A method, system, equipment, and medium for fusing on-site monitoring and experimental data.

Abstract

This application relates to a method, system, equipment and medium for fusion of field monitoring and test data, and involves the field of multi-technical integrated technology, including: according to the detection area divided by each adjacent pile number, all-round monitoring of the construction area is realized, and the video data in the detection area provides a detailed record of the construction process; according to the priority, the video data of the detection area is allocated to the appropriate cache area to balance the cache load of video data upload; the pile foundation and human portrait in the video can be automatically identified through the Faster R-CNN algorithm. This application can avoid problems such as human dependence and human errors to the greatest extent, and try to ensure that the data is true and valid, and traceable.

Description

Technical Field

[0001] The present application relates to the field of multi-technical integrated technology, and in particular to a method, system, device and medium for fusion of on-site monitoring and test data. Background Art

[0002] In the process of engineering quality inspection, there are many problems in the process of obtaining comprehensive, true and effective on-site inspection data. Some on-site tests have realized automatic data collection and real-time wireless transmission, and some data generated during the test process can be understood through remote terminals. Therefore, this method is crucial to the quality and efficiency of on-site inspection.

[0003] At present, during the process of uploading on-site inspection videos, some inspection agencies use cameras to shoot on-site as an auxiliary supervision method. However, the video and inspection data are separated, and it is difficult to extract videos of key links. The camera only serves as a reminder, and it is impossible to achieve synchronous fusion, correlation and storage based on video surveillance and test data. In addition, there is a lack of supervision on key links, and the authenticity of the data needs to be verified.

[0004] In view of the problems arising from the above-mentioned prior art, a method, system, equipment and medium for fusing field monitoring and test data are now needed to solve the problem of how to synchronously fuse, associate and store video surveillance and test data to minimize human dependence and human errors. Summary of the invention

[0005] The first purpose of this application is to provide a method for fusing on-site monitoring and test data, which can avoid problems such as manpower dependence and human errors to the greatest extent, and try to ensure that the data is authentic, valid, and traceable.

[0006] In the first aspect, the present application provides a method for fusing field monitoring and test data, which adopts the following technical solution:

[0007] A method for fusing field monitoring and test data, comprising:

[0008] Acquire a number of video data, wherein the video data includes an image of a pile foundation;

[0009] Marking the pile foundations in the video data to obtain pile number markings;

[0010] According to the geographical location marked by the stake number, the area between each adjacent stake number is divided into a detection area;

[0011] Using the Faster R-CNN algorithm, annotate the human portrait in the detection area, annotate and locate the human portrait, and obtain the construction status of the human portrait, wherein the construction status includes clock-in information and construction status;

[0012] Determine the priority of the video data of the detection area according to the preset construction plan and the construction status marked by the portrait;

[0013] According to the priority arrangement order of the detection areas, the video data of each detection area is written into a cache area of ​​the memory arranged from large to small, and distributedly stored;

[0014] According to the movement of the portrait annotations in each detection area in the video data located in the distributed storage area, the movement information of multiple workers in the detection area is obtained;

[0015] A construction report is generated based on the movement information and construction status of the portrait annotations, and the construction report includes the clock-in information, movement information and construction status of each worker.

[0016] By adopting the above technical solution, the detection area is divided according to each adjacent pile number, which can monitor the construction area more comprehensively. At the same time, it can also trace the source according to the pile number marking to ensure the transparency and compliance of the construction process as much as possible. The Faster R-CNN algorithm can automatically detect the marking of pile foundations and portraits in the video, reduce dependence on manual marking, and avoid problems such as manpower dependence and human errors to the greatest extent, improve the efficiency and accuracy of data processing, and can also allocate the video data of the detection area to the appropriate cache area according to priority, balancing the cache load of video data upload.

[0017] In a preferred example, the present application can be further configured as follows: the Faster R-CNN algorithm is used to mark the human portrait in the detection area, the human portrait is marked and located, and the construction status of the human portrait is obtained, wherein the construction status includes clock-in information and the step of obtaining the clock-in information includes:

[0018] In a time period, the area at the junction of each detection area in the video data is divided into an entry and punch-in area;

[0019] If in one of the detection areas, the entry punching area only has one portrait mark, then the entry punching information that the portrait mark is located in the detection area is generated;

[0020] If in one of the detection areas, there are two portrait markings in the entry clock-in area, then the departure clock-in information of the portrait marking being located in the detection area is generated;

[0021] The entry clock-in information and the exit clock-in information are integrated into the clock-in information annotated with the portrait.

