Steel enterprise personnel violation identification management and control method and system based on AI analysis

Abstract

The invention provides a steel enterprise personnel violation identification control method and system based on AI analysis, and the method comprises the steps: carrying out the behavior tracking of each steel enterprise personnel in a real-time video stream when the steel enterprise personnel enter a monitoring region, and obtaining the dynamic behavior information corresponding to each steel enterprise personnel, the method comprises the following steps: performing deep identification on each piece of dynamic behavior information by utilizing an AI technology, screening illegal steel enterprise personnel, determining an illegal project corresponding to each illegal steel enterprise personnel, establishing an illegal reminding instruction corresponding to the illegal steel enterprise personnel based on the illegal project, and transmitting the illegal reminding instruction to a use terminal of the illegal steel enterprise personnel for reminding. According to the method, the violation duration and the violation item corresponding to each violation steel enterprise person are respectively counted, a violation report of a monitoring area is established and displayed, and technologies such as AI visual analysis, person real-time positioning and intelligent bracelets are comprehensively applied, so that a prevention and control means which can discover violation problems and can be more accurately pushed to people to ensure that violation behaviors stop is created.

Description

Technical Field

[0001] The present invention relates to the technical field of factory violation identification, and in particular to a steel enterprise personnel violation identification and control method and system based on AI analysis. Background Art

[0002] Steel enterprises refer to companies or factories that specialize in steel production and processing. These enterprises are mainly responsible for extracting steel from iron ore and carrying out subsequent processing and manufacturing. Steel is one of the basic materials of modern industry and is widely used in many fields such as construction, transportation, and energy. my country's steel enterprises play an important role in the global economy. Their products are key components of modern infrastructure and manufacturing. However, the work of steel enterprises is dangerous, so workers entering the factory must operate in a standardized manner. However, it is inevitable that some employees with weak awareness of prevention do not operate in a standardized manner. Therefore, many companies have made a series of monitoring work for this phenomenon, such as: human sampling monitoring through video, or using video recognition technology to identify violations, but existing technologies can often only play the role of discovery and recording, and there is a lag. It is impossible to automatically, promptly, and effectively remind the violator in time, and naturally there is no way to prevent and control this risk. In the end, although the violation was discovered, it still happened. Therefore, there is an urgent need for a technical means that can not only monitor and discover problems in real time, but also send information to the violator and his manager in a timely and accurate manner to prevent accidents.

[0003] Therefore, the present invention provides a method and system for identifying and controlling violations of steel enterprise personnel based on AI analysis. Summary of the invention

[0004] The present invention provides a steel enterprise personnel violation identification and control method and system based on AI analysis, which comprehensively uses AI visual analysis, real-time personnel positioning, smart bracelets and other technologies to create a prevention and control means that can not only detect violation problems, but also accurately push them to people to ensure that the illegal behavior stops.

[0005] The present invention provides a method for identifying and controlling steel enterprise personnel violations based on AI analysis, comprising:

[0006] Step 1: When steel enterprise personnel enter the monitoring area, the behavior of each steel enterprise personnel is tracked in the real-time video stream to obtain the dynamic behavior information corresponding to each steel enterprise personnel;

[0007] Step 2: Use AI technology to deeply identify each of the dynamic behavior information, screen out the steel company personnel who violate the rules, and determine the violation items corresponding to each of the steel company personnel who violate the rules;

[0008] Step 3: Based on the violation items, a violation reminder instruction is established for the corresponding steel enterprise personnel who violate the regulations and the instruction is transmitted to the user terminal of the steel enterprise personnel who violate the regulations for reminder;

[0009] Step 4: Count the violation duration and violation items corresponding to each of the steel enterprise personnel who violated the regulations, establish and display the violation report of the monitoring area.

[0010] In one practicable manner,

[0011] Also includes:

[0012] Each of the steel enterprise personnel is provided with a corresponding user terminal;

[0013] When the steel enterprise personnel enter the monitoring area, the steel enterprise personnel are reminded to wear and turn on the corresponding user terminal.

[0014] In one practicable manner,

[0015] The step 1 comprises:

[0016] Step 11: Monitor the monitoring area in real time to obtain a real-time video stream of the monitoring area. When the steel enterprise personnel enter the monitoring area, locate the entry position of the steel enterprise personnel in the real-time video stream to establish the RIO information of the monitoring area;

[0017] Step 12: When the steel enterprise personnel are active in the monitoring area, the activity characteristics corresponding to the steel enterprise personnel at different times are tracked in the real-time video stream, and the RIO information is updated using the activity characteristics to obtain the real-time RIO information of the steel enterprise personnel;

[0018] Step 13: Decompose the RIO information to obtain the posture presentation of the steel company personnel at different times and the position changes of the steel company personnel at different times, map the posture presentation to the position changes of the steel company personnel, and construct the dynamic behavior information of the steel company personnel.

