Fire Smoke Propagation Path Detection Using Image Centroid Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional fire safety systems rely on point-based detection methods, leading to latency and inefficiencies in detecting fire and smoke, especially in high-ceiling environments like industries, resulting in potential delays in evacuation and inadequate information for planning safe evacuation routes.
Innovation Solution
A system and method utilizing image processing and deep learning to determine the propagation path of fire or smoke by analyzing image frames for regions of interest, applying motion detection and color segmentation, and tracking the displacement of centroids to render an output on the direction of propagation, enabling timely and informed evacuation planning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If point-based detection devices are used, then the system structure is simple, but the detection latency increases and detection accuracy decreases
Solution Approach 1:
The patent segments the detection task by dividing the field of view into multiple zones and using multiple imaging devices positioned at different locations. Each device captures a specific region, and the system processes multiple image streams simultaneously to detect fire and smoke across the entire area, reducing detection latency while maintaining manageable system complexity through modular architecture
Solution Approach 2:
The patent transitions from point-based detection to area-based detection by capturing two-dimensional image frames. This dimensional change allows simultaneous monitoring of multiple points across the field of view, eliminating detection latency while the structured zone division keeps the system organized and manageable
2Device complexity
If point-based detection devices are used, then the device complexity is low, but the field of view coverage is limited
Solution Approach 1:
The field of view is segmented into multiple zones that can be monitored by different imaging devices. Each device focuses on specific regions, and the system integrates these segmented views to achieve comprehensive area coverage while keeping each individual detector's configuration simple
Solution Approach 2:
The imaging devices serve multiple functions: capturing visible light for fire detection, detecting smoke through image analysis, and providing wide-area coverage. This multi-functionality expands the effective field of view without proportionally increasing device complexity
3Ease of operation
If conventional detection methods are used, then the system is easy to operate, but false alarms increase and detection reliability decreases
Solution Approach 1:
The patent introduces an image processing unit as an intermediary that analyzes visual data before triggering alarms. This intermediary layer processes image frames to distinguish actual fire/smoke from other visual patterns, reducing false alarms while the system remains easy to operate through automated analysis
Solution Approach 2:
The patent replaces conventional mechanical/chemical detection methods with optical image processing. This substitution uses visual analysis algorithms to detect fire and smoke characteristics, improving reliability by analyzing patterns rather than relying on simple threshold-based sensors that cause false alarms
Data Source
AI summary
A method and system of determining a propagation path of fire or smoke is disclosed that includes receiving, by a processor, a plurality of image frames captured by an imaging device. A plurality of regions of interests are determined based on determination of one or more object and masks based on motion detection and color segmentation. A class for each of the plurality of regions of interest is determined to be one of a fire class or a smoke class using a deep learning model. Further, a direction of propagation path of fire or smoke based on a displacement in coordinates of a centroid of each of the plurality of regions of interest is determined in each of the plurality of image frames. An output is rendered based on the detection of the class along with the direction of propagation path of fire or smoke.