[0022] By adopting the above technical solution, video data is used to automatically identify the entry and exit of workers in the detection area, and punch-in information is automatically generated, which reduces the need for manual punch-in and improves the automation level of punch-in management. It can also accurately record the entry and exit time of workers in the detection area, provide accurate time data for construction management, and help improve construction efficiency and the accuracy of resource scheduling. Combining punch-in information with video data provides a detailed record of construction personnel entering and exiting the detection area, which is easy to trace and verify, and tries to ensure the transparency and compliance of the construction process.

[0023] In a preferred example, the present application may be further configured as follows: in the step of annotating the portrait in the detection area by the Faster R-CNN algorithm, annotating and locating the portrait, and obtaining the construction status of the portrait annotation, wherein the construction status includes clock-in information and the construction status, the step of obtaining the construction status includes:

[0024] Divide the area of ​​each video data screen except the entry punch-in area into a non-entry punch-in area;

[0025] The Faster R-CNN algorithm is used to train the model of the actions annotated with the portraits, and matches them with the behavior database in the preset construction plan to obtain the construction status of each portrait annotated in the current non-entry clock-in area.

[0026] By adopting the above technical solution, the video screen is divided into non-entry clocking-in areas, so that the activities of workers in the construction area can be fully monitored, and every corner of the construction process can be monitored, thereby improving the comprehensiveness and accuracy of construction monitoring. The Faster R-CNN algorithm can analyze the portrait annotations of workers in the non-entry clocking-in areas in real time, and by establishing a learning model, learn the construction actions of the portrait annotations, generate a construction action model, and then obtain the construction status of the workers according to the construction actions in the construction action model, providing real-time construction status information for construction management, which is conducive to real-time monitoring and adjustment of the construction progress.

[0027] In a preferred example, the present application may be further configured as follows: after the step of writing the video data of each detection area into a cache area of ​​a memory arranged from large to small according to the priority arrangement order of the detection areas and performing distributed storage, the step further includes:

[0028] The video data in the buffer area is quickly compressed by a VDMS compression algorithm;

[0029] According to the distributed storage architecture, the video data in the cache area is stored in a plurality of storage devices or servers in a dispersed manner;

[0030] According to the priority of the detection area in each time period, the video data is divided into hot data, warm data and cold data, the hot data and warm data are stored in a distributed database, and the cold data is stored in a distributed file system.

[0031] By adopting the above technical solutions, the VDMS compression algorithm can quickly compress video data, reduce storage space requirements, and improve storage efficiency. At the same time, it can maintain the quality of video data as much as possible, making the data authentic and effective. Distributed storage and data classification management can ensure the security and compliance of video data, improve the scalability and fault tolerance of the system, and reduce the occurrence of data leakage and illegal access.

[0032] In a preferred example, the present application can be further configured as follows: the step of dividing the video data into hot data, warm data and cold data according to the priority of the construction arrangement in each time period includes:

[0033] According to the construction situation, the non-entry clocking-in area is divided into a preparation area and a working area;

[0034] According to the construction conditions marked on each portrait in the non-entry clocking-in area, the video data of the working area in the detection area is divided into hot data, the video data of the entry clocking-in area in the detection area is divided into warm data, and the video data of the preparation work area in the detection area is divided into cold data.

[0035] By adopting the above technical solution, the video data of the working area is defined as hot data. When supervising, the data of the most active and frequently accessed working areas can get a quick response, which improves the efficiency of data retrieval and processing. The video data of the entry and exit area is defined as warm data. When it is necessary to query the entry and exit records of workers, the more active and frequently accessed entry and exit data can get a quicker response. By classifying and storing video data, storage resources can be reasonably allocated. Limited high-speed storage space can be used to store hot data and warm data, while cold data can be stored on lower-cost storage media, reducing storage costs.

[0036] In a preferred example, the present application can be further configured as follows: the step of obtaining movement information of multiple workers in the detection area according to the movement of the portrait annotations in each detection area in the video data located in the distributed storage area includes:

[0037] Obtaining the entry and exit clocking information of each of the portraits;

[0038] The movement of the portrait annotation is obtained by combining the geographical locations of the detection areas where the entry clock-in information and the exit clock-in information are located.

[0039] By adopting the above technical solution, the entry and exit information of each portrait can be obtained, so as to accurately record the entry and exit time of workers, improve the accuracy of personnel management, reduce human errors in the entry and exit of workers, facilitate traceability and verification, and ensure the transparency and compliance of the construction process as much as possible.

