standard time of the working point in the work ticket information management system to obtain the construction behavior recognition result of the working point according to the comparison result, and identify whether there is an abnormal construction behavior of working without a ticket or plan. The present invention improves the real-time monitoring of all working points, reduces the workload of supervisors, and improves the efficiency and accuracy of abnormal construction behavior recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 is a flowchart of a construction behavior recognition method provided by an embodiment of the present invention;

[0043] Figure 2 is a system framework diagram applied to the construction behavior recognition method provided by an embodiment of the present invention;

[0044] Figure 3 is another flowchart of a construction behavior recognition method provided by an embodiment of the present invention;

[0045] Figure 4 is a structural block diagram of a construction behavior recognition system provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0046] 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 only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0047] See Figure 1 , Figure 1 is a flowchart of a construction behavior recognition method provided by an embodiment of the present invention, and the construction behavior recognition method includes:

[0048] S1. Obtain a real-time remote sensing image of a working point in the current detection period by a camera device;

[0049] S2. Determine the real-time construction progress of the working point according to the real-time remote sensing image;

[0050] S3. Calculate the construction progress difference between the real-time construction progress and the reference construction progress, where the reference construction progress is the construction progress calculated in the previous detection cycle.

[0051] S4. Calculate the construction duration according to the construction progress difference, the preset error coefficient, and the construction progress coefficient.

[0052] S5. Determine the construction time of the current operation point according to the construction duration and the reference time, where the reference time is the shooting time of the remote sensing image in the previous detection cycle.

[0053] S6. Compare the construction time with the standard time of the operation point in the work ticket information management system, and obtain the construction behavior recognition result of the operation point according to the comparison result.

[0054] Exemplarily, see Figure 2 , Figure 2 which is the system framework diagram applied to the construction behavior recognition method provided by the embodiments of the present invention. The construction behavior recognition method described in the embodiments of the present invention is implemented by the safety management platform. The safety management platform and the camera device are pre-established with a communication connection. The camera device can be a near-earth satellite, a drone, etc. Based on its operating trajectory and the coordinate information of the operation point, the camera device periodically shoots the real-time remote sensing image of the operation point. The safety management platform provides the number and position of the operation points to be photographed in real time, updates the shooting rules, so that the camera device polls and shoots each target operation point one by one, and then sends the real-time remote sensing image to the safety management platform. The safety management platform provides computing and storage capabilities, and is used to deploy the target recognition and comparison algorithm based on the remote sensing image, process and analyze the real-time remote sensing image to obtain the construction time of the operation point. The safety management platform also interacts with the work ticket information management system. The work ticket information management system records detailed construction details (including construction personnel, construction materials, equipment, construction progress, pre-planned construction start time and construction end time, etc.). The safety management platform can identify whether there is abnormal construction behavior at the operation point by comparing the standard time provided by the work ticket information management system and the calculated construction time. In addition, based on the analysis of the remote sensing image, the real-time monitoring of all operation points is improved, the work burden of the supervision personnel is reduced, and the efficiency and accuracy of abnormal construction behavior recognition are improved.

[0055] See Figure 3 , Figure 3 which is another flow chart of a construction behavior recognition method provided by the embodiments of the present invention. The above steps S1-S6 are described in detail according to Figure 3 this.

[0056] Specifically, in step S1, obtain the real-time remote sensing image of the operation point by the camera device in the current detection cycle.

[0057] It should be noted that in order to ensure that abnormal construction behavior can be identified in a timely manner, the detection period can be set according to the actual construction situation, such as being set to 3 to 7 days. Of course, if the camera device is a low-Earth satellite, the detection period must also be determined based on the trajectory period of the low-Earth satellite. The security management platform can communicate with more than one low-Earth satellite. For example, a low-Earth satellite needs to pass through the work site every 7 days to take remote sensing images, but the security management platform expects to obtain a remote sensing image once within 3 days. At this time, the security management platform can use the remaining low-Earth satellites to obtain remote sensing images, thereby ensuring that the detection period can be shorter.

