Video monitoring method and system based on construction equipment

Abstract

The invention discloses a video monitoring method based on construction equipment, and the method comprises the steps: carrying out the risk assessment of a vulnerable part of newly introduced equipment, configuring a monitoring node for the recognized vulnerable part, carrying out the real-time and intelligent monitoring of the operation state of the equipment, and uploading data to a background system through a wireless network. And the monitoring video data of the equipment is checked and stored in real time. A video monitoring system based on construction equipment comprises a data acquisition module, a data storage module, a data processing module, a communication module and a user interface module. According to the invention, the function of time-phased automatic video recording can be realized, independent monitoring is carried out on easily damaged points of each device, a more comprehensive and more intelligent monitoring means is provided through combination of fusion of various sensors, machine learning identification and a data processing technology, prediction and preventive maintenance of potential faults of the device can be realized, and the reliability of the device is improved. Therefore, the non-planned downtime of equipment is reduced, and the overall production efficiency of a construction site is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of video monitoring, and in particular to a video monitoring method and system based on construction equipment. Background Art

[0002] In the remote management equipment work scenario, managers often need to understand the operating status of mechanical equipment in order to conduct effective monitoring and management. Traditional monitoring methods have certain limitations. Managers can often only rely on the punch-in schedule of mechanical equipment to obtain information, and cannot fully understand the actual working status of the equipment. Managers sometimes need to go to the site in person for supervision, which consumes a lot of human resources and time costs.

[0003] Most existing video surveillance methods use a one-to-many live broadcast method. Although remote monitoring can be achieved to a certain extent, there is a significant disadvantage, that is, the monitoring data cannot be automatically saved. This means that once the live broadcast ends, the relevant monitoring data will be lost and subsequent analysis and review will not be possible. In addition, fixed-point monitoring is usually used to monitor the entire large scene. Although it can cover a wider range, it cannot accurately display the working details and vulnerable points of the equipment, making it difficult for managers to discover potential problems with the equipment in a timely manner, thus affecting the maintenance and maintenance of the equipment.

[0004] Therefore, a more comprehensive and automated monitoring system is needed to monitor the specific operating conditions of construction equipment in real time, especially to monitor the vulnerable points of the equipment individually to ensure the safety and efficient management of the equipment. Summary of the invention

[0005] In view of the defects of the prior art, the present invention provides a video monitoring method and system based on construction equipment. By individually monitoring each vulnerable point of the equipment, automatically recording and saving the monitoring content and uploading it to the background, the damaged points can be intelligently identified, and more comprehensive and accurate equipment operation data can be provided to management personnel.

[0006] To achieve the above purpose, the technical solution adopted by the present invention is as follows:

[0007] In a first aspect, the present invention provides a video monitoring method based on construction equipment, the method comprising the following steps:

[0008] S1, conduct risk assessment on the vulnerable parts of the newly introduced equipment, including analysis of the mechanical structure, working environment, historical failure data, etc. of the equipment, to determine the key parts of the equipment that are most likely to have problems during operation;

[0009] S2, configure a monitoring node for each identified vulnerable part to ensure that the vulnerable part information can be monitored in real time during the equipment introduction process; the monitoring node includes a high-resolution camera, temperature sensor, vibration sensor, etc., which are used to monitor the equipment status in multiple dimensions;

[0010] S3, deploy monitoring nodes in the factory or construction site to achieve real-time monitoring of the equipment operation status; the monitoring nodes are connected to the background system through a wireless network to ensure the real-time and accuracy of the data;

[0011] S4, reports the device introduction to the background system, activates all monitoring nodes, and uploads the real-time live broadcast address of each node; the system automatically assigns a unique identifier to each device for easy management and data matching;

[0012] S5, during the operation of the equipment, all video surveillance nodes automatically record and upload the video files of this period to the background database every hour; the video files use efficient compression coding to reduce storage space occupation;

[0013] S6, each monitoring node remains in standby mode at the same time, waiting for the real-time access request from the management end; the monitoring node adopts a low-power design and enters standby mode when not connected to save energy;

[0014] S7 pushes the real-time video stream to the live broadcast address to ensure that the management end can view the equipment operation status and vulnerable parts in real time when accessing the live broadcast address; the video stream uses an encrypted transmission protocol to ensure data security;

[0015] S8, damage point determination and reporting;

[0016] S81, each monitoring node on the device side takes a photo at a certain interval and sends the current device information to the background; the photo shooting frequency can be dynamically adjusted according to the device operation status to increase the monitoring density during high-risk periods;

[0017] S82, the backend performs optimization processing on the image, including noise reduction and sharpening, to ensure that the image quality is suitable for subsequent analysis;

[0018] S83, using a machine learning model to identify the tuned image and determine whether the device is damaged;

[0019] S84, if the device is identified as damaged, the image and device information are pushed to the management end, and the management personnel conduct a secondary inspection to determine whether the device is damaged and related subsequent operations;

[0020] S9, the management end checks the historical video records and current operating status of each device;

[0021] S10, when the equipment is withdrawn, a notification is sent to the background system and all camera monitoring nodes are closed; the system will automatically save all relevant data before closing the node to ensure that the information is not lost.