[0040] In a preferred example, the present application may be further configured as follows: after the step of generating a construction report according to the construction arrangement of multiple workers in the preset construction plan, wherein the worker trajectory report includes the punch-in record, movement trajectory and construction status of each worker, the following may also be included:

[0041] The construction report is uploaded to the blockchain network for further review and approval. The approved construction report is automatically stored on the blockchain through a smart contract and can be viewed, downloaded, printed, etc.;

[0042] By inputting the ID and uploading the construction report to be verified, the hash value of the on-chain file and the construction report to be verified is compared to determine whether the construction report to be verified has been tampered with, and the verification result is returned.

[0043] By adopting the above technical solution, the construction report is uploaded to the blockchain network. By utilizing the decentralized storage, data immutability and smart contract automated approval of the blockchain, efficient storage, secure management and real-time verification of the construction report can be achieved, thereby improving the efficiency and security of construction management, reducing human errors, and ensuring the authenticity and traceability of the data.

[0044] In the second aspect, the present application provides a field monitoring and test data fusion system, which adopts the following technical solution:

[0045] A field monitoring and test data fusion system, comprising:

[0046] Data acquisition module: the data acquisition module is used to acquire a number of video data, the video data including images of pile foundations;

[0047] Pile foundation marking module: the pile foundation marking module is used to mark the pile foundations in the video data and obtain the pile number marking;

[0048] Detection area division module: The detection area division module is used to divide the area between each adjacent stake number into a detection area according to the geographical location marked by the stake number;

[0049] Portrait marking module: The portrait marking module is used to mark the portraits in the detection area through the Faster R-CNN algorithm, mark and locate the portraits, and obtain the construction status of the portraits, which includes the clock-in information and the construction status;

[0050] Priority confirmation module: The priority confirmation module is used to confirm the priority of the video data of the detection area according to the preset construction plan and the construction situation marked by the portrait;

[0051] Cache area writing module: The cache area writing module is used to write the video data of each detection area into the cache area of ​​the memory arranged from large to small according to the priority arrangement order of the detection areas, and perform distributed storage;

[0052] Movement information acquisition module: the movement information acquisition module is used to obtain movement information of multiple workers in the detection area according to the movement of the portrait annotations in each detection area in the video data located in the distributed storage area;

[0053] Report generation module: The report generation module is used to generate a construction report based on the movement information and construction status of the portrait annotation, and the construction report includes the clock-in information, movement information and construction status of each worker.

[0054] In a third aspect, the present application provides an electronic device, which adopts the following technical solution:

[0055] An electronic device comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the above-mentioned method for fusion of field monitoring and test data are implemented.

[0056] In a fourth aspect, the present application provides a computer storage medium, the following technical solution:

[0057] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above-mentioned method for fusing field monitoring and test data are implemented.

[0058] In summary, this application has the following beneficial technical effects:

[0059] 1. This application realizes comprehensive monitoring of the construction area according to the detection area divided by each adjacent pile number. The video data in the detection area provides a detailed record of the construction process, which is convenient for subsequent tracing and verification, thereby ensuring the transparency and compliance of the construction process. In addition, according to the priority, the video data of the detection area is allocated to the appropriate cache area to balance the cache load of video data upload.

[0060] 2. This application can automatically identify pile foundations and portraits in videos through the Faster R-CNN algorithm, reducing the need for manual labeling, effectively avoiding various problems caused by manpower dependence and human errors, and improving the efficiency and accuracy of data processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 It is a flow chart of a method for fusing field monitoring and test data in one embodiment of the present application.

[0062] Figure 2 This is a flow chart of the steps for obtaining the punch-in information in step S4 in one embodiment of the present application.

[0063] Figure 3 It is a flow chart of the steps for obtaining the construction status in step S4 in one embodiment of the present application.

[0064] Figure 4 This is a flowchart of steps added after step S6 in one embodiment of the present application.

[0065] Figure 5 It is a sub-step flow chart of step S62 in one embodiment of the present application.

[0066] Figure 6 It is a sub-step flow chart of step S7 in one embodiment of the present application.

[0067] Figure 7 It is a sub-step flow chart of step S8 in one embodiment of the present application.

[0068] Figure 8 It is a structural diagram of a field monitoring and test data fusion system according to one embodiment of the present application.

[0069] Fig. 9 It is a principle block diagram of an electronic device in one embodiment of the present application.

[0070] Figure numerals: 1. Data acquisition module; 2. Pile foundation marking module; 3. Detection area division module; 4. Portrait marking module; 5. Priority confirmation module; 6. Cache area writing module; 7. Mobile information acquisition module; 8. Report generation module. DETAILED DESCRIPTION

[0071] The following is combined with Figure 1-9 This application is described in further detail.

[0072] refer to Figure 1 , a method for fusing field monitoring and test data, specifically comprising:

[0073] S1. Acquire some video data, where the video data includes images of pile foundations.

[0074] S2. Mark the pile foundations in a number of video data to obtain pile number markings.