[0019] In one practicable manner,

[0020] The step 2 comprises:

[0021] Step 21: Perform color training on the dynamic behavior information using AI technology to obtain RGB dynamic behavior information and D dynamic behavior information corresponding to the steel enterprise personnel, and perform layered training on the RGB dynamic behavior information using the AI ​​technology to obtain a plurality of monochrome information corresponding to each of the steel enterprise personnel;

[0022] Step 22: construct skeleton features corresponding to the steel enterprise personnel according to the D dynamic information, map each of the monochrome information to the corresponding skeleton features for color improvement training, and generate dynamic improvement behaviors corresponding to the steel enterprise personnel;

[0023] Step 23: Determine the appearance color distribution characteristics of the steel company personnel according to the monochrome information, determine the clothing violation information of the steel company personnel, use AI technology to respectively identify the behavior results corresponding to each of the dynamic improvement behaviors, determine the operation violation information of the steel company personnel according to the behavior results, respectively count the clothing violation information and operation violation information corresponding to each of the steel company personnel, and establish violation items corresponding to each of the steel company personnel.

[0024] In one practicable manner,

[0025] The step 22 comprises:

[0026] Step 221: respectively searching for the maximum dynamic pixel and the minimum dynamic pixel contained in each of the D dynamic information, establishing a pixel variation range of the D dynamic information based on the maximum dynamic pixel and the minimum dynamic pixel, normalizing the pixel variation range to generate a normalized pixel correspondence list corresponding to the D dynamic information, and using the normalized pixel correspondence list to perform pixel adjustment on the D dynamic information to obtain stable D dynamic information of each steel enterprise employee;

[0027] Step 222: performing contour detection in the stable D dynamic information to obtain a plurality of object contours, using the AI ​​technology to respectively identify each of the object contours, determining the appearance contour of the corresponding steel enterprise personnel, performing body classification on the appearance contour, and constructing the skeleton features of the corresponding steel enterprise personnel;

[0028] Step 223: Based on the feature specifications corresponding to the skeleton features, the corresponding monochrome information is adjusted and mapped to the skeleton features, and the monochrome information is merged with the skeleton features to obtain the color distribution information within the specified range of the steel company personnel, and obtain the dynamic improvement behavior of the corresponding steel company personnel.

[0029] In one practicable manner,

[0030] The step 3 comprises:

[0031] Step 31: respectively obtain the reminder method corresponding to each of the violation items, and establish a violation reminder instruction for each of the steel enterprise's violators according to the reminder method;

[0032] Step 32: Obtain basic information of the steel company personnel who violated the regulations, and determine the personality characteristics corresponding to each of the steel company personnel;

[0033] Step 33: The violation reminder instruction is personalized according to the personality characteristics, and the selected violation reminder instruction is transmitted to the user terminal of the corresponding steel enterprise personnel for reminder.

[0034] In one practicable manner,

[0035] The step 4 comprises:

[0036] Step 41: Obtain the violation items of each of the illegal steel enterprise personnel respectively, and establish the violation time distribution axis of the corresponding illegal steel enterprise personnel according to the violation start and end time corresponding to each of the illegal items;

[0037] Step 42: adding a corresponding frequency weight to the violation time distribution axis according to the violation frequency corresponding to each violation item, and adding a corresponding risk weight to the violation time distribution axis according to the risk level corresponding to each violation item;

[0038] Step 43: Find the violation duration corresponding to each violation item in the violation time distribution axis, and establish semantic violation description information of the corresponding steel enterprise personnel in combination with the corresponding frequency weight and danger weight;

[0039] Step 44: Generate a violation report for the monitoring area based on the semantic violation description information corresponding to all steel enterprise personnel who violated the regulations, and display it.

[0040] The present invention provides a steel enterprise personnel violation identification and control system based on AI analysis, comprising:

[0041] The behavior collection module is used to track the behavior of each steel enterprise personnel in the real-time video stream when the steel enterprise personnel enter the monitoring area, and obtain the dynamic behavior information corresponding to each steel enterprise personnel;

[0042] The violation identification module is used to use AI technology to deeply identify each of the dynamic behavior information, screen out the steel company personnel who violate the rules, and determine the violation items corresponding to each of the steel company personnel who violate the rules;

[0043] A reminder execution module is used to establish a violation reminder instruction for the corresponding steel enterprise personnel based on the violation items and transmit it to the user terminal of the steel enterprise personnel for reminder;

[0044] The violation supervision module is used to count the violation duration and violation items corresponding to each of the steel enterprise personnel who violated the regulations, establish and display the violation report of the monitoring area.

[0045] In one practicable manner,

[0046] Also includes:

[0047] The network layout module is used to set up several available networks in the monitoring area and control each of the user terminals to connect to the available networks.

