Camera-Server Accident Detection Using Digital Twin
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for detecting accident risks in working places face inefficiencies due to excessive data transmission and resource consumption, and limitations in analyzing complex three-dimensional workspaces, making it difficult to accurately determine work stages and potential accidents.
Innovation Solution
A system that uses a camera to photograph a working place, recognize objects, and transmit coordinates and object information to a server, which generates a virtual space using digital twin techniques to simulate the workspace, allowing for determination of work stages and potential accidents by comparing recognized objects' positions and types with stored data, and sends warning messages when necessary.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the camera transmits photographed images directly to the server for analysis, then the server can analyze the images to detect accidents and risks, but excessive data is transmitted and the server consumes excessive CPU or GPU resources
Solution Approach 1:
The system segments the image processing task into two parts: the camera performs preliminary object recognition and extracts only essential information (coordinates and object types), while the server processes only this extracted data rather than full images. This segmentation reduces server computational load while maintaining accident detection capability.
Solution Approach 2:
The camera extracts only the necessary information (object coordinates and types) from the photographed images and transmits only this extracted data to the server. This extraction process eliminates unnecessary data transmission and reduces server processing requirements while preserving the essential information needed for accident and risk detection.
2Area of stationary object
If the server processes images from multiple cameras, then comprehensive coverage of the working place is achieved, but the server cannot analyze the excessive data from multiple sources
Solution Approach 1:
Each camera extracts essential information (coordinates and object types) from its captured images before transmission. This extraction at the source reduces the data volume from multiple cameras, enabling the server to process comprehensive coverage data from multiple sources without being overwhelmed by excessive image data.
Solution Approach 2:
The system divides the processing workload by having each camera independently perform preliminary object recognition and data extraction. This segmentation allows parallel processing at the camera level, reducing the burden on the server and enabling efficient handling of data from multiple cameras simultaneously.
3Measurement precision
If two-dimensional images are analyzed directly, then changes in state of specific detection targets are easy to detect, but risky situations occurring in complex three-dimensional work space cannot be detected
Solution Approach 1:
The system transitions from analyzing two-dimensional images to constructing three-dimensional spatial relationships by collecting coordinate information from multiple cameras and reconstructing the positions of detection targets in three-dimensional space. This dimensional transformation enables the detection of complex spatial risks while maintaining the ability to detect state changes.
Data Source
AI summary
A system for detecting an accident risk in a working place that can determine a work stage of the working place where various kinds of works, such as a cargo loading and unloading workshop, are performed or determine whether or not there is a risk that an accident may occurs in the working place is disclosed. In the disclosed system for detecting the accident risk in the working place, a camera is used to photograph the working place, only information about an object recognized by analyzing the image is transmitted to the server without transmitting the photographed image to the server, thereby minimizing an amount of information transmitted from the camera to the server.