KR-20260062865-A - AI-BASED CHEMICAL LABORATORY SAFETY MONITORING SYSTEM

KR20260062865AKR 20260062865 AKR20260062865 AKR 20260062865AKR-20260062865-A

Abstract

The present invention provides an artificial intelligence-based chemical laboratory safety monitoring system comprising: an image collection module for collecting image data inside a chemical laboratory; and a server that communicates with the image collection module and receives and analyzes image data collected by the image collection module, wherein the server includes: a chatbot unit that recognizes voice input by an experimenter terminal and provides information on experiment procedures and risk factors; a storage unit that receives and stores the image data; an anomaly detection unit that analyzes the image data stored in the storage unit through an artificial intelligence model to detect in real time whether an event regarding an abnormal state of the experimenter has occurred; and a safety control unit that outputs an alarm signal and operates safety equipment according to the detection result of the anomaly detection unit.

Inventors

곽표성

Assignees

곽표성

Dates

Publication Date: 20260507
Application Date: 20251027
Priority Date: 20241028

Claims (9)

An image acquisition module for collecting video images inside a chemical laboratory; and A server that communicates with the above-mentioned image acquisition module and receives and analyzes image data collected by the above-mentioned image acquisition module Includes, and the above server A chatbot unit that recognizes voice input by an experimenter terminal and provides information on experimental procedures and risk factors; A storage unit that receives and stores the above-mentioned video image; An anomaly detection unit that analyzes video images stored in the above storage unit through an artificial intelligence model to detect in real time whether an event regarding the experimenter's abnormal state has occurred; and A safety control unit that outputs an alarm signal and operates safety equipment based on the detection result of the above-mentioned abnormality detection unit. An artificial intelligence-based chemical laboratory safety monitoring system including
In Article 1, The above abnormal detection unit An artificial intelligence-based chemical laboratory safety monitoring system that detects whether an event has occurred by analyzing the above video image in a multi-threaded manner using multiple object recognition models and adopting the analysis result of the object recognition model having the highest probability among the event occurrence probabilities derived from each object recognition model.
In Article 2, The above abnormality detection unit Environmental characteristics for the above video image are derived, and weights are set for each object recognition model according to the derived environmental characteristics, An AI-based chemical laboratory safety monitoring system that applies weights to the event occurrence probabilities of each object recognition model and adopts the analysis result of the object recognition model having the highest probability among the weighted event occurrence probabilities.
In Paragraph 3, An artificial intelligence-based chemical laboratory safety monitoring system in which the above environmental characteristics are calculated based on at least one of illuminance, brightness, object size, and object angle for the above image.
In Article 1, The above image acquisition module is A surveillance camera that films the interior area of a chemical laboratory while adjusting the shooting angle; and An image control unit that controls the operation of the surveillance camera according to a camera control signal transmitted from the above server An artificial intelligence-based chemical laboratory safety monitoring system including
In Article 5, The above surveillance camera adjusts the shooting angle by rotating 360° around a vertical axis via a pan motor and tilting up and down via a tilting motor, and The above image control unit is, A camera angle adjustment unit that controls the operation of the fan motor and the tilting motor so that the surveillance camera rotates 360° in increments of a certain rotation angle around a vertical axis, and tilts alternately in the up and down directions in each increment of rotation; and A camera shooting control unit that controls the operating state of the surveillance camera to adjust the zoom function of the surveillance camera. An artificial intelligence-based chemical laboratory safety monitoring system including
In Article 6, The above abnormality detection unit Monitoring the abnormal state of the experimenter through an object recognition model that recognizes the experimenter as a target object in the above video image, and When the object recognition accuracy on the above video image is below a threshold, a stop signal is generated to stop the rotation and tilting of the surveillance camera and a zoom signal is generated to perform the zoom function of the surveillance camera. An artificial intelligence-based chemical laboratory safety monitoring system that reanalyzes enlarged video images captured by the surveillance camera within the corresponding shooting angle range according to the stop signal and zoom signal.
In Article 7, The above camera shooting control unit adjusts the zoom function according to a preset magnification ratio in the process of adjusting the zoom function of the surveillance camera according to the zoom signal of the above abnormal detection unit, and An artificial intelligence-based chemical laboratory safety monitoring system, wherein the camera angle adjustment unit adjusts the shooting angle of the surveillance camera so that the surveillance camera sequentially zooms in and shoots for each divided area of the entire area of the basic shooting video image captured when the surveillance camera is in a state without a zoom function.
In Article 1, The above abnormal detection unit The abnormal state of the experimenter is monitored by determining whether the experimenter's falling event and static event occur through an object recognition model that recognizes the experimenter as a target object in the above video image, and An AI-based chemical laboratory safety monitoring system that determines that a static event has occurred when the experimenter is recognized as an object in the above video image and remains in a state where there is no change in the experimenter's movement for a reference period of time.