[0048] In one practicable manner,

[0049] The violation identification module comprises:

[0050] The first recognition unit is used to perform color training on the dynamic behavior information using AI technology to obtain RGB dynamic behavior information and D dynamic behavior information corresponding to the steel enterprise personnel, and perform layered training on the RGB dynamic behavior information using the AI ​​technology to obtain a plurality of monochrome information corresponding to each of the steel enterprise personnel;

[0051] The second recognition unit is used to construct the skeleton features corresponding to the steel enterprise personnel according to the D dynamic information, and respectively map each of the monochrome information to the corresponding skeleton features for color improvement training to generate dynamic improvement behaviors corresponding to the steel enterprise personnel;

[0052] The third identification unit is used to determine the appearance color distribution characteristics of the corresponding steel company personnel based on the monochrome information, determine the clothing violation information of the corresponding steel company personnel, use AI technology to respectively identify the behavior results corresponding to each of the dynamic improvement behaviors, determine the operation violation information of the corresponding steel company personnel based on the behavior results, respectively count the clothing violation information and operation violation information corresponding to each of the steel company personnel, and establish violation items corresponding to each of the steel company personnel.

[0053] The achievable beneficial effects of the above technical solution are: in order to determine the legality of the behavior of each steel company personnel in real time, when the steel company personnel enter the monitoring area, their dynamic behavior information is determined by video tracking, and then AI technology is used to identify whether they have violated the rules, and corresponding prompts are given when they violate the rules. At the same time, all violations in the monitoring area are determined by real-time statistics, and a violation report is generated. The technology first analyzes the violation through the camera in the area where the violation occurred (with a built-in AI recognition algorithm for typical violation scenarios of steel companies), and then fixes the personnel information based on the personnel positioning information. The system background pushes the violation analysis results to his bracelet, and the bracelet informs the person at the first time through vibration reminder, thus forming an automated closed loop of discovering problems - background push - person receiving - stopping violations. Violation information will be pushed to relevant managers according to management requirements, and dual measures will ensure that violations are stopped in time.

[0054] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description and the accompanying drawings.

[0055] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0057] Figure 1 This is a schematic diagram of the workflow of a method for identifying and controlling steel enterprise personnel violations based on AI analysis in an embodiment of the present invention;

[0058] Figure 2 The present invention is a schematic diagram of the composition of a steel enterprise personnel violation identification and control system based on AI analysis in an embodiment of the present invention. DETAILED DESCRIPTION

[0059] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0060] Example 1

[0061] This embodiment provides a method and system for identifying and controlling violations of steel enterprise personnel based on AI analysis, such as Figure 1 As shown, including:

[0062] Step 1: When steel enterprise personnel enter the monitoring area, the behavior of each steel enterprise personnel is tracked in the real-time video stream to obtain the dynamic behavior information corresponding to each steel enterprise personnel;

[0063] Step 2: Use AI technology to deeply identify each of the dynamic behavior information, screen out the steel company personnel who violate the rules, and determine the violation items corresponding to each of the steel company personnel who violate the rules;

[0064] Step 3: Based on the violation items, a violation reminder instruction is established for the corresponding steel enterprise personnel who violate the regulations and the instruction is transmitted to the user terminal of the steel enterprise personnel who violate the regulations for reminder;

[0065] Step 4: Count the violation duration and violation items corresponding to each of the steel enterprise personnel who violated the regulations, establish and display the violation report of the monitoring area.

[0066] In this example, steel company personnel refers to the staff who enter the steel company's factory;

[0067] In this example, a camera is installed in the monitoring area, and everyone who enters the monitoring area has clearly entered the real-time monitoring area and is under monitoring;

[0068] In this example, the violations include: not wearing a safety helmet, not wearing work clothes, playing with mobile phones, smoking, falling, leaving the post, and other illegal operations;

[0069] In this example, the user terminal can be a wristband, and each steel company employee is equipped with a user terminal;

[0070] In this example, the violation report includes the violations of all steel company personnel in the monitoring area.

[0071] The working principle and beneficial effects of the above technical solution: In order to determine the legality of the behavior of each steel company employee in real time, when the steel company employee enters the monitoring area, his dynamic behavior information is determined by video tracking, and then AI technology is used to identify whether he has violated the rules, and corresponding prompts are given when he violates the rules. At the same time, all violations in the monitoring area are determined by real-time statistics, and a violation report is generated. The technology first analyzes the violation through the camera in the area where the violation occurred (with a built-in AI recognition algorithm for typical violation scenarios of steel companies), and then fixes the personnel information based on the personnel positioning information. The system background pushes the violation analysis results to his bracelet, and the bracelet informs the person at the first time through vibration reminder, thus forming an automated closed loop of discovering problems - background push - person receiving - stopping violations. Violation information will be pushed to relevant managers according to management requirements, and dual measures will ensure that violations are stopped in time.