[0058] Specifically, in step S2, determining the real-time construction progress of the work point based on the real-time remote sensing image includes: performing target recognition on the real-time remote sensing image; and determining the real-time construction progress of the work point based on pixel points of the recognized construction target.

[0059] It should be noted that high-precision remote sensing images can be used to capture the overall construction picture of the work site with high definition, but the work site is often in a complex natural geographical environment, such as forests, rivers, etc., or it may be in a complex urban environment, surrounded by factories and other civil construction. In addition, various construction materials and construction equipment are often scattered in the work site. The above environmental factors cause certain difficulties in determining the construction target of the work site. The present invention will identify the construction target through a deep learning model based on remote sensing images, and correct the edge of the construction target by identifying the noise target in the surrounding environment of the construction target, thereby improving the recognition accuracy. For example, after acquiring the real-time remote sensing image, the real-time remote sensing image is used to identify the construction target based on the SAM segmentation model and the YOLOv8 multi-target recognition model.

[0060] For example, the real-time remote sensing image is processed by the SAM segmentation model, wherein the prompt part in the SAM segmentation model is updated by YOLOv8. After being processed by the SAM segmentation model, the construction target can be segmented from the complex and cluttered background image to obtain the initial construction target. , the initial construction target is actually the construction scope of the work point. In addition, after the real-time remote sensing image of the work point is processed by YOLOv8, typical construction materials, equipment, personnel, buildings, etc. in the image can be The influencing factors are divided one by one and expressed as , and thus fine-tune the initial construction goals After the loss function of the SAM segmentation model is processed, the edge-corrected construction target is obtained. , and then use the construction target The number of pixels determines the real-time construction progress of the work point 。

[0061] Exemplarily, assume that the remote sensing image seen is of a iron tower. After image recognition by the computer, based on the number of pixel points of the iron tower recognized in the real-time remote sensing image, compared with the expected complete iron tower data (data model, which can also be represented by pixel points), it is considered that its approximate completion degree is about 80%.

[0062] In the embodiment of the present invention, for the target operation point, the position of the high-precision remote sensing satellite is not fixed during each periodic shooting, so it is impossible to simply calculate the construction progress of the operation point. The present invention corrects the captured image according to the longitude and latitude where the imaging device is located and the construction comparison algorithm to obtain the actual construction progress for subsequent construction progress calculation.

[0063] Specifically, in step S3, calculate the construction progress difference between the real-time construction progress and the reference construction progress; wherein, the reference construction progress is the construction progress calculated in the previous detection cycle.

[0064] Exemplarily, obtain the reference construction progress calculated for this operation point in the previous detection cycle from the operation ticket information management system , and then based on the real-time construction progress calculated in this detection cycle , take the difference between the two to obtain the construction progress difference , the construction progress difference The calculation process satisfies the following formula:

[0065] (1).

[0066] Specifically, in step S4, calculating the construction duration according to the construction progress difference, a preset error coefficient, and a construction progress coefficient includes: calculating the ratio of the construction progress difference to the construction progress coefficient; adding the ratio to the preset error coefficient to obtain the construction duration.

[0067] Exemplarily, the calculation process of the construction duration satisfies the following formula:

[0068] (2).

[0069] Wherein, is the construction duration, is the construction progress coefficient, is the error coefficient.

[0070] Furthermore, the method for determining the construction progress coefficient includes: obtaining at least one construction influencing factor of the work point, and determining the construction impact degree value of each construction influencing factor and its corresponding influencing factor coefficient; calculating the construction progress coefficient according to a preset basic construction duration, the construction impact degree value and the corresponding influencing factor coefficient.