[0075] Specifically, the video playback technology based on streaming media can be used through RTSP (Real time Streaming Protocol) or MMS (Microsoft Media Server) to meet the needs of real-time streaming transmission, which requires special network protocols and specific servers for video control. The Faster R-CNN algorithm is used in the streaming video data to select piles as pre-training targets according to specific needs, and the identified piles are labeled, and then the labeled piles are trained on the model.

[0076] S3. According to the geographical location marked by the stake numbers, the area between each adjacent stake number is divided into a detection area.

[0077] Specifically, dividing the inspection area according to each adjacent pile foundation is conducive to monitoring and management of the construction plan of each inspection area.

[0078] S4. Use the Faster R-CNN algorithm to mark and locate the portraits in the detection area, and obtain the construction status of the portrait annotations, including the clock-in information and construction status.

[0079] Specifically, the Faster R-CNN algorithm is used to select portraits as pre-training targets according to specific needs, and the portraits are labeled according to their recognized actions. Then, model training is performed on the construction conditions of the labeled portraits to identify the construction conditions corresponding to the portraits' actions, and gradually improve the recognition accuracy.

[0080] S5. According to the preset construction plan and the construction status marked by the portrait, the priority of the video data in the detection area is determined.

[0081] Specifically, for example, in the vertical compressive strength test of the static load test, including the construction conditions of the preparation work and the start of hoisting, the priority of the start of hoisting is set to the highest, and the priority of the preparation work is second.

[0082] S6. According to the priority arrangement order of the detection areas, the video data of each detection area is written into a cache area of ​​the memory arranged from large to small, and distributed storage is performed.

[0083] Specifically, the higher the priority, the larger the cache area written to the memory.

[0084] S7. Obtain movement information of multiple workers in the detection area according to the movement of the portrait annotations in each detection area in the video data located in the distributed storage area.

[0085] For example, the pile numbers of each adjacent pile foundation are integrated into a detection area number. For example, the detection area number corresponding to the inspection area between pile numbers 01 and 02 is 0102, and the detection area number corresponding to the inspection area between pile numbers 02 and 03 is 0203. The movement information is 0102-0203.

[0086] S8. Generate a construction report based on the movement information and construction status of the portrait annotation. The construction report includes each worker's clock-in information, movement information, and construction status.

[0087] In this embodiment, comprehensive monitoring of the construction area is achieved by dividing the detection area according to adjacent pile numbers, and the source of the problem can be traced through pile number marking, which is convenient for tracing and verification, thereby ensuring the transparency and compliance of the construction process.

[0088] The Faster R-CNN algorithm can automatically detect piles and portraits in videos, reducing reliance on manual annotation, thereby minimizing supervision errors caused by manpower dependence and human errors, and improving the efficiency and accuracy of data processing. In addition, the algorithm can also reasonably allocate video data in the detection area to the appropriate cache area according to priority to balance the cache load when uploading video data.

[0089] refer to Figure 2 Further, in one embodiment, in step S4, the step of obtaining the punch-in information is refined into the following sub-steps:

[0090] S40: Within a time period, the area at the junction of each detection area in the video data is divided into an entry and check-in area.

[0091] Specifically, in a rectangular video screen of video data, a rectangular pixel area at the edge of the video screen is divided into entry and exit punch-in areas, which are respectively located at the top, bottom, and both sides of the video screen.

[0092] S41. If in a detection area, there is only one portrait mark in the entry punching area, then the entry punching information of the portrait mark located in the detection area is generated.

[0093] Specifically, when a portrait annotation appears only once at the top, bottom, and both sides of the video screen, it is considered that the portrait annotation enters the detection area for the first time and is clocked in.

[0094] S42: If two portrait annotations appear in the entry clocking-in area in a detection area, then the departure clocking-in information is generated in which the portrait annotation is located in the detection area.

[0095] Specifically, the first time a portrait annotation appears at the top, bottom, and both sides of the video screen is when the portrait annotation enters the detection area and clocks in, and the second time is when the portrait annotation leaves the detection area and clocks in. After the portrait annotation appears twice, the number of appearances of the portrait annotation is re-recorded.

[0096] S43, integrating the entry clock-in information and the exit clock-in information into the clock-in information annotated with the portrait.

[0097] In this embodiment, the punch-in area of ​​the video screen of the video data is monitored, and the number of occurrences of the portrait annotations is recorded, which can reflect the entry and exit of the workers in the detection area, and automatically generate punch-in records, thereby improving the degree of automation of punch-in management. It can improve the accuracy of capturing the specific time points of workers entering and exiting the detection area, thereby improving construction efficiency and accurate resource scheduling, and can also effectively trace and verify subsequent inquiries about portrait annotations, which also helps to ensure the transparency and compliance of the construction process.