[0072] Example 2

[0073] On the basis of Example 1, the method for identifying and controlling steel enterprise personnel violations based on AI analysis further includes:

[0074] Each of the steel enterprise personnel is provided with a corresponding user terminal;

[0075] When the steel enterprise personnel enter the monitoring area, the steel enterprise personnel are reminded to wear and turn on the corresponding user terminal.

[0076] The working principle and beneficial effects of the above technical solution are as follows: each steel enterprise employee is assigned a user terminal, which can be used to prompt him when he violates the rules.

[0077] Example 3

[0078] On the basis of Example 1, the method for identifying and controlling steel enterprise personnel violations based on AI analysis, step 1, comprises:

[0079] Step 11: Monitor the monitoring area in real time to obtain a real-time video stream of the monitoring area. When the steel enterprise personnel enter the monitoring area, locate the entry position of the steel enterprise personnel in the real-time video stream to establish the RIO information of the monitoring area;

[0080] Step 12: When the steel enterprise personnel are active in the monitoring area, the activity characteristics corresponding to the steel enterprise personnel at different times are tracked in the real-time video stream, and the RIO information is updated using the activity characteristics to obtain the real-time RIO information of the steel enterprise personnel;

[0081] Step 13: Decompose the RIO information to obtain the posture presentation of the steel company personnel at different times and the position changes of the steel company personnel at different times, map the posture presentation to the position changes of the steel company personnel, and construct the dynamic behavior information of the steel company personnel.

[0082] In this example, the RIO information indicates the information presented in the real-time video stream when the steel company personnel enter the real-time video stream;

[0083] In this example, the activity characteristics represent the characteristics of the steel company personnel's limb activities at different times;

[0084] In this example, real-time RIO information represents information about work activities generated by steel company personnel as they perform continuous activities in the monitoring area;

[0085] In this example, the posture presentation represents the external manifestation of steel company personnel at different moments;

[0086] In this example, the purpose of decomposition is to: decompose the posture and position changes at different moments;

[0087] In this example, dynamic behavior information refers to the information generated when steel company personnel perform activities.

[0088] The working principle and beneficial effects of the above technical solution: In order to reduce the capture delay and achieve synchronous detection, when the steel company personnel enter the monitoring area, the corresponding RIO information is determined in the real-time video stream, and the activity characteristics of the steel company personnel in the monitoring area are collected from this moment on, so as to update the RIO information and construct the real-time RIO information of the steel company personnel. Then, the posture and position changes of the steel company personnel are determined by decomposition and mapping, and finally the dynamic behavior information of the steel company personnel is constructed. In this way, not only real-time capture can be carried out, but also in-depth analysis of the behavior of the steel company personnel can be carried out during capture, thereby improving the accuracy and effectiveness of the dynamic behavior information.

[0089] Example 4

[0090] On the basis of Example 1, the method for identifying and controlling steel enterprise personnel violations based on AI analysis, step 2, comprises:

[0091] Step 21: Perform color training on the dynamic behavior information using AI technology to obtain RGB dynamic behavior information and D dynamic behavior information corresponding to the steel enterprise personnel, and perform layered training on the RGB dynamic behavior information using the AI ​​technology to obtain a plurality of monochrome information corresponding to each of the steel enterprise personnel;

[0092] Step 22: construct skeleton features corresponding to the steel enterprise personnel according to the D dynamic information, map each of the monochrome information to the corresponding skeleton features for color improvement training, and generate dynamic improvement behaviors corresponding to the steel enterprise personnel;

[0093] Step 23: Determine the appearance color distribution characteristics of the steel company personnel according to the monochrome information, determine the clothing violation information of the steel company personnel, use AI technology to respectively identify the behavior results corresponding to each of the dynamic improvement behaviors, determine the operation violation information of the steel company personnel according to the behavior results, respectively count the clothing violation information and operation violation information corresponding to each of the steel company personnel, and establish violation items corresponding to each of the steel company personnel.

[0094] In this example, RGB dynamic behavior information represents color dynamic behavior information, and D dynamic behavior information represents depth dynamic behavior information;

[0095] In this example, the monochrome information represents the dynamic behavior information about a single color formed by dividing the colors of the RGB dynamic behavior information after hierarchical training;

[0096] In this example, the skeleton features represent the limb contour features of the steel company personnel;

[0097] In this example, the dynamic improvement behavior represents the behavior obtained after distinguishing colors. The purpose of color separation is to identify helmets and clothing;

[0098] In this example, the operation violation information indicates the irregular behavior of the steel company personnel when performing work, and the clothing violation information indicates the irregular dressing behavior of the steel company personnel.

