A construction site monitoring system and method based on 5G network architecture

Abstract

The present invention provides a construction site monitoring system and method based on a 5G network architecture, wherein the monitoring system includes a video camera and a monitoring background, wherein the monitoring background includes a video image processing module and a video image analysis module, wherein the video camera is connected to the monitoring background in communication, and the video image processing module is connected to the video image analysis module. The monitoring method specifically comprises first collecting video of the construction site in real time, wherein the video image processing module performs frame-by-frame segmentation of the video image in the video, and performs image block segmentation on each frame of the image, and then screens out the video image in which construction personnel exist, extracts the image block in which facial information exists in the video image, identifies the facial information through a facial recognition algorithm, compares the identification result with the facial information in the employee facial database, and issues an alarm when the comparison result is a mismatch. The present invention can identify the identities of all personnel at the construction site, and effectively prevent the entry of irrelevant personnel at the construction site.

Description

Technical Field

[0001] The present invention relates to the technical field of infrastructure site management, and in particular to an infrastructure site monitoring system and method based on a 5G network architecture. Background Art

[0002] With the continuous development of infrastructure construction, the personnel involved in the construction have become more complicated. When personnel unrelated to infrastructure construction enter the infrastructure construction site, it may cause material losses at the infrastructure construction site, and the safety of personnel entering the infrastructure construction site cannot be guaranteed. However, most of the existing technologies use gate recognition to identify the identities of people entering and leaving the infrastructure site, but gate recognition cannot completely prevent illegal intrusions, and there may be situations such as following in. Moreover, gate recognition cannot identify the identities of construction personnel who have already entered the infrastructure site, and the management personnel of the infrastructure site cannot grasp the identity information of each construction personnel at the infrastructure site. Summary of the invention

[0003] The purpose of the present invention is to overcome the shortcomings of the prior art and provide a real-time video intelligent recognition system and method based on 5G network architecture.

[0004] The purpose of the present invention is achieved through the following technical solutions:

[0005] A real-time video intelligent recognition method based on 5G network architecture includes the following steps:

[0006] Step 1: Use a video camera to collect video of the infrastructure construction site in real time. The video camera transmits the video of the infrastructure construction site to the video image processing module in the monitoring background through the 5G network.

[0007] Step 2: The video image processing module segments the video image in the video of the infrastructure site frame by frame, and segments each frame of the video image into N image blocks of the same size. The video image processing module sends each segmented frame of the video image to the video image analysis module;

[0008] Step 3: The video image analysis module selects the video images containing construction workers, selects one of the video images containing construction workers, and extracts the image block containing facial information in the selected video image. The video image analysis module then recognizes the facial information through a facial recognition algorithm.

[0009] In step 4, the video image analysis module compares the recognition result with the facial information in the employee facial information storage unit. If the comparison result is that the facial information of the construction worker does not match the facial information in the employee facial information storage unit, the monitoring background will issue an alarm and notify the infrastructure site management personnel to verify the identity of the construction worker; if the comparison result is that the facial information of the construction worker matches the facial information in the employee facial information storage unit, return to step 3 and re-select a video image containing a construction worker for facial information recognition until all video images containing construction workers have completed facial information recognition.

[0010] Furthermore, after completing the comparison between the face recognition result and the face information in the employee face database in step four, the video image analysis module also performs safety equipment wearing detection based on the face recognition result. The specific process of the safety equipment wearing detection is: select one of the video images with construction workers, capture the face contour area in the video image, and obtain the corresponding standard wearing area of ​​safety equipment in the video image based on the face contour area. The video image analysis module performs safety equipment identification on the standard wearing area of ​​safety equipment in the video image. If no safety equipment is detected in the standard wearing area of ​​safety equipment, the video image analysis module extracts the comparison result corresponding to the video image, obtains the employee information corresponding to the construction worker in the video image, and the monitoring background sends an alarm message to the safety equipment corresponding to the construction worker through the 5G network. The safety equipment issues an alarm after receiving the alarm message. If the safety equipment is detected in the standard wearing area of ​​safety equipment, a video image with construction workers is reselected for safety equipment wearing detection until all video images with construction workers complete the safety equipment wearing detection.