[0071] For example, the construction progress factor and The construction influencing factors include the actual construction personnel involved in the work order, the construction environment, the risk level of the construction unit, the weather conditions and other factors. Each construction influencing factor corresponds to a construction impact degree value. The construction impact value represents the positive impact on the construction process. The greater the construction impact, the greater the positive impact, which in turn leads to faster construction progress and increased construction progress. Optionally, the construction impact value can be obtained based on historical construction data analysis. In addition, each construction influencing factor also corresponds to an influencing factor coefficient, which is expressed as Characterization, the influencing factor coefficient is used to characterize the importance of each construction influencing factor. For example, if weather has the greatest impact on the construction progress, the corresponding influencing factor coefficient can be set to the maximum. The following are examples of several construction influencing factors:

[0072] 1) The number of construction workers is positively correlated with the corresponding construction impact value, because the more construction workers there are, the faster the construction efficiency is, and the greater the positive impact on the construction process is;

[0073] 2) The severity of the construction environment is inversely proportional to the corresponding construction impact value. The worse the construction environment (such as on hillsides, rock-covered areas, wetlands, etc.), the slower the construction efficiency, and the smaller the positive impact on the construction process (it can even be negative);

[0074] 3) The risk level of the construction unit is inversely proportional to the corresponding construction impact value. The higher the risk level, the greater the probability of problems occurring during the construction process, and the corresponding construction efficiency is slower. At this time, the positive impact on the construction process is smaller (it can even be negative);

[0075] 4) The severity of the weather is inversely proportional to its corresponding construction impact value. The worse the weather (such as heavy rain, typhoon, etc.), the slower the construction efficiency (even the construction may be suspended), and the smaller the positive impact on the construction process (it can even be negative).

[0076] Exemplarily, the embodiment of the present invention provides a construction basic unit schedule , the progress of the basic construction unit It can be the construction progress increment defined in advance by the staff after a unit time (such as 1 hour) within a construction period, or it can be calculated by the system itself based on historical construction data. In the present invention, by adding the value of the increase / decrease in construction progress caused by the above-mentioned construction influencing factors to the progress of the basic construction unit , the construction progress coefficient is finally obtained. Combining the above analysis, according to the preset basic construction duration, the construction influence degree value, and the corresponding influence factor coefficient, the construction progress coefficient is calculated, satisfying the following formula:

[0077] (3);

[0078] Among them, represents the th influence factor coefficient, represents the th construction influence degree value, .

[0079] Furthermore, the method for determining the error coefficient includes: determining the average time difference according to the historical construction duration and the historical standard duration; wherein, the historical construction duration is the construction duration calculated in the historical detection period, and the historical standard duration is the standard construction duration recorded in the work ticket information management system; adding the average time difference to the error coefficient calculated in the previous detection period to obtain the updated error coefficient.

[0080] Exemplarily, the determining the average time difference according to the historical construction duration and the historical standard duration includes: taking the historical construction duration and its corresponding historical standard duration as a set of time; calculating the difference between the historical construction duration and the historical standard duration in k sets of time, obtaining k time differences, k≥1 and k is an integer; calculating the average time difference of the k time differences. In the embodiment of the present invention, since the construction data is recorded after each construction is completed, the error coefficient can be obtained by comparing the difference between the actual construction duration (i.e., the standard duration) and the calculated construction duration in each detection period , the error coefficient is updated with the update of the detection period, and the calculation process of the error coefficient satisfies the following formula:

[0081] (4);

[0082] Among them, is the error coefficient calculated last time, is the historical standard duration in the th detection period, is the The historical construction duration in a detection period , is the number of detection periods and also the number of groups in the time set.

[0083] Specifically, in step S5, the construction time of the current working point is determined according to the construction duration and the reference time; wherein, the reference time is the shooting time of the remote sensing image in the previous detection period.