[0099] The working principle and beneficial effects of the above technical solution are as follows: the dynamic behavior information is color trained through AI technology, and the behavior information of steel company personnel is divided into monochrome information and D dynamic information. Then, the skeleton features of the steel company personnel are constructed according to the D dynamic information, and the skeleton features are improved by mapping, thereby generating the dynamic and perfect behavior of the steel company personnel. Then, the appearance color distribution characteristics and behavior results of the steel company personnel are analyzed to further determine whether the clothing and operation of the steel company personnel are in violation of regulations. Finally, the violation items of the steel company personnel are generated. In this way, the behavior of the steel company personnel can be analyzed from the outside to the inside to determine their violation items. In this way, the violations of the steel company personnel can be decomposed to improve the quality of violation identification.

[0100] Example 5

[0101] On the basis of Example 4, the method for identifying and controlling steel enterprise personnel violations based on AI analysis, step 22, comprises:

[0102] Step 221: respectively searching for the maximum dynamic pixel and the minimum dynamic pixel contained in each of the D dynamic information, establishing a pixel variation range of the D dynamic information based on the maximum dynamic pixel and the minimum dynamic pixel, normalizing the pixel variation range to generate a normalized pixel correspondence list corresponding to the D dynamic information, and using the normalized pixel correspondence list to perform pixel adjustment on the D dynamic information to obtain stable D dynamic information of each steel enterprise employee;

[0103] Step 222: performing contour detection in the stable D dynamic information to obtain a plurality of object contours, using the AI ​​technology to respectively identify each of the object contours, determining the appearance contour of the corresponding steel enterprise personnel, performing body classification on the appearance contour, and constructing the skeleton features of the corresponding steel enterprise personnel;

[0104] Step 223: Based on the feature specifications corresponding to the skeleton features, the corresponding monochrome information is adjusted and mapped to the skeleton features, and the monochrome information is merged with the skeleton features to obtain the color distribution information within the specified range of the steel company personnel, and obtain the dynamic improvement behavior of the corresponding steel company personnel.

[0105] In this example, the maximum dynamic pixel represents the maximum pixel in the D dynamic pixel information, and the minimum dynamic pixel represents the minimum pixel in the D dynamic pixel information;

[0106] In this example, the normalization process means converting the pixel value of the D dynamic information into the range of [0, 1]. The purpose of the normalization process is to balance the contrast of the D dynamic information and reduce interference pixels.

[0107] In this example, the object contour represents the outer contour of the object contained in the stable D dynamic information;

[0108] In this example, the process of fusing monochrome information with skeleton features represents the process of mapping, adjusting specifications, and fusing details of monochrome information;

[0109] In this example, the specified range represents a range with the steel company personnel as the center and the arm length as the radius. Since the arm lengths of personnel from different steel companies are different, the specified ranges of personnel from different steel companies are also different.

[0110] The working principle and beneficial effects of the above technical solution: In order to better improve the dynamic behavior of steel company personnel and ensure the quality of recognition work, the maximum dynamic pixel and the minimum dynamic pixel in the 3D dynamic information are first used to establish the pixel change range of the 3D dynamic information, and then the 3D dynamic information is adjusted by normalization processing to determine the stable 3D dynamic information of the steel company personnel, and then the contour of the object contained in the stable 3D dynamic information is determined by contour detection, and the appearance contour of the steel company personnel is identified therein, and the skeleton features of the steel company personnel are further constructed by limb classification, and then merged with the monochrome information to generate the dynamic and perfect behavior of the steel company personnel. In this way, the interference in the dynamic information can be eliminated, and the behavior of each steel company personnel is effectively analyzed and identified, which is convenient for better management and control.

[0111] Example 6

[0112] On the basis of Example 1, the method for identifying and controlling steel enterprise personnel violations based on AI analysis, step 3, comprises:

[0113] Step 31: respectively obtain the reminder method corresponding to each of the violation items, and establish a violation reminder instruction for each of the steel enterprise's violators according to the reminder method;

[0114] Step 32: Obtain basic information of the steel company personnel who violated the regulations, and determine the personality characteristics corresponding to each of the steel company personnel;

[0115] Step 33: The violation reminder instruction is personalized according to the personality characteristics, and the selected violation reminder instruction is transmitted to the user terminal of the corresponding steel enterprise personnel for reminder.

[0116] The working principle and beneficial effects of the above technical solution are as follows: When steel enterprise personnel violate regulations, corresponding reminder instructions are constructed according to the actual situation so as to give reminders through the use of terminals to reduce the risk of illegal operations.

[0117] Example 7

[0118] On the basis of Example 1, the method for identifying and controlling steel enterprise personnel violations based on AI analysis, step 4, comprises:

[0119] Step 41: Obtain the violation items of each of the illegal steel enterprise personnel respectively, and establish the violation time distribution axis of the corresponding illegal steel enterprise personnel according to the violation start and end time corresponding to each of the illegal items;

[0120] Step 42: adding a corresponding frequency weight to the violation time distribution axis according to the violation frequency corresponding to each violation item, and adding a corresponding risk weight to the violation time distribution axis according to the risk level corresponding to each violation item;

[0121] Step 43: Find the violation duration corresponding to each violation item in the violation time distribution axis, and establish semantic violation description information of the corresponding steel enterprise personnel in combination with the corresponding frequency weight and danger weight;

[0122] Step 44: Generate a violation report for the monitoring area based on the semantic violation description information corresponding to all steel enterprise personnel who violated the regulations, and display it.