[0011] Furthermore, in step 4, when the facial information of the construction worker does not match the facial information in the employee facial information storage unit, the operating authority of each construction equipment in the infrastructure site is also restricted. Before the construction worker starts the construction equipment, the monitoring background also verifies the operating authority of the construction worker, obtains the image of the construction worker through the video camera at the started construction equipment, and the video image analysis module performs facial recognition on the image of the construction worker, and matches the facial recognition result with the employee facial database to obtain the employee information of the construction worker, and at the same time extracts the operating authority list of the started construction equipment, and matches the employee information of the construction worker with the operating authority list of the started construction equipment. If the construction worker is not on the operating authority list, the construction worker's operating authority is not passed, the construction equipment cannot be started, and an alarm is issued; if the construction worker is on the operating authority list, the construction worker's operating authority is passed, and the construction equipment is started.

[0012] Furthermore, when the comparison result in step four is that the facial information of the construction worker does not match the facial information in the employee facial information storage unit, the facial recognition result of each image block containing facial information is retrieved, and all video images corresponding to the construction workers whose facial information does not match the facial information in the employee facial information storage unit are screened out. The movement trajectory of the construction worker is obtained based on all the screened video images, and the movement trajectory of the construction worker is displayed in the monitoring background.

[0013] A construction site monitoring system based on a 5G network architecture includes a video camera and a monitoring background, the monitoring background includes a video image processing module and a video image analysis module, the video camera is set at the construction site, the video camera and the monitoring background are communicatively connected via a 5G network, the video image processing module is connected to the video image analysis module, the video camera is used to collect video images of the construction site, the video image processing module is used to divide the video images frame by frame and divide each frame of the video image into image blocks, the video image analysis module is used to perform face recognition and comparison based on the video image, and the video image analysis module is also used to perform safety equipment wearing detection.

[0014] Furthermore, the video camera includes a fixed camera, a control ball and a camera built into the construction equipment.

[0015] Furthermore, the monitoring background also includes a video storage module and an employee data storage module. The video storage module is connected to the video camera and is used to store video images captured by the video camera; the employee data storage module includes an employee facial information storage unit, an employee information storage unit and a construction equipment operation authority storage unit. The employee facial information storage unit, the employee information storage unit and the construction equipment operation authority storage unit are all connected to the video image analysis module. The employee facial information storage unit is used to provide facial information for the video image analysis module to perform facial recognition. The employee information storage unit is used to store the employee information of all construction personnel at the infrastructure site. The construction equipment operation authority storage unit is used to store the operation authority list corresponding to each construction equipment at the infrastructure site.

[0016] The beneficial effects of the present invention are:

[0017] Real-time and comprehensive monitoring of the infrastructure site. Through video monitoring, the identities of the personnel at the infrastructure site can be identified, and the identity information of all personnel at the infrastructure site can be obtained, which is convenient for the management of the infrastructure site. It can also issue an alarm in time when personnel unrelated to the infrastructure construction enter the infrastructure site to prevent material losses at the infrastructure site and reduce the risk of accidents. When personnel unrelated to the infrastructure construction enter the infrastructure site, the operating authority of the construction equipment is restricted to prevent illegal operation of the construction equipment and improve the safety of the construction process at the infrastructure site. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a structural schematic diagram of the present invention;

[0019] Figure 2 It is a schematic diagram of a process of the present invention;

[0020] Among them: 1. Video camera, 2. Monitoring background, 21. Video image processing module, 22. Video image analysis module, 23. Video storage module, 24. Employee data storage module, 241. Employee face information storage unit, 242. Employee information storage unit, 243. Construction equipment operation authority storage unit. DETAILED DESCRIPTION

[0021] The present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0022] Example:

[0023] A 5G network architecture infrastructure site monitoring system, such as Figure 1 As shown, it includes a video camera 1 and a monitoring background 2, the monitoring background 2 includes a video image processing module 21 and a video image analysis module 22, the video camera 1 is set at the infrastructure site, the video camera 1 and the monitoring background 2 are communicated and connected through a 5G network, the video image processing module 21 is connected to the video image analysis module 22, the video camera 1 is used to collect video images at the infrastructure site, the video image processing module 21 is used to divide the video image frame by frame and divide each frame of the video image into image blocks, the video image analysis module 22 is used to perform face recognition and comparison according to the video image, and the video image analysis module 22 is also used for safety equipment wearing detection.