[0084] Exemplarily, the shooting of the remote sensing image is in a fixed cycle, but the construction is random. For example, on December 30th, the near-earth satellite passes through the airspace and takes the image of the day, but there is no construction from December 28th to December 30th. Therefore, the progress shown in the image taken on December 30th is the progress at the end of the construction on December 27th. Therefore, based on the time of the remote sensing image taken in the previous detection period, the construction time of this detection period is determined. From the construction progress, it can be inferred that the construction duration required for the construction progress difference between the two remote sensing images before and after is . After obtaining the time (year, month, day + specific time) of the remote sensing image taken by the camera device last time, according to the standard working duration per day (such as 8 hours) recorded in the work ticket information management system, the construction duration is calculated The ratio to the standard working duration is obtained to get the number of days. Then, combined with the time of the remote sensing image taken last time, the construction time within this detection period can be inferred. For example, the time of the remote sensing image taken last time is 21:30 on December 20th, and the construction duration is 24 hours, so the construction lasts for 3 days. If the detection period is 5 days, the construction time should be any three days between 21:30 on December 20th and 21:30 on December 24th.

[0085] Specifically, in step S6, the construction time is compared with the standard time of the working point in the work ticket information management system. When the comparison result shows that the construction time matches the standard time, it is determined that the construction behavior is a normal construction behavior; when the comparison result shows that the construction time does not match the standard time, it is determined that the construction behavior is an abnormal construction behavior.

[0086] Exemplarily, check whether there is a work ticket with the construction time of the same working point in the work ticket information management system. For example, if the detection period is 5 days, the time of the remote sensing image taken this time is 21:30 on December 24th. Check the work tickets from 21:30 on December 20th to 21:30 on December 24th in the work ticket information management system. If it is found that only 1 day, 2 days, 4 days or 5 days of construction are carried out within these 5 days, it is determined that there is an abnormal construction behavior in this detection period; conversely, if it is found that the expected construction is 3 days within these 5 days, it is determined that the construction behavior in this detection period is normal.

[0087] Further, after step S6 is executed, the method further includes: when an abnormal construction behavior is detected, a prompt message is sent.

[0088] Exemplarily, when an abnormal construction behavior is detected, it means that the construction time in this detection period cannot correspond to the expected time in the work ticket information management system, which may be caused by weather reasons, improper staff arrangement, out-of-scope construction, etc. The staff can readjust the construction arrangement.

[0089] Further, the real-time remote sensing images taken each time will be stored in the safety management platform, and will be associated with the current construction type, operation point, operation time, construction progress difference (construction duration) obtained compared with the previous remote sensing shooting time, associated work ticket id, etc. It is convenient for the staff to consult at any time and provides an indexing basis for the safety management platform.

[0090] Compared with the prior art, the construction behavior recognition method disclosed by the present invention has the following beneficial effects:

[0091] 1. By periodically taking real-time remote sensing images of the operation point through the camera device, the low-earth orbit near-earth satellite can periodically take pictures of the operation point at a high frequency, quickly obtain the real-time remote sensing images of the operation point, and timely grasp the construction progress and changes of the operation point. Compared with the traditional manual inspection or other monitoring methods, the monitoring efficiency and frequency are greatly improved.

[0092] 2. According to the real-time remote sensing images, the real-time construction progress of the operation point can be accurately determined. Through advanced image recognition technology and algorithms, the scope of the construction area can be accurately measured, providing an accurate data basis for the subsequent construction time calculation.

[0093] 3. Using data such as the real-time construction progress, through a series of scientific calculations, the construction time of the current operation point can be obtained. This calculation method is more accurate than the traditional estimation method and can provide a more reliable time basis for construction management.

[0094] 4. By comparing the calculated construction time with the standard time of the operation point in the work ticket information management system, it is possible to quickly identify whether there is an abnormal construction behavior of working without a ticket and without a plan. Once an abnormality is found, an early warning can be issued in a timely manner, facilitating relevant departments to take measures to stop and correct it, and avoiding potential safety hazards and economic losses caused by illegal construction behaviors.

[0095] 5. Through the real-time monitoring and recognition of construction behaviors, it helps to timely discover illegal operations and potential safety hazards during the construction process, such as out-of-scope construction, too fast or too slow construction progress, etc., thus ensuring the safety and quality of the construction.

[0096] 6. This monitoring method can provide comprehensive and accurate information for construction management, helping managers adjust the construction plan and resource allocation in a timely manner, improving the efficiency and scientific nature of construction management. At the same time, it can also reduce manual intervention and management costs, and achieve intelligent construction management.