[0098] In addition, reference Figure 3 Further, in one embodiment, in step S4, the step of obtaining the construction status is refined into the following sub-steps:

[0099] S44, dividing the area of ​​each video data screen except the entry punch-in area into a non-entry punch-in area.

[0100] S45. The Faster R-CNN algorithm is used to train the model of the actions annotated with the portraits, and matches them with the behavior database in the preset construction plan to obtain the construction status of each portrait annotated in the current non-entry clock-in area.

[0101] In this embodiment, the edge of the video screen is divided into non-entry punch-in areas, which realizes comprehensive monitoring of the activities of workers in the construction area, and thus every detail of the construction process is under monitoring, thereby enhancing the comprehensiveness and accuracy of monitoring. The Faster R-CNN algorithm can analyze the portrait annotations of workers in the non-entry punch-in area in real time, and by building a learning model, absorb and learn the construction actions in the portrait annotations, and create a construction action model. Using the construction action model, the construction situation of the workers can be captured and analyzed in real time, providing real-time construction status information for construction management, which helps to realize real-time monitoring and timely adjustment of the construction progress.

[0102] In addition, reference Figure 4 Further, in one embodiment, after step S6, steps S60, S61, and S62 are added:

[0103] S60. Rapidly compress the video data in the cache area using a VDMS compression algorithm.

[0104] Specifically, the instant and uninterrupted ultra-large-scale video data storage method can significantly reduce the size of video files by quickly compressing videos through a VDMS compression algorithm.

[0105] S61. According to the distributed storage architecture, the video data in the cache area is stored in a plurality of storage devices or servers in a dispersed manner.

[0106] S62. According to the priority of the detection area in each time period, the video data is divided into hot data, warm data and cold data, the hot data and warm data are stored in a distributed database, and the cold data is stored in a distributed file system.

[0107] In this embodiment, the VDMS compression algorithm has an efficient compression capability, which can quickly reduce the volume of video data, thereby reducing the demand for storage space, not only improving storage efficiency, but also retaining the quality of video data as much as possible, ensuring the authenticity and validity of the data. In addition, through distributed storage and classified management of data, the security and compliance of video data can be enhanced, while improving the scalability and fault tolerance of the system, effectively reducing the risk of video data leakage and illegal access.

[0108] In addition, reference Figure 5 Further, in one embodiment, step S62 is further divided into the following sub-steps:

[0109] S620. According to the construction situation, the non-entry clocking-in area is divided into the preparation area and the working area.

[0110] S621. According to the construction conditions of each portrait in the non-entry clocking-in area, the video data of the working area in the detection area is divided into hot data, the video data of the entry clocking-in area in the detection area is divided into warm data, and the video data of the preparation work area in the detection area is divided into cold data.

[0111] In this embodiment, the video data of the work area is divided into hot data. The work area is the most active and frequently accessed work area during supervision. When querying the video data of the work area, a quick response can be obtained, so that key construction activities can be monitored in real time and any problems or changes in the construction can be quickly responded to, thereby improving the efficiency of data retrieval and processing. The video data of the entry and exit area is defined as warm data. When it is necessary to query the entry and exit records of workers, the more active and frequently accessed entry and exit data can get a faster response. The data of the preparation work area is usually queried and recorded after an abnormality occurs in the later stage, so it is processed as cold data. The video data is classified and stored, and storage resources are reasonably allocated. The limited high-speed storage space is used to store hot data and warm data, and the cold data is stored on a storage medium with lower cost, thereby reducing the storage cost.

[0112] In addition, reference Figure 6 Further, in one embodiment, step S7 is refined into the following sub-steps:

[0113] S70: Obtain entry clock-in information and exit clock-out information marked on each portrait.

[0114] S71. Acquire movement status of the portrait annotation based on the geographical locations of the detection areas where the entry clock-in information and the exit clock-in information are located.

[0115] In this embodiment, the clock-in information is associated with the geographic location of the detection area, and the extracted entry clock-in information is used as the starting point and the exit clock-in information is used as the end point. After being integrated into the movement situation, it is integrated with the corresponding portrait annotation to form a complete clock-in record, clarifying the clock-in behavior of each worker in a specific area, and obtaining the entry clock-in information and exit clock-in information of each portrait annotation. This can accurately record the entry and exit time of the workers, improve the accuracy of personnel management, reduce human errors in the entry and exit clocking of workers, facilitate traceability and verification, and try to ensure the transparency and compliance of the construction process.