[0024] The monitoring background 2 also includes a video storage module 23 and an employee data storage module 24. The video storage module 23 is connected to the video camera 1, and the video storage module 23 is used to store the video images collected by the video camera 1; the employee data storage module 24 includes an employee facial information storage unit 241, an employee information storage unit 242 and a construction equipment operation authority storage unit 243. The employee facial information storage unit 241, the employee information storage unit 242 and the construction equipment operation authority storage unit 243 are all connected to the video image analysis module 22. The employee facial information storage unit 241 is used to provide facial information for the video image analysis module 22 to perform facial recognition. The employee information storage unit 242 is used to store the employee information of all construction personnel at the infrastructure site. The construction equipment operation authority storage unit 243 is used to store the operation authority list corresponding to each construction equipment at the infrastructure site.

[0025] The employee facial information storage unit 241 stores the names of all construction workers at the infrastructure site and their corresponding facial information, and the employee information storage unit 242 stores the information of the construction workers including their names, types of work, numbers, and bound safety equipment numbers.

[0026] When an accident occurs, the source can be quickly traced through the video data in the video storage module 23. When the video is stored in the video storage module 23, the video is also time-stamped, which can effectively improve the efficiency in the subsequent tracing investigation process.

[0027] The video camera 1 includes a fixed camera, a control ball and a camera built into the construction equipment. The video camera 1 is also provided with a communication antenna, which is a 5DB antenna.

[0028] The built-in cameras of the construction equipment are also used to collect video images in real time, and monitor the status of personnel in the construction equipment and the working process of the construction equipment in real time. Taking the tower crane as an example, the built-in cameras of the construction equipment include the tower crane hook camera, the boom camera, the control room camera and the elevator car camera. The tower crane hook camera and the boom camera can obtain the working process of the tower crane in real time, and can quickly trace the source when an accident or failure occurs, thereby improving the processing efficiency. The control room camera can monitor the working status of the tower crane control personnel in real time. The elevator bridge car camera can monitor the number of operators to prevent the elevator from being overloaded, and ensure the safety of construction personnel and tower crane equipment.

[0029] A method for monitoring infrastructure sites based on 5G network architecture, such as Figure 2 As shown, the following steps are included:

[0030] The video camera 1 collects the video of the infrastructure site in real time, and the video camera 1 transmits the video of the infrastructure site to the video image processing module 21 in the monitoring background 2 through the 5G network;

[0031] Step 2: The video image processing module 21 segments the video image in the video of the infrastructure site frame by frame, and segments each frame of the video image into N image blocks of the same size. The video image processing module 21 sends each segmented video image frame to the video image analysis module 22;

[0032] Step 3: The video image analysis module 22 selects the video images containing construction workers, selects one of the video images containing construction workers, and extracts the image block containing facial information in the selected video image. The video image analysis module 22 then recognizes the facial information through a facial recognition algorithm.

[0033] Step 4: The video image analysis module 22 compares the recognition result with the facial information in the employee facial information storage unit 241. If the comparison result shows that the facial information of the construction worker does not match the facial information in the employee facial information storage unit 241, the monitoring background 2 issues an alarm and notifies the infrastructure site manager to verify the identity of the construction worker. If the comparison result shows that the facial information of the construction worker matches the facial information in the employee facial information storage unit 241, the system returns to step 3 and reselects a video image containing a construction worker for facial information recognition until all video images containing construction workers have completed facial information recognition. The method of notifying the infrastructure site manager includes text messages, voice broadcasts, etc.

[0034] By using the video camera 1 to monitor the construction site in real time, all construction workers on the construction site can be identified. And through face comparison, people who illegally enter the construction site can be found in time to reduce the risk of material loss or accidents. Dividing each frame of video image into image blocks and only analyzing and identifying image blocks with facial information can effectively reduce the amount of calculation and improve the efficiency of analysis and identification.