[0097] See Figure 4 , Figure 4 which is a structural block diagram of a construction behavior recognition system 100 provided by an embodiment of the present invention. The construction behavior recognition system 100 includes:

[0098] A real-time remote sensing image acquisition module 11, configured to acquire a real-time remote sensing image of an operation point by a camera device in a current detection period;

[0099] A real-time construction progress determination module 12, configured to determine the real-time construction progress of the operation point according to the real-time remote sensing image;

[0100] A construction progress difference calculation module 13, configured to calculate a construction progress difference between the real-time construction progress and a reference construction progress; wherein, the reference construction progress is the construction progress calculated in the previous detection period;

[0101] A construction duration calculation module 14, configured to calculate a construction duration according to the construction progress difference, a preset error coefficient, and a construction progress coefficient;

[0102] A construction time determination module 15, configured to determine the construction time of the current operation point according to the construction duration and a reference time; wherein, the reference time is the shooting time of the remote sensing image in the previous detection period;

[0103] A construction behavior recognition module 16, configured to compare the construction time with a standard time of the operation point in a work ticket information management system, and obtain a construction behavior recognition result of the operation point according to a comparison result.

[0104] Specifically, the real-time construction progress determination module 12 is specifically configured to: perform target recognition on the real-time remote sensing image; determine the real-time construction progress of the operation point according to pixel points of the recognized construction target.

[0105] Specifically, the determining the real-time construction progress of the operation point according to pixel points of the recognized construction target includes: determining a target construction progress of the operation point in the real-time remote sensing image by using the number of pixel points of the construction target; calculating the real-time construction progress of the operation point according to position information of the camera device, position information of the operation point, and the target construction progress.

[0106] Specifically, the construction duration calculation module 14 is specifically configured to: calculate the ratio of the construction progress difference to the construction progress coefficient; add the ratio to a preset error coefficient to obtain the construction duration.

[0107] Specifically, the method for determining the construction progress coefficient includes: obtaining at least one construction influencing factor of the operation point, and determining the construction influence degree value and the corresponding influence factor coefficient of each construction influencing factor; calculating the construction progress coefficient according to a preset basic construction duration, the construction influence degree value, and the corresponding influence factor coefficient.

[0108] Specifically, the method for determining the error coefficient includes: determining an average time difference according to the historical construction duration and the historical standard duration; wherein, the historical construction duration is the construction duration calculated in the historical detection period, and the historical standard duration is the standard construction duration recorded in the work ticket information management system; adding the average time difference to the error coefficient calculated in the previous detection period to obtain an updated error coefficient.

[0109] Specifically, the determining of the average time difference according to the historical construction duration and the historical standard duration includes: taking the historical construction duration and its corresponding historical standard duration as a set of time; calculating the difference between the historical construction duration and the historical standard duration in k sets of time to obtain k time differences, where k≥1 and k is an integer; calculating the average time difference of the k time differences.

[0110] Specifically, the construction behavior recognition module 16 is specifically configured to: when the comparison result is that the construction time matches the standard time, determine that the construction behavior is a normal construction behavior; when the comparison result is that the construction time does not match the standard time, determine that the construction behavior is an abnormal construction behavior.

[0111] Specifically, the construction behavior recognition system 100 further includes:

[0112] A prompt module, configured to send a prompt message when an abnormal construction behavior is detected.

[0113] It should be noted that the working processes of the various modules in the construction behavior recognition system 100 described in the embodiments of the present invention may refer to the working process of the construction behavior recognition method described in the above embodiments, and will not be elaborated herein.

[0114] The above is the preferred embodiment of the present invention. It should be pointed out that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements are also regarded as the protection scope of the present invention.