[0123] In this example, the violation start and end times indicate the start and end times of a steel company employee's violation;

[0124] In this example, the violation time distribution axis includes the start and end times of all violation items of a steel company employee, and one steel company employee corresponds to one violation time distribution axis.

[0125] The working principle and beneficial effects of the above technical solution are as follows: the violations of steel company personnel are analyzed by constructing a violation time distribution axis, and then the semantic violation description information of steel company personnel is established to generate a violation report for the monitoring area, which is convenient for management personnel to carry out unified management.

[0126] Example 8

[0127] This embodiment provides a steel enterprise personnel violation identification and control system based on AI analysis, such as Figure 2 As shown, including:

[0128] The behavior collection module is used to track the behavior of each steel enterprise personnel in the real-time video stream when the steel enterprise personnel enter the monitoring area, and obtain the dynamic behavior information corresponding to each steel enterprise personnel;

[0129] The violation identification module is used to use AI technology to deeply identify each of the dynamic behavior information, screen out the steel company personnel who violate the rules, and determine the violation items corresponding to each of the steel company personnel who violate the rules;

[0130] A reminder execution module is used to establish a violation reminder instruction for the corresponding steel enterprise personnel based on the violation items and transmit it to the user terminal of the steel enterprise personnel for reminder;

[0131] The violation supervision module is used to count the violation duration and violation items corresponding to each of the steel enterprise personnel who violated the regulations, establish and display the violation report of the monitoring area.

[0132] In this example, steel company personnel refers to the staff who enter the steel company's factory;

[0133] In this example, a camera is installed in the monitoring area, and everyone who enters the monitoring area has clearly entered the real-time monitoring area and is under monitoring;

[0134] In this example, the violations include: not wearing a safety helmet, not wearing work clothes, playing with mobile phones, smoking, falling, leaving the post, and other illegal operations;

[0135] In this example, the user terminal can be a wristband, and each steel company employee is equipped with a user terminal;

[0136] In this example, the violation report includes the violations of all steel company personnel in the monitoring area.

[0137] The working principle and beneficial effects of the above technical solution: In order to determine the legality of the behavior of each steel company employee in real time, when the steel company employee enters the monitoring area, his dynamic behavior information is determined by video tracking, and then AI technology is used to identify whether he has violated the rules, and corresponding prompts are given when he violates the rules. At the same time, all violations in the monitoring area are determined by real-time statistics, and a violation report is generated. The technology first analyzes the violation through the camera in the area where the violation occurred (with a built-in AI recognition algorithm for typical violation scenarios of steel companies), and then fixes the personnel information based on the personnel positioning information. The system background pushes the violation analysis results to his bracelet, and the bracelet informs the person at the first time through vibration reminder, thus forming an automated closed loop of discovering problems - background push - person receiving - stopping violations. Violation information will be pushed to relevant managers according to management requirements, and dual measures will ensure that violations are stopped in time.

[0138] Example 9

[0139] On the basis of Example 8, the steel enterprise personnel violation identification and control system based on AI analysis further includes:

[0140] The network layout module is used to set up several available networks in the monitoring area and control each of the user terminals to connect to the available networks.

[0141] The working principle and beneficial effects of the above technical solution are as follows: Ultra-wideband (UWB), Bluetooth, Wi-Fi or other high-precision positioning technologies are used to achieve real-time tracking of the positions of personnel inside the workshop of a steel enterprise.

[0142] Example 10

[0143] On the basis of Example 8, the steel enterprise personnel violation identification and control system based on AI analysis, the violation identification module includes:

[0144] The first recognition unit is used to perform color training on the dynamic behavior information using AI technology to obtain RGB dynamic behavior information and D dynamic behavior information corresponding to the steel enterprise personnel, and perform layered training on the RGB dynamic behavior information using the AI ​​technology to obtain a plurality of monochrome information corresponding to each of the steel enterprise personnel;

[0145] The second recognition unit is used to construct the skeleton features corresponding to the steel enterprise personnel according to the D dynamic information, and respectively map each of the monochrome information to the corresponding skeleton features for color improvement training to generate dynamic improvement behaviors corresponding to the steel enterprise personnel;

[0146] The third identification unit is used to determine the appearance color distribution characteristics of the corresponding steel company personnel based on the monochrome information, determine the clothing violation information of the corresponding steel company personnel, use AI technology to respectively identify the behavior results corresponding to each of the dynamic improvement behaviors, determine the operation violation information of the corresponding steel company personnel based on the behavior results, respectively count the clothing violation information and operation violation information corresponding to each of the steel company personnel, and establish violation items corresponding to each of the steel company personnel.

