High-altitude operation safety monitoring method and system based on image recognition

Abstract

According to the aerial work safety monitoring method and system based on image recognition, through real-time image acquisition and processing, feature analysis is carried out in combination with the deep learning technology, potential risks in aerial work are effectively recognized, an alarm is given in time, corresponding measures are taken, and the safety management level of aerial work is remarkably improved. The description is only the summary of the patent, and the specific embodiment and the detailed operation process need to be adjusted according to the actual development and application environment. The protection range is based on the claims, and the specification and the drawings are used for explaining the claims.

Description

Technical Field

[0001] The present invention relates to the technical field of aerial work safety monitoring, and in particular to an aerial work safety monitoring method and system based on image recognition. Background Art

[0002] With the rapid development of modern industrial technology, aerial work is increasingly used in many industries such as construction, electricity, and communications. However, the aerial work environment is complex and changeable, with many potential safety hazards, posing a serious threat to the life safety of workers. Traditional safety monitoring methods mainly rely on manual inspections and fixed equipment monitoring, which have problems such as low efficiency, easy omissions, and poor real-time performance. Therefore, the development of an efficient and accurate aerial work safety monitoring method has important practical significance and application value. Summary of the invention

[0003] The present invention aims to provide a method and system for monitoring the safety of aerial work based on image recognition, which realizes intelligent monitoring and management of aerial work safety by collecting image information of the aerial work area in real time and using image recognition technology to monitor and analyze the movements of the operators and the surrounding environment.

[0004] The technical solution of the present invention includes: A method for monitoring the safety of aerial work based on image recognition comprises the following steps: (1) Real-time image information of the aerial work area is obtained through image acquisition equipment. These image acquisition devices can be high-definition cameras, cameras mounted on drones, etc., which can cover different angles and positions of the aerial work area.

[0005] (2) Preprocess the acquired image information, including denoising, contrast enhancement and other operations, to improve image quality and recognition accuracy. Preprocessing can eliminate interference information in the image and make key features more prominent.

[0006] (3) Use image recognition algorithms to extract and identify features from preprocessed images. These features include key features such as operators, operating equipment, and surrounding environment. Through deep learning algorithms, such as convolutional neural networks (CNNs), accurate identification and positioning of these features can be achieved.

[0007] (4) Based on the recognition results, conduct a safety analysis of the operator's actions and the surrounding environment. For example, it can determine whether the operator is wearing protective equipment such as a helmet and safety belt, whether the operating equipment is in normal working condition, and whether there are obstacles or danger sources in the surrounding environment. Through comparative analysis, potential safety hazards can be discovered in a timely manner.

[0008] (5) If there is a safety hazard, the alarm mechanism will be triggered. The alarm can be issued in various forms such as sound, light flash, wireless signal, etc., to promptly remind operators and managers to take corresponding measures. At the same time, the alarm information can also be connected to the central control equipment to achieve remote monitoring and management.

[0009] A high-altitude operation safety monitoring system based on image recognition, comprising: (1) Image acquisition equipment, used to obtain real-time image information of the aerial work area; (2) Image processing equipment, used for preprocessing and feature extraction of image information; (3) Image recognition equipment, used to identify and analyze the pre-processed images; (4) Alarm and response equipment, used to trigger alarms when safety hazards are discovered; (5) Central control equipment, used to control and manage the entire system.

[0010] The beneficial effects of the present invention are: High real-time performance: Real-time monitoring of the operation process is achieved by collecting image information of the high-altitude operation area in real time; High accuracy: Using advanced image recognition technology, it can accurately identify and locate key features of operators, equipment and the surrounding environment; High degree of intelligence: Through automatic analysis and judgment of the operator's actions and the safety of the surrounding environment, intelligent monitoring and management of high-altitude operation safety is achieved; Timely warning: Once a safety hazard is discovered, the alarm mechanism will be triggered immediately to remind operators and managers to take timely measures to effectively prevent accidents. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 It is a structural design diagram of the system of the present invention; Figure 2 is a structural diagram of the system of the present invention; Figure 3 is a flow chart of the method of the present invention; Figure 4 This is the test result diagram of the target detection algorithm; Figure 5 This is the test result diagram of the human posture recognition algorithm; As shown in the figure: Figure 1 This is a schematic diagram of the system structure of the present invention, showing the connection relationship and workflow between the image acquisition device, image processing device, image recognition device, alarm device and central control device.

[0012] Start-Stop Layer: Mainly includes the start and stop of the algorithm. The start-stop layer is the control center of the algorithm. By starting the script of the algorithm, it realizes the start and end of all algorithms in the system.

[0013] Data acquisition layer: mainly includes data from network cameras, local videos and pictures obtained by users. As the data source of the monitoring system, the data acquisition layer is the foundation of the entire system. The algorithm supports online and local detection, so the data sources are divided into online and local. Considering that when reviewing historical information, it may be necessary to analyze and judge historical videos frame by frame, the detection of local pictures and videos is added.