Claims

1. A method for identifying construction behaviors, characterized in that, Including: Obtain the real-time remote sensing image of the operation point of the camera device in the current detection period; Determine the real-time construction progress of the operation point according to the real-time remote sensing image; Calculate the construction progress difference between the real-time construction progress and the reference construction progress; wherein, the reference construction progress is the construction progress calculated in the previous detection period; Calculate the construction duration according to the construction progress difference, the preset error coefficient and the construction progress coefficient; Determine the construction time of the current operation point according to the construction duration and the reference time; wherein, the reference time is the shooting time of the remote sensing image in the previous detection period; Compare the construction time with the standard time of the operation point in the work ticket information management system, and obtain the construction behavior recognition result of the operation point according to the comparison result; Wherein, the method for determining the construction progress coefficient includes: Obtain at least one construction influencing factor of the operation point, and determine the construction influence degree value and the corresponding influence factor coefficient of each construction influencing factor; Calculate the construction progress coefficient according to the preset basic construction duration, the construction influence degree value and the corresponding influence factor coefficient.

2. The construction behavior recognition method according to claim 1, characterized in that, The determining the real-time construction progress of the operation point according to the real-time remote sensing image includes: Perform target recognition on the real-time remote sensing image; Determine the real-time construction progress of the operation point according to the pixel points of the recognized construction target.

3. The construction behavior recognition method according to claim 1, characterized in that The calculating the construction duration according to the construction progress difference, the preset error coefficient and the construction progress coefficient includes: Calculate the ratio of the construction progress difference and the construction progress coefficient; Add the ratio to the preset error coefficient to obtain the construction duration.

4. The construction behavior recognition method according to claim 1 or 3, characterized in that, The method for determining the error coefficient includes: Determine the average time difference according to the historical construction duration and the historical standard duration; wherein, the historical construction duration is the construction duration calculated in the historical detection period, and the historical standard duration is the standard construction duration recorded in the work ticket information management system; Superimpose the average time difference on the error coefficient calculated in the previous detection period to obtain the updated error coefficient.

5. The construction behavior recognition method according to claim 4, wherein, The determining the average time difference according to the historical construction duration and the historical standard duration includes: Take the historical construction duration and its corresponding historical standard duration as a set of time; Calculate the difference between the historical construction duration and the historical standard duration in k sets of time to obtain k time differences, k≥1 and k is an integer; Calculate the average time difference of the k time differences.

6. The construction behavior recognition method according to claim 1, wherein The obtaining the construction behavior recognition result of the operation point according to the comparison result includes: When the comparison result is that the construction time matches the standard time, determine that the construction behavior is a normal construction behavior; When the comparison result is that the construction time does not match the standard time, determine that the construction behavior is an abnormal construction behavior.

7. The construction behavior recognition method according to claim 6, characterized in that, The method further includes: When an abnormal construction behavior is detected, send out a prompt message.

8. A construction behavior recognition system, characterized in that Including: A real-time remote sensing image acquisition module, configured to acquire the real-time remote sensing image of the operation point of the camera device in the current detection period; A real-time construction progress determination module, configured to determine the real-time construction progress of the operation point according to the real-time remote sensing image; The construction progress difference calculation module is used to calculate the construction progress difference between the real-time construction progress and the reference construction progress; wherein, the reference construction progress is the construction progress calculated in the previous detection period. The construction duration calculation module is used to calculate the construction duration according to the construction progress difference, a preset error coefficient, and a construction progress coefficient. The construction time determination module is used to determine the construction time of the current operation point according to the construction duration and the reference time; wherein, the reference time is the shooting time of the remote sensing image in the previous detection period. The construction behavior recognition module is used to compare the construction time with the standard time of the operation point in the work ticket information management system, and obtain the construction behavior recognition result of the operation point according to the comparison result. Wherein, the determination method of the construction progress coefficient includes: Obtain at least one construction influencing factor of the operation point, and determine the construction influence degree value and the corresponding influence factor coefficient of each construction influencing factor. Calculate the construction progress coefficient according to the preset basic construction duration, the construction influence degree value, and the corresponding influence factor coefficient.

9. The construction behavior recognition system according to claim 8, wherein, The system further includes: The prompt module is used to send a prompt message when an abnormal construction behavior is detected.

Images