[0022] Furthermore, in step S81, pictures larger than 2M are processed by a resolution compression algorithm before being transmitted to the background; the compression algorithm is based on subsampling and chroma sampling to minimize the impact of compression on image quality.

[0023] Furthermore, in step S82, the image optimization process includes:

[0024] a. Noise reduction: Use a median filter to reduce noise and better preserve the edge and detail information of the image;

[0025] b. Sharpening: Use an image sharpening filter with an image color value brightening algorithm to enhance the edge and detail features of the image, making the image of the vulnerable point clearer;

[0026] c. Edge enhancement: Use an adaptive threshold algorithm to further enhance edge details in the image, making potential damage points more obvious.

[0027] Furthermore, in step S83, the machine learning model adopts an image recognition solution and performs a large amount of data training on vulnerable points of each type of equipment, so as to accurately identify damaged equipment. The model uses a multi-layer convolutional neural network structure and is pre-trained with a large amount of historical fault data to achieve adaptability to different equipment types and working environments. In addition, the model also integrates an anomaly detection algorithm to automatically discover potential risks through analysis of real-time data.

[0028] Furthermore, in step S10, the management terminal provides data visualization functions, including video thumbnail preview, key event marking, etc., to quickly find key monitoring content.

[0029] In a second aspect, the present invention also provides a video monitoring system based on construction equipment, including a data acquisition module, a data storage module, a data processing module, a communication module, and a user interface module.

[0030] Data acquisition module: used to collect construction equipment data information;

[0031] Data storage module: used to store equipment entry and exit information, video recording files, monitoring node status information, etc.;

[0032] Data processing module: including image processing and machine learning analysis functions, processing and analyzing the data collected from the device end;

[0033] Communication module: responsible for data transmission between the device side, the background and the management side, using encrypted communication protocols to ensure data security;

[0034] User interface module: used for management personnel to access and control the system, and to monitor and manage equipment through PC or mobile terminals.

[0035] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0036] The present invention can realize the function of automatic recording in different time periods, monitor each vulnerable point of the equipment separately, and upload the content to the server to achieve the effect of remote real-time monitoring and viewing historical recordings. Through the combination of multiple sensor fusion, machine learning recognition and data processing technology, it provides a more comprehensive and intelligent monitoring method, which can realize the prediction and preventive maintenance of potential equipment failures, thereby reducing the unplanned downtime of equipment and improving the overall production efficiency of the construction site. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 A schematic diagram of a flow chart of a video monitoring method based on construction equipment provided by the present invention;

[0038] Figure 2 A schematic diagram of a video monitoring system module based on construction equipment provided by the present invention. DETAILED DESCRIPTION

[0039] In this embodiment, the entire system is divided into three main parts: the device side, the background side, and the management side, which are described as follows:

[0040] Device-side process:

[0041] (1) Conduct a comprehensive vulnerable point assessment on all incoming equipment to identify key parts of the equipment that may cause problems during use; the vulnerable point assessment includes analysis of equipment historical data and material property assessment of vulnerable parts.

[0042] (2) A dedicated monitoring node is configured for each vulnerable point to facilitate real-time monitoring of the status of each vulnerable point; the monitoring node includes high-definition cameras, temperature and vibration sensors, etc., and the monitoring accuracy is improved through multimodal sensor fusion technology.

[0043] (3) Monitoring nodes are arranged in the factory or construction site to ensure that the working status of the equipment can be monitored in real time so that potential problems can be discovered and dealt with in a timely manner; the monitoring nodes adopt a modular design to facilitate quick installation and maintenance.

[0044] (4) After the equipment enters the site, the background system is notified and all monitoring nodes are started. At the same time, the live broadcast address of each node is uploaded so that the background and management end can view the equipment status in real time; each node is equipped with an independent power supply to ensure that the equipment can continue to work in the event of a power outage.

[0045] (5) During the operation of the equipment, all video nodes will automatically record video and automatically upload the video files within this time period to the background database every hour for historical data analysis and storage; the video data is stored in multiple copies to improve data reliability.

[0046] (6) Each monitoring node is on standby so that managers can view real-time monitoring videos at any time; the nodes have edge computing capabilities and can process part of the video data locally, reducing the processing load of the background server.

[0047] (7) Push the live video stream to the live address to ensure that when the management end accesses the live address, it can see the working status of the current equipment and the status of vulnerable points in real time; the video stream uses dynamic bandwidth adjustment technology to ensure a relatively smooth viewing experience even when the network conditions are poor.

[0048] (8) When the equipment is removed from the site, the background system is notified and all camera nodes are shut down to save resources and protect device privacy; the system automatically backs up all video and monitoring data before shutting down the node.