[0035] After completing the comparison between the face recognition result and the face information in the employee face database in step 4, the video image analysis module 22 also performs safety equipment wearing detection based on the face recognition result. The specific process of the safety equipment wearing detection is: select one of the video images with construction workers, capture the face contour area in the video image, and obtain the corresponding safety equipment standard wearing area in the video image based on the face contour area. The video image analysis module 22 performs safety equipment identification on the safety equipment standard wearing area in the video image. If no safety equipment is detected in the safety equipment standard wearing area, the video image analysis module 22 extracts the comparison result corresponding to the video image, obtains the employee information corresponding to the construction worker in the video image, and the monitoring background 2 sends an alarm message to the safety equipment corresponding to the construction worker through the 5G network. The safety equipment issues an alarm after receiving the alarm message. If the safety equipment is detected in the safety equipment standard wearing area, a video image with construction workers is reselected for safety equipment wearing detection until all video images with construction workers complete the safety equipment wearing detection.

[0036] The safety equipment is worn on the top of the construction worker's head. The size of the standard wearing area of ​​the safety equipment is determined according to the size of the face contour with a preset ratio. The facial feature points are obtained through the construction worker's face contour. Taking the ear as an example, the position of the ear is used as the lower boundary of the standard wearing area of ​​the safety equipment. Then, according to the size of the standard wearing area of ​​the safety equipment, the corresponding standard wearing area of ​​the safety equipment in the video image is determined. By detecting the wearing of the safety equipment, the construction workers can be reminded in time when the safety equipment is not worn correctly to ensure safety during the construction process. And the identity of the person who does not wear the safety equipment correctly can be identified, so as to accurately convey the alarm information.

[0037] In step 4, when the facial information of the construction worker does not match the facial information in the employee facial information storage unit 241, the operating authority of each construction equipment in the infrastructure site is also restricted. Before the construction worker starts the construction equipment, the monitoring background 2 also verifies the operating authority of the construction worker, obtains the image of the construction worker through the video camera 1 at the started construction equipment, and the video image analysis module 22 performs facial recognition on the image of the construction worker, and matches the facial recognition result with the employee facial database to obtain the employee information of the construction worker, and at the same time extracts the operating authority list of the started construction equipment, and matches the employee information of the construction worker with the operating authority list of the started construction equipment. If the construction worker is not on the operating authority list, the construction worker's operating authority is not passed, the construction equipment cannot be started, and an alarm is issued; if the construction worker is on the operating authority list, the construction worker's operating authority is passed, and the construction equipment is started.

[0038] When there are people who illegally break into the infrastructure site, prevent them from operating the construction equipment, restrict the operating authority of each construction equipment, and conduct an operating authority check before the construction equipment is started. If the check fails, the construction equipment will be shut down directly, reducing the risk of accidents and further reducing the risk of material loss.

[0039] When the comparison result in step four is that the facial information of the construction worker does not match the facial information in the employee facial information storage unit 241, the facial recognition result of each image block that contains facial information is also retrieved, and all video images corresponding to the construction workers whose facial information does not match the facial information in the employee facial information storage unit 241 are screened out. The movement trajectory of the construction worker is obtained based on all the screened video images, and the movement trajectory of the construction worker is displayed in the monitoring background 2.

[0040] For construction workers who do not break in illegally, their movements are also recorded, and when they enter a dangerous area, an alarm is sent to their corresponding safety equipment.

[0041] The above-described embodiment is only a preferred solution of the present invention and does not limit the present invention in any form. There are other variations and modifications without exceeding the technical solution described in the claims.