[0147] In this example, RGB dynamic behavior information represents color dynamic behavior information, and D dynamic behavior information represents depth dynamic behavior information;

[0148] In this example, the monochrome information represents the dynamic behavior information about a single color formed by dividing the colors of the RGB dynamic behavior information after hierarchical training;

[0149] In this example, the skeleton features represent the limb contour features of the steel company personnel;

[0150] In this example, the dynamic improvement behavior represents the behavior obtained after distinguishing colors. The purpose of color separation is to identify helmets and clothing;

[0151] In this example, the operation violation information indicates the irregular behavior of the steel company personnel when performing work, and the clothing violation information indicates the irregular dressing behavior of the steel company personnel.

[0152] The working principle and beneficial effects of the above technical solution are as follows: the dynamic behavior information is color trained through AI technology, and the behavior information of steel company personnel is divided into monochrome information and D dynamic information. Then, the skeleton features of the steel company personnel are constructed according to the D dynamic information, and the skeleton features are improved by mapping, thereby generating the dynamic and perfect behavior of the steel company personnel. Then, the appearance color distribution characteristics and behavior results of the steel company personnel are analyzed to further determine whether the clothing and operation of the steel company personnel are in violation of regulations. Finally, the violation items of the steel company personnel are generated. In this way, the behavior of the steel company personnel can be analyzed from the outside to the inside to determine their violation items. In this way, the violations of the steel company personnel can be decomposed to improve the quality of violation identification.

[0153] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for identifying and controlling violations of steel enterprise personnel based on AI analysis, characterized in that: include: Step 1: When steel enterprise personnel enter the monitoring area, the behavior of each steel enterprise personnel is tracked in the real-time video stream to obtain the dynamic behavior information corresponding to each steel enterprise personnel; Step 2: Use AI technology to deeply identify each of the dynamic behavior information, screen out the steel company personnel who violate the rules, and determine the violation items corresponding to each of the steel company personnel who violate the rules; Step 3: Based on the violation items, a violation reminder instruction is established for the corresponding steel enterprise personnel who violate the regulations and the instruction is transmitted to the user terminal of the steel enterprise personnel who violate the regulations for reminder; Step 4: Count the violation duration and violation items corresponding to each of the steel enterprise personnel who violated the regulations, establish and display the violation report of the monitoring area.

2. A steel enterprise personnel violation identification and control method based on AI analysis as claimed in claim 1, characterized in that: Also includes: Each of the steel enterprise personnel is provided with a corresponding user terminal; When the steel enterprise personnel enter the monitoring area, the steel enterprise personnel are reminded to wear and turn on the corresponding user terminal.

3. A steel enterprise personnel violation identification and control method based on AI analysis as claimed in claim 1, characterized in that: The step 1 comprises: Step 11: Monitor the monitoring area in real time to obtain a real-time video stream of the monitoring area. When the steel enterprise personnel enter the monitoring area, locate the entry position of the steel enterprise personnel in the real-time video stream to establish the RIO information of the monitoring area; Step 12: When the steel enterprise personnel are active in the monitoring area, the activity characteristics corresponding to the steel enterprise personnel at different times are tracked in the real-time video stream, and the RIO information is updated using the activity characteristics to obtain the real-time RIO information of the steel enterprise personnel; Step 13: Decompose the RIO information to obtain the posture presentation of the steel company personnel at different times and the position changes of the steel company personnel at different times, map the posture presentation to the position changes of the steel company personnel, and construct the dynamic behavior information of the steel company personnel.

4. A steel enterprise personnel violation identification and control method based on AI analysis as claimed in claim 1, characterized in that: The step 2 comprises: Step 21: Perform color training on the dynamic behavior information using AI technology to obtain RGB dynamic behavior information and D dynamic behavior information corresponding to the steel enterprise personnel, and perform layered training on the RGB dynamic behavior information using the AI ​​technology to obtain a plurality of monochrome information corresponding to each of the steel enterprise personnel; Step 22: construct skeleton features corresponding to the steel enterprise personnel according to the D dynamic information, map each of the monochrome information to the corresponding skeleton features for color improvement training, and generate dynamic improvement behaviors corresponding to the steel enterprise personnel; Step 23: Determine the appearance color distribution characteristics of the steel company personnel according to the monochrome information, determine the clothing violation information of the steel company personnel, use AI technology to respectively identify the behavior results corresponding to each of the dynamic improvement behaviors, determine the operation violation information of the steel company personnel according to the behavior results, respectively count the clothing violation information and operation violation information corresponding to each of the steel company personnel, and establish violation items corresponding to each of the steel company personnel.