[0014] Communication layer: mainly includes interacting with the monitoring system to obtain detection information. According to the message detection information, the algorithm that actually needs to be started is obtained. Then all other related algorithms are stopped and the current algorithm is started.

[0015] Algorithm layer: mainly includes further research and analysis of the data obtained by the data acquisition layer. The processing and analysis functions of each algorithm module are realized through the image detection and human posture recognition algorithm in the system.

[0016] Alarm layer: uploads the violations detected and identified by the algorithm as alarms. The alarm layer mainly matches the results of the algorithm layer processing and analysis with the specified alarm type, determines the specific alarm type, and then uploads it.

[0017] Transport layer: uploads the violation identification alarm to the monitoring platform, sends the identification result to the server and sends the server address to the safety monitoring system. The transport layer receives the alarm information from the alarm layer and the identification information from the algorithm layer, and uploads it to the interface of the aerial work safety monitoring system through the network transmission protocol.

[0018] Display layer: mainly includes the alarm display and real-time detection result display of the security monitoring system, Web browsing after the recognition results are sent to the server, and local display of image detection and recognition results.

[0019] Figure 3 This is a flow chart of the method of the present invention, which describes in detail the various steps and operation procedures of the high-altitude work safety monitoring method based on image recognition.

[0020] Figure 4 This is the test result diagram of the target detection algorithm. Figure 5 This is a test result diagram of the human posture recognition algorithm, which demonstrates the actual effect of feature extraction and recognition of high-altitude working areas through image recognition technology. DETAILED DESCRIPTION

[0021] The specific implementation of the present invention is as follows: First, the image information of the aerial work area is acquired in real time through an image acquisition device (such as a high-definition camera) and transmitted to an image processing device. The image processing device pre-processes the image, including denoising, contrast enhancement and other operations, to improve image quality and recognition accuracy.

[0022] Next, the pre-processed image is transmitted to the image recognition device. The image recognition device uses deep learning algorithms (such as convolutional neural networks) to extract and identify features of the image. Through the trained model, the key features of the operator, operating equipment and surrounding environment can be accurately identified and located.

[0023] Then, based on the recognition results, the operator's actions and the surrounding environment are analyzed for safety. For example, it can determine whether the operator is wearing protective equipment such as a helmet and a safety belt, whether the operating equipment is in normal working condition, and whether there are obstacles or danger sources in the surrounding environment.

[0024] Finally, if the analysis results show that there is a safety hazard, the alarm device will immediately trigger an alarm. The alarm can be issued in the form of sound, light flash or wireless signal, reminding operators and managers to take timely measures to eliminate the hazard.

[0025] Practical application scenarios The high-altitude operation safety monitoring method and system based on image recognition of the present invention have a wide range of application scenarios, and several typical examples are listed below: Construction industry: On construction sites, aerial work is frequent and dangerous. By deploying the monitoring system of the present invention, the actions and surrounding environment of workers during aerial work can be monitored in real time, and violations such as not wearing safety helmets and safety belts, as well as possible obstacles or falling risks can be discovered in a timely manner, thereby effectively preventing accidents such as falling from heights and being hit by objects.

[0026] Power industry: During high-altitude operations such as power line inspection and tower maintenance, operators face multiple risks such as electric shock and falling. The monitoring system of the present invention can monitor the wearing of safety equipment by operators, the safety of the working environment, and the operating status of power equipment in real time to ensure the safety and controllability of the operation process.

[0027] Communications industry: In high-altitude operations such as communication base station construction and antenna installation, operators need to perform precise operations at high altitudes. The monitoring system of the present invention can monitor the operating process of operators in real time to determine whether there are risks such as improper operation or fatigue, and whether there are interferences or safety hazards in the surrounding environment, thereby ensuring the installation quality of communication equipment and the safety of operators.

[0028] The high-altitude operation safety monitoring method and system based on image recognition of the present invention realizes intelligent monitoring and management of high-altitude operation safety by real-time acquisition and analysis of image information of the high-altitude operation area. The implementation of the present invention will effectively improve the safety of high-altitude operations and reduce the accident rate, and has broad application prospects.

Claims

1. A method for monitoring the safety of aerial work based on image recognition, characterized in that: It includes obtaining real-time image information of the high-altitude working area through image acquisition equipment, extracting and identifying features of the image using image recognition technology, conducting safety analysis based on the recognition results, and triggering an alarm when safety hazards are found.

2. The method according to claim 1, wherein the feature Extraction and analysis use deep learning technology to improve recognition accuracy and efficiency.

3. The method according to claim 1 or 2, wherein the alarm and response device can be connected to the mobile device of the manager via a wireless network to realize remote alarm.

4. A high-altitude work safety monitoring system based on image recognition, including image acquisition equipment, image processing equipment, image recognition equipment, safety monitoring equipment, alarm and response equipment, data storage and management equipment and central control equipment. The devices work together to achieve real-time monitoring and management of high-altitude work safety.

Images