Video Smoke Detection Using Neural Network Trend Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing smoke detection methods are limited in their ability to effectively measure smoke conditions throughout an entire monitored area, often resulting in false alarms and requiring multiple sensors or complex installations, and fail to accurately detect thick smoke or smoke beyond the vicinity of light sources.
Innovation Solution
A system that captures successive video images, identifies potential smoke areas by analyzing pixel changes, and uses a neural network to analyze temporal changes and patterns indicative of smoke, allowing for comprehensive smoke detection within a monitored area and integration with existing CCTV surveillance systems.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If multiple sampling tubes are installed at assorted locations throughout the monitored area to increase spatial sampling, then the extent of spatial sampling is improved, but the device complexity and installation requirements worsen
Solution Approach 1:
The monitored area is segmented into multiple zones, each monitored by a separate video camera. This allows comprehensive spatial coverage without requiring physical sampling tubes at multiple locations, as each camera independently monitors its assigned zone for smoke conditions.
Solution Approach 2:
The video cameras serve dual functions: general surveillance and smoke detection. By using existing surveillance infrastructure for both purposes, the system achieves extensive spatial sampling without adding dedicated sampling equipment, thereby reducing installation complexity while maintaining broad monitoring coverage.
2Area of stationary object
If a laser beam is directed across the monitored area to sense smoke along a line, then the sensing area is improved, but the coverage of all points within the monitored area worsens
Solution Approach 1:
The system transitions from linear sensing (laser beam along a line) to two-dimensional area coverage using video cameras. Multiple cameras positioned at different angles capture overlapping fields of view that collectively cover the entire monitored volume, ensuring no areas are missed while maintaining comprehensive smoke detection capability.
3Measurement precision
If statistical analysis of pixel brightness and color intensity is used to identify smoke, then smoke detection capability is improved, but the false alarm rate worsens due to lighting changes and moving objects
Solution Approach 1:
The system performs preliminary learning during an initial period to establish baseline characteristics of the monitored environment, including normal lighting variations and moving objects. This pre-learning phase enables the smoke detection algorithm to distinguish between normal fluctuations and actual smoke conditions, significantly reducing false alarms while maintaining detection sensitivity.
Solution Approach 2:
The system continuously compares current pixel analysis results against the learned baseline characteristics and adjusts its detection thresholds dynamically. This feedback mechanism allows the system to adapt to changing environmental conditions while maintaining reliable smoke detection, reducing false alarms caused by lighting changes or moving objects that differ from the established baseline.
4Measurement precision
If image contrast analysis is used to detect smoke, then smoke detection is improved under certain conditions, but the method fails when smoke increases high frequency content or when background conditions are plain
Solution Approach 1:
The system employs multiple detection parameters beyond just image contrast, including pixel intensity changes over time, color space transformations, and temporal analysis of pixel sequences. By monitoring multiple parameters simultaneously, the system can detect smoke across diverse environmental conditions, including plain backgrounds and situations where smoke increases high frequency content, thereby improving environmental adaptability.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A system for detecting smoke in a monitored area includes: (a) a video device for capturing a series of successive video images of the monitored area as a series of two-dimensional bitmaps having a specified number of pixels, (b) a processing device having memory capability for storing said series of images and processing capability for analyzing the series of images, and (c) an analysis algorithm that runs on the processing device and has: (i) an identification portion for examining this series of bitmaps to identify indicator areas in successive bitmaps of adjacent pixels that have the potential for being used as indicators for the existence of smoke in the monitored area, (ii) a tracking portion for identifying the trends in the growth and movement of the indicator areas, and (iii) a trend comparison portion for comparing the identified trends to determine which of the trends are consistent with those produced by a smoke cloud.