5. A steel enterprise personnel violation identification and control method based on AI analysis as claimed in claim 4, characterized in that: The step 22 comprises: Step 221: respectively searching for the maximum dynamic pixel and the minimum dynamic pixel contained in each of the D dynamic information, establishing a pixel variation range of the D dynamic information based on the maximum dynamic pixel and the minimum dynamic pixel, normalizing the pixel variation range to generate a normalized pixel correspondence list corresponding to the D dynamic information, and using the normalized pixel correspondence list to perform pixel adjustment on the D dynamic information to obtain stable D dynamic information of each steel enterprise employee; Step 222: performing contour detection in the stable D dynamic information to obtain a plurality of object contours, using the AI ​​technology to respectively identify each of the object contours, determining the appearance contour of the corresponding steel enterprise personnel, performing body classification on the appearance contour, and constructing the skeleton features of the corresponding steel enterprise personnel; Step 223: Based on the feature specifications corresponding to the skeleton features, the corresponding monochrome information is adjusted and mapped to the skeleton features, and the monochrome information is merged with the skeleton features to obtain the color distribution information within the specified range of the steel company personnel, and obtain the dynamic improvement behavior of the corresponding steel company personnel.

6. A steel enterprise personnel violation identification and control method based on AI analysis as claimed in claim 1, characterized in that: The step 3 comprises: Step 31: respectively obtain the reminder method corresponding to each of the violation items, and establish a violation reminder instruction for each of the steel enterprise's violators according to the reminder method; Step 32: Obtain basic information of the steel company personnel who violated the regulations, and determine the personality characteristics corresponding to each of the steel company personnel; Step 33: The violation reminder instruction is personalized according to the personality characteristics, and the selected violation reminder instruction is transmitted to the user terminal of the corresponding steel enterprise personnel for reminder.

7. A steel enterprise personnel violation identification and control method based on AI analysis as claimed in claim 1, characterized in that: The step 4 comprises: Step 41: Obtain the violation items of each of the illegal steel enterprise personnel respectively, and establish the violation time distribution axis of the corresponding illegal steel enterprise personnel according to the violation start and end time corresponding to each of the illegal items; Step 42: adding a corresponding frequency weight to the violation time distribution axis according to the violation frequency corresponding to each violation item, and adding a corresponding risk weight to the violation time distribution axis according to the risk level corresponding to each violation item; Step 43: Find the violation duration corresponding to each violation item in the violation time distribution axis, and establish semantic violation description information of the corresponding steel enterprise personnel in combination with the corresponding frequency weight and danger weight; Step 44: Generate a violation report for the monitoring area based on the semantic violation description information corresponding to all steel enterprise personnel who violated the regulations, and display it.

8. A steel enterprise personnel violation identification and control system based on AI analysis, characterized in that: include: The behavior collection module is used to track the behavior of each steel enterprise personnel in the real-time video stream when the steel enterprise personnel enter the monitoring area, and obtain the dynamic behavior information corresponding to each steel enterprise personnel; The violation identification module is used to use AI technology to deeply identify each of the dynamic behavior information, screen out the steel company personnel who violate the rules, and determine the violation items corresponding to each of the steel company personnel who violate the rules; A reminder execution module is used to establish a violation reminder instruction for the corresponding steel enterprise personnel based on the violation items and transmit it to the user terminal of the steel enterprise personnel for reminder; The violation supervision module is used to count the violation duration and violation items corresponding to each of the steel enterprise personnel who violated the regulations, establish and display the violation report of the monitoring area.

9. A steel enterprise personnel violation identification and control system based on AI analysis as claimed in claim 8, characterized in that: Also includes: The network layout module is used to set up several available networks in the monitoring area and control each of the user terminals to connect to the available networks.

10. A steel enterprise personnel violation identification and control system based on AI analysis as claimed in claim 8, characterized in that: The violation identification module comprises: The first recognition unit is used to perform color training on the dynamic behavior information using AI technology to obtain RGB dynamic behavior information and D dynamic behavior information corresponding to the steel enterprise personnel, and perform layered training on the RGB dynamic behavior information using the AI ​​technology to obtain a plurality of monochrome information corresponding to each of the steel enterprise personnel; The second recognition unit is used to construct the skeleton features corresponding to the steel enterprise personnel according to the D dynamic information, and respectively map each of the monochrome information to the corresponding skeleton features for color improvement training to generate dynamic improvement behaviors corresponding to the steel enterprise personnel; The third identification unit is used to determine the appearance color distribution characteristics of the corresponding steel company personnel based on the monochrome information, determine the clothing violation information of the corresponding steel company personnel, use AI technology to respectively identify the behavior results corresponding to each of the dynamic improvement behaviors, determine the operation violation information of the corresponding steel company personnel based on the behavior results, respectively count the clothing violation information and operation violation information corresponding to each of the steel company personnel, and establish violation items corresponding to each of the steel company personnel.

Images