Claims

1. A method for infrastructure site monitoring based on 5G network architecture, characterized in that: The following steps are involved: Step 1: Use a video camera to collect video of the infrastructure site in real time. The video camera transmits the video of the infrastructure site to the video image processing module in the monitoring background through the 5G network. Step 2: The video image processing module segments the video image in the video of the infrastructure site frame by frame, and segments each frame of the video image into N image blocks of the same size. The video image processing module sends each segmented frame of the video image to the video image analysis module; Step 3: The video image analysis module selects the video images containing construction workers, selects one of the video images containing construction workers, and extracts the image block containing facial information in the selected video image. The video image analysis module then recognizes the facial information through a facial recognition algorithm. Step 4: The video image analysis module compares the recognition result with the information in the employee face information storage unit. If the comparison result shows that the face information of the construction worker does not match the information in the employee face information storage unit, the monitoring background issues an alarm and notifies the infrastructure site management personnel to verify the identity of the construction worker. If the comparison result shows that the face information of the construction worker matches the information in the employee face information storage unit, the module returns to step 3 and reselects a video image containing a construction worker for face information recognition until all video images containing construction workers have completed face information recognition. After completing the comparison between the face recognition result and the face information in the employee face database in step 4, the video image analysis module also performs safety equipment wearing detection based on the face recognition result. The specific process of the safety equipment wearing detection is as follows: select one of the video images with construction workers, intercept the face contour area in the video image, and obtain the corresponding safety equipment standard wearing area in the video image based on the face contour area. The video image analysis module performs safety equipment identification on the safety equipment standard wearing area in the video image. If no safety equipment is detected in the safety equipment standard wearing area, the video image analysis module extracts the comparison result corresponding to the video image, obtains the employee information corresponding to the construction worker in the video image, and the monitoring background sends an alarm message to the safety equipment corresponding to the construction worker through the 5G network. The safety equipment issues an alarm after receiving the alarm message. If the safety equipment is detected in the safety equipment standard wearing area, reselect a video image with construction workers for safety equipment wearing detection until all video images with construction workers complete the safety equipment wearing detection. The safety equipment is worn on the top of the construction workers' heads. The size of the standard wearing area of ​​the safety equipment is determined according to the size of the facial contour with a preset ratio. The facial feature points are obtained through the construction workers' facial contour as the lower boundary of the standard wearing area of ​​the safety equipment. Then, according to the size of the standard wearing area of ​​the safety equipment, the corresponding standard wearing area of ​​the safety equipment in the video image is determined.

2. According to claim 1, a method for infrastructure site monitoring based on 5G network architecture is characterized in that: In step 4, when the facial information of the construction worker does not match the information in the employee facial information storage unit, the operation authority of each construction equipment in the infrastructure site is also restricted. Before the construction worker starts the construction equipment, the monitoring background also verifies the operation authority of the construction worker, obtains the image of the construction worker through the video camera at the started construction equipment, and the video image analysis module performs facial recognition on the image of the construction worker, and matches the facial recognition result with the employee facial database to obtain the employee information of the construction worker. At the same time, the operation authority list of the started construction equipment is extracted, and the employee information of the construction worker is matched with the operation authority list of the started construction equipment. If the construction worker is not on the operation authority list, the construction worker's operation authority is not passed, the construction equipment cannot be started, and an alarm is issued; if the construction worker is on the operation authority list, the construction worker's operation authority is passed, and the construction equipment is started.

3. The infrastructure site monitoring method based on 5G network architecture according to claim 1 is characterized in that: When the comparison result in step four is that the facial information of the construction workers does not match the information in the employee facial information storage unit, the facial recognition results of each image block that contains facial information are also retrieved, and all video images corresponding to the construction workers that do not match the information in the employee facial information storage unit are screened out. The movement trajectory of the construction workers is obtained based on all the screened video images, and the movement trajectory of the construction workers is displayed in the monitoring background.

4. A construction site monitoring system based on a 5G network architecture, used to execute a construction site monitoring method based on a 5G network architecture as described in any one of claims 1 to 3, characterized in that: It includes a video camera and a monitoring background, the monitoring background includes a video image processing module and a video image analysis module, the video camera is set at the infrastructure site, the video camera and the monitoring background are communicated and connected via a 5G network, the video image processing module is connected to the video image analysis module, the video camera is used to collect video images at the infrastructure site, the video image processing module is used to divide the video image frame by frame and divide each frame of the video image into image blocks, the video image analysis module is used to perform face recognition and comparison based on the video image, and the video image analysis module is also used to perform safety equipment wearing detection.

5. The infrastructure site monitoring system based on 5G network architecture according to claim 4 is characterized in that: The video cameras include fixed cameras, surveillance balls and cameras built into construction equipment.

6. The infrastructure site monitoring system based on 5G network architecture according to claim 4 is characterized in that: The monitoring background also includes a video storage module and an employee data storage module. The video storage module is connected to the video camera and is used to store video images captured by the video camera. The employee data storage module includes an employee facial information storage unit, an employee information storage unit and a construction equipment operation authority storage unit. The employee facial information storage unit, the employee information storage unit and the construction equipment operation authority storage unit are all connected to the video image analysis module. The employee facial information storage unit is used to provide facial information for the video image analysis module to perform facial recognition. The employee information storage unit is used to store the employee information of all construction personnel at the infrastructure site. The construction equipment operation authority storage unit is used to store the operation authority list corresponding to each construction equipment at the infrastructure site.

Images