[0116] In addition, reference Figure 7 Further, in one embodiment, step S8 is refined into the following sub-steps:

[0117] S80. Upload the construction report to the blockchain network for further review and approval. The approved construction report is automatically stored on the blockchain through a smart contract and can be viewed, downloaded, printed, etc.

[0118] S81. By inputting the ID and uploading the construction report to be verified, the hash value of the on-chain file and the construction report to be verified is compared to determine whether the construction report to be verified has been tampered with, and the verification result is returned.

[0119] In this embodiment, the construction report is uploaded to the blockchain network, and blockchain technology is used through distributed ledgers, consensus mechanisms, smart contracts, encryption and other core technologies to achieve decentralization, immutability, openness, transparency, traceability and collective maintainability.

[0120] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0121] The embodiment of the present application also provides a field monitoring and test data fusion system, which corresponds one-to-one to the field monitoring and test data fusion method in the embodiment.

[0122] refer to Figure 8, a field monitoring and test data fusion system includes: data acquisition module 1, pile foundation marking module 2, detection area division module 3, portrait marking module 4, priority confirmation module 5, cache area writing module 6, mobile information acquisition module 7, report generation module 8. The detailed description of each functional module is as follows:

[0123] Data acquisition module 1: Data acquisition module 1 is used to acquire a number of video data, the video data including images of pile foundations;

[0124] Pile foundation marking module 2: Pile foundation marking module 2 is used to mark the pile foundations in a number of video data and obtain the pile number marking;

[0125] Detection area division module 3: The detection area division module 3 is used to divide the area between each adjacent pile number into a detection area according to the geographical location marked by the pile number;

[0126] Portrait annotation module 4: Portrait annotation module 4 is used to annotate the portraits in the detection area through the Faster R-CNN algorithm, annotate and locate the portraits, and obtain the construction status of the portrait annotations, including the clock-in information and the construction status;

[0127] Priority confirmation module 5: The priority confirmation module 5 is used to confirm the priority of the video data of the detection area according to the preset construction plan and the construction situation of the portrait annotation;

[0128] Cache area writing module 6: The cache area writing module 6 is used to write the video data of each detection area into the cache area of ​​the memory arranged from large to small according to the priority arrangement order of the detection areas, and perform distributed storage;

[0129] Movement information acquisition module 7: the movement information acquisition module 7 is used to obtain movement information of multiple workers in the detection area according to the movement of the portrait annotations in each detection area in the video data located in the distributed storage area;

[0130] Report generation module 8: Report generation module 8 is used to generate a construction report based on the movement information and construction status of the portrait annotation. The construction report includes the clock-in information, movement information and construction status of each worker.

[0131] Among them, the data acquisition module 1 collects video data containing pile foundation images; the pile foundation annotation module 2 annotates the pile foundations in the video and obtains the pile number information; the detection area division module 3 divides the area between adjacent pile numbers into detection areas according to the geographical location of the pile numbers; the portrait annotation module 4 uses the Faster R-CNN algorithm to annotate and locate the portraits in the detection area and record the construction status of the workers, including the punch-in information and the construction status; the priority confirmation module 5 confirms the priority of the video data in the detection area according to the preset construction plan and the construction status of the portrait annotation; the cache area writing module 6 writes the video data into the cache areas of different memories according to the priority of the detection area and performs distributed storage; the mobile information acquisition module 7 analyzes the movement of the portraits in each detection area in the video data located in the distributed storage area to obtain the movement information of the workers; finally, the report generation module 8 integrates the movement information and construction status of the portrait annotation to generate a detailed construction report, which includes the punch-in information, movement trajectory and construction status of each worker, realizes the effective integration of on-site monitoring and test data, and improves the efficiency and accuracy of construction management.

[0132] For the specific definition of the field monitoring and test data fusion system, please refer to the definition of the field monitoring and test data fusion method in the context, which will not be repeated here. Each module in the above-mentioned field monitoring and test data fusion system can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the electronic device in the form of hardware, or can be stored in the memory of the electronic device in the form of software, so that the processor can call and execute the operations corresponding to the above modules. In one embodiment, an electronic device is provided, which is a user terminal. Reference Fig. 9 , the electronic device includes a processor, a memory, a network interface and a database connected through a system bus. Among them, the processor of the electronic device is used to provide computing and control capabilities. The memory of the electronic device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the electronic device is used to store a detection data table. The network interface of the electronic device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a method for fusion of on-site monitoring and test data is implemented.

[0133] In one embodiment, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program:

[0134] S1. Acquire some video data, where the video data includes images of pile foundations.

[0135] S2. Mark the pile foundations in a number of video data to obtain pile number markings.