[0049] Backstage:

[0050] The background is responsible for data storage and management, including the entry and exit status of equipment, the live broadcast address of each monitoring node, the equipment's video files, etc.; these data will be transmitted to the management end for management personnel to view and analyze; the background is also responsible for data compression, backup and recovery to ensure system stability and data security.

[0051] Management process:

[0052] (1) Managers enter the system through the management page to monitor and manage the device status. The management page supports multi-user access and can set different permissions for different users to ensure data security.

[0053] (2) The management end can view the historical video records and current operating status of each device, making it easier to understand the device usage and historical problems; historical records can be filtered by multiple dimensions such as time and device type.

[0054] (3) When checking the historical operation status of the equipment, the system will automatically play the video content of the equipment in the corresponding period according to the time period of the video recording, which is convenient for the management personnel to analyze in detail; the management end supports functions such as fast forward, slow play, screenshot, etc., so as to conduct more detailed analysis.

[0055] (4) When checking the current operating status, the system will connect to the camera node and play the current monitoring content in real time. If the device has not entered the site, it will prompt that it has not entered the site. The system also provides an automatic alarm function, which will automatically send a notification to the relevant management personnel when an abnormality is detected.

[0056] Damaged point report:

[0057] (1) Each monitoring node on the device side takes photos regularly and sends the device information to the background. Images larger than 2M will be compressed to save resources. After compression, the image must still have sufficient clarity for fault analysis.

[0058] (2) The background optimizes the received images, uses a median filter to reduce noise, and enhances the images through sharpening and color brightening algorithms. The processed images are saved in the database for further analysis.

[0059] (3) Use a deep learning-based machine learning model to identify the tuned image and accurately determine whether the equipment is damaged; the model automatically records the recognition results and compares them with historical data to improve the accuracy of the judgment.

[0060] (4) If the equipment is found to be damaged, the image and equipment information will be pushed to the management end, and the management personnel will conduct a second confirmation and take corresponding measures. The management end can also generate a detailed report, including information such as the location of the damage point and the degree of damage, so that maintenance personnel can perform targeted maintenance operations.

Claims

1. A video monitoring method based on construction equipment, characterized in that: include: S1, conduct risk assessment on vulnerable parts of newly introduced equipment; S2, configure monitoring nodes for each vulnerable part of the equipment; S3, deploy monitoring nodes in the factory or construction site; S4, reports the device introduction status to the background system, activates all monitoring nodes, and uploads the real-time live broadcast address of each node. The system automatically assigns a unique identifier to each device; S5: During the operation of the equipment, all video surveillance nodes automatically record and upload the video files of this period to the background database every hour; S6, each monitoring node remains in standby state at the same time, waiting for the real-time access request from the management end; S7, pushes the real-time video stream to the live broadcast address, ensuring that the management end can view the equipment operation status and vulnerable parts in real time when accessing the live broadcast address; S8, damage point determination and reporting; S81, each monitoring node on the device side takes a photo at a certain interval and sends the current device information to the background; S82, the backend optimizes the image to ensure that the image quality is suitable for subsequent analysis; S83, using a machine learning model to identify the tuned image and determine whether the device is damaged; S84, if the device is identified as damaged, the image and device information are pushed to the management end, and the management personnel conduct a secondary inspection to determine whether the device is damaged and related subsequent operations; S9, the management end checks the historical video records and current operating status of each device; S10, when the equipment is withdrawn, a notification is sent to the background system to shut down all camera monitoring nodes.

2. A video monitoring method based on construction equipment according to claim 1, characterized in that: In step S5, the video file is encoded using efficient compression to reduce storage space usage.

3. A video monitoring method based on construction equipment according to claim 1, characterized in that: In the step S6, the monitoring node enters a standby mode when not connected to save energy.

4. The video monitoring method based on construction equipment according to claim 1, characterized in that: In step S7, the video stream uses an encrypted transmission protocol to ensure data security.

5. The video monitoring method based on construction equipment according to claim 1, characterized in that: In step S81, the photo taking frequency can be dynamically adjusted according to the operating status of the equipment to increase the monitoring density during high-risk periods.

6. A video monitoring method based on construction equipment according to claim 1, characterized in that: In the step S10, the background system will automatically save all relevant data before closing the node.

7. The video monitoring method based on construction equipment according to claim 1, characterized in that: In step S82, the tuning process includes: a. Noise reduction: Use a median filter to reduce noise and better preserve the edge and detail information of the image; b. Sharpening: Use an image sharpening filter with an image color value brightening algorithm to enhance the edge and detail features of the image, making the image of the vulnerable point clearer; c. Edge enhancement: Use an adaptive threshold algorithm to further enhance edge details in the image, making potential damage points more obvious.

8. A video monitoring system based on construction equipment, characterized in that: include: Data acquisition module, data storage module, data processing module, communication module, and user interface module.

Images