Description

AI-Based Chemical Laboratory Safety Monitoring System The present invention relates to an artificial intelligence-based chemical laboratory safety monitoring system. More specifically, it relates to an artificial intelligence-based chemical laboratory safety monitoring system capable of comprehensively monitoring the safety status within a chemical laboratory by analyzing video images captured of the internal space of the chemical laboratory in real time through an artificial intelligence model. In recent years, safety accidents such as gas leaks, explosions, and suffocation have frequently occurred in chemical laboratories at domestic universities and research institutions, and casualties resulting from these incidents are also continuously being reported. According to statistical data on laboratory accidents, the highest rate of accidents occurs in chemical laboratories, and accidents caused by a lack of safety awareness, such as failure to wear protective equipment, careless handling of equipment, failure to follow safety rules, and insufficient education, account for more than 76% of the total. In particular, there are approximately 45,000 university laboratories, but the government's annual budget for establishing a safe laboratory environment is only about 6 billion won, resulting in a significant shortage of safety management personnel and follow-up management systems in each laboratory. Generally, chemical laboratories are required to install and manage gas detection sensors and safety equipment in accordance with gas safety management guidelines; however, in reality, accidents caused by malfunctions or detection failures occur frequently because regular maintenance or sensor replacement is not properly carried out after installation. Furthermore, as most laboratories rely on an institutional self-management system, problems such as a lack of familiarity with safety protocols, the absence of managers, and delays in responding to emergencies are currently occurring. Due to this situation, existing chemical laboratory safety systems rely solely on hardware detection by sensors, and there is a lack of accident prevention systems capable of recognizing and responding in real time to the experimenter's behavioral status or hazardous situations during the experiment. In particular, when an experimenter conducts research alone, there are limitations in that immediate detection by the surroundings is difficult in the event of an accident, and it is impossible to respond to situations that sensors fail to detect (e.g., falling, fainting, failure to wear protective equipment). FIG. 1 is a block diagram conceptually illustrating the overall configuration of an artificial intelligence-based chemical laboratory safety monitoring system according to one embodiment of the present invention. FIG. 2 is a block diagram illustrating the functional classification of a server configuration according to one embodiment of the present invention. FIG. 3 is a block diagram illustrating the functional classification of the configuration of an image acquisition module according to one embodiment of the present invention. FIG. 4 is a conceptually illustrated diagram to explain a multi-threaded monitoring method according to an embodiment of the present invention. FIG. 5 is a conceptually illustrated diagram to explain the process of applying weights according to environmental characteristics in accordance with one embodiment of the present invention. FIG. 6 is a diagram exemplarily illustrating the shooting angle adjustment state of a surveillance camera according to one embodiment of the present invention. FIG. 7 is a drawing illustrating a segmented area for a captured image according to an embodiment of the present invention. FIG. 8 is a drawing illustrating a method for enlarging and photographing a segmented area of a captured image according to an embodiment of the present invention. FIG. 9 is a flowchart illustrating, step-by-step, a method for monitoring the safety status of a laboratory through a server according to one embodiment of the present invention. Hereinafter, specific embodiments for implementing the present invention will be described in detail with reference to the drawings. First, it should be noted that when assigning reference numerals to the components of each drawing, the same components are assigned the same reference numeral whenever possible, even if they are shown in different drawings. Furthermore, in describing the present invention, if it is determined that a detailed description of related known components or functions could obscure the essence of the invention, such detailed description is omitted. Furthermore, when it is stated that one component is 'connected,' 'supported,' 'connected,' 'supplied,' 'transmitted,' or 'contacted' with another component, it should be understood that while the connection, support, connection, supply, transmission, or contact may be direct to that other component