[0136] S3. According to the geographical location marked by the stake numbers, the area between each adjacent stake number is divided into a detection area.

[0137] S4. Use the Faster R-CNN algorithm to mark and locate the portraits in the detection area, and obtain the construction status of the portrait annotations, including the clock-in information and construction status.

[0138] S5. According to the preset construction plan and the construction status marked by the portrait, the priority of the video data in the detection area is determined.

[0139] S6. According to the priority arrangement order of the detection areas, the video data of each detection area is written into a cache area of ​​the memory arranged from large to small, and distributed storage is performed.

[0140] S7. Obtain movement information of multiple workers in the detection area according to the movement of the portrait annotations in each detection area in the video data located in the distributed storage area.

[0141] S8. Generate a construction report based on the movement information and construction status of the portrait annotation. The construction report includes each worker's clock-in information, movement information, and construction status.

[0142] In one embodiment, in step S4, the sub-steps of obtaining the punch-in information include:

[0143] S40: Within a time period, the area at the junction of each detection area in the video data is divided into an entry and check-in area.

[0144] S41. If in a detection area, there is only one portrait mark in the entry punching area, then the entry punching information of the portrait mark located in the detection area is generated.

[0145] S42: If two portrait annotations appear in the entry clocking-in area in a detection area, then the departure clocking-in information is generated in which the portrait annotation is located in the detection area.

[0146] S43, integrating the entry clock-in information and the exit clock-in information into the clock-in information annotated with the portrait.

[0147] In one embodiment, in step S4, the sub-steps of obtaining the construction status include:

[0148] S44, dividing the area of ​​each video data screen except the entry punch-in area into a non-entry punch-in area.

[0149] S45. The Faster R-CNN algorithm is used to train the model of the actions annotated with the portraits, and matches them with the behavior database in the preset construction plan to obtain the construction status of each portrait annotated in the current non-entry clock-in area.

[0150] In one embodiment, the steps added after step S6 include:

[0151] S60. Rapidly compress the video data in the cache area using a VDMS compression algorithm.

[0152] S61. According to the distributed storage architecture, the video data in the cache area is stored in a plurality of storage devices or servers in a dispersed manner.

[0153] S62. According to the priority of the detection area in each time period, the video data is divided into hot data, warm data and cold data, the hot data and warm data are stored in a distributed database, and the cold data is stored in a distributed file system.

[0154] In one embodiment, the detailed sub-steps of step S62 include:

[0155] S620. According to the construction situation, the non-entry clocking-in area is divided into the preparation area and the working area.

[0156] S621. According to the construction conditions of each portrait in the non-entry clocking-in area, the video data of the working area in the detection area is divided into hot data, the video data of the entry clocking-in area in the detection area is divided into warm data, and the video data of the preparation work area in the detection area is divided into cold data.

[0157] In one embodiment, the sub-steps of step S7 include:

[0158] S70: Obtain entry clock-in information and exit clock-out information marked on each portrait.

[0159] S71. Obtain movement status of the portrait annotation based on the combination of the geographical locations of the detection areas of the entry clock-in information and the exit clock-in information.

[0160] In one embodiment, the sub-steps of step S8 include:

[0161] S80. Upload the construction report to the blockchain network for further review and approval. The approved construction report is automatically stored on the blockchain through a smart contract and can be viewed, downloaded, printed, etc.

[0162] S81. By inputting the ID and uploading the construction report to be verified, the hash value of the on-chain file and the construction report to be verified is compared to determine whether the construction report to be verified has been tampered with, and the verification result is returned.

[0163] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0164] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

Claims

1. A method for fusing field monitoring and test data, characterized in that: include: Acquire a number of video data, wherein the video data includes an image of a pile foundation; Marking the pile foundations in the video data to obtain pile number markings; According to the geographical location marked by the stake number, the area between each adjacent stake number is divided into a detection area; Using the Faster R-CNN algorithm, annotate the human portrait in the detection area, annotate and locate the human portrait, and obtain the construction status of the human portrait, wherein the construction status includes clock-in information and construction status; Determine the priority of the video data of the detection area according to the preset construction plan and the construction status marked by the portrait; According to the priority arrangement order of the detection areas, the video data of each detection area is written into a cache area of ​​a memory arranged from large to small, and distributedly stored; According to the movement of the portrait annotations in each detection area in the video data located in the distributed storage area, the movement information of multiple workers in the detection area is obtained; A construction report is generated according to the movement information and construction status of the portrait annotation, and the construction report includes the clock-in information, movement information and construction status of each worker.

2. The method according to claim 1, characterized in that In the step of marking the human portrait in the detection area by the Faster R-CNN algorithm, marking and locating the human portrait, and obtaining the construction status of the human portrait, wherein the construction status includes clock-in information and the construction status, the step of obtaining the clock-in information includes: In a time period, the area at the junction of each detection area in the video data is divided into an entry and punch-in area; If in one of the detection areas, the entry punching area only has one portrait mark, then the entry punching information that the portrait mark is located in the detection area is generated; If in one of the detection areas, there are two portrait markings in the entry clock-in area, then the departure clock-in information of the portrait marking being located in the detection area is generated; The entry clock-in information and the exit clock-in information are integrated into the clock-in information annotated with the portrait.

3. The method according to claim 2, characterized in that In the step of marking the portrait in the detection area by the Faster R-CNN algorithm, marking and locating the portrait, and obtaining the construction status of the portrait marking, wherein the construction status includes clock-in information and the construction status, the step of obtaining the construction status includes: Divide the area of ​​each video data screen except the entry punch-in area into a non-entry punch-in area; The Faster R-CNN algorithm is used to train the model of the actions annotated with the portraits, and matches them with the behavior database in the preset construction plan to obtain the construction status of each portrait annotated in the current non-entry clock-in area.

4. The method according to claim 3, characterized in that After the step of writing the video data of each detection area into a cache area of ​​a memory arranged from large to small according to the priority arrangement order of the detection areas and performing distributed storage, the method further includes: The video data in the buffer area is quickly compressed by a VDMS compression algorithm; According to the distributed storage architecture, the video data in the cache area is stored in a plurality of storage devices or servers in a dispersed manner; According to the priority of the detection area in each time period, the video data is divided into hot data, warm data and cold data, the hot data and warm data are stored in a distributed database, and the cold data is stored in a distributed file system.

5. The method according to claim 4, characterized in that The step of dividing the video data into hot data, warm data and cold data according to the priority of the construction arrangement in each time period comprises: According to the construction situation, the non-entry clocking-in area is divided into a preparation area and a working area; According to the construction conditions marked on each portrait in the non-entry clocking-in area, the video data of the working area in the detection area is divided into hot data, the video data of the entry clocking-in area in the detection area is divided into warm data, and the video data of the preparation work area in the detection area is divided into cold data.

6. The method according to claim 2, characterized in that The step of obtaining movement information of multiple workers in the detection area according to the movement of portrait annotations in each detection area in the video data located in the distributed storage area includes: Obtaining the entry and exit clocking information of each of the portraits; The movement of the portrait annotation is obtained by combining the geographical locations of the detection areas where the entry clock-in information and the exit clock-in information are located.

7. The method according to claim 1, characterized in that After the step of generating a construction report based on the movement information and construction status of the portrait annotation, wherein the construction report includes the clock-in information, movement information and construction status of each worker, the step further includes: The construction report is uploaded to the blockchain network for further review and approval. The approved construction report is automatically stored on the blockchain through a smart contract and can be viewed, downloaded and printed; By inputting the ID and uploading the construction report to be verified, the hash value of the on-chain file and the construction report to be verified is compared to determine whether the construction report to be verified has been tampered with, and the verification result is returned.

8. A field monitoring and test data fusion system, characterized in that: include: Data acquisition module (1): the data acquisition module (1) is used to acquire a number of video data, wherein the video data includes an image of a pile foundation; Pile foundation marking module (2): the pile foundation marking module (2) is used to mark the pile foundations in the video data and obtain the pile number marking; Detection area division module (3): the detection area division module (3) is used to divide the area between each adjacent pile number into a detection area according to the geographical location marked by the pile number; Portrait marking module (4): the portrait marking module (4) is used to mark the portraits in the detection area by using the Faster R-CNN algorithm, mark and locate the portraits, and obtain the construction status of the portraits, wherein the construction status includes check-in information and construction status; Priority confirmation module (5): the priority confirmation module (5) is used to confirm the priority of the video data of the detection area according to the preset construction plan and the construction situation marked by the portrait; Cache area writing module (6): the cache area writing module (6) is used to write the video data of each detection area into the cache area of ​​the memory arranged from large to small according to the priority arrangement order of the detection areas, and perform distributed storage; Movement information acquisition module (7): the movement information acquisition module (7) is used to acquire movement information of multiple workers in the detection area according to the movement of the portrait annotations in each detection area in the video data located in the distributed storage area; Report generation module (8): The report generation module (8) is used to generate a construction report based on the movement information and construction status marked by the portrait, and the construction report includes the clocking-in information, movement information and construction status of each worker.

9. An electronic device, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and executes a method for fusing field monitoring and test data as claimed in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer program is stored which can be loaded by a processor and executes the method for fusing field monitoring and test data according to any one of claims 1 to 7.

Images