Method for fire detection based on flame color template

Abstract

The invention relates to a fire detection method using a flame color template, which adopts a color CCD camera system, collects a plurality of flame samples in advance according to the distribution law of the three primary colors of the flame in the color space, and analyzes and counts the color values ​​of the flame to obtain The sample mean and variance of the three primary colors, and the statistical results are input into the computer as parameters. When the detector is working, the color CCD converts the on-site image into a digital signal, and sends it to the host memory through the image card. After the processor obtains the image data, it compares it with the parameters determined in advance to judge whether it conforms to the Gaussian distribution law. The area that satisfies the conditions is divided as a suspicious area, and it is differentially processed to further analyze whether the change of the target over time meets the characteristics of fire jumping, flickering, and instability, which improves the reliability of fire detection and can meet fire detection, Different needs such as linkage alarm, image monitoring, fire protection and anti-theft.

Description

Technical field [0001] The invention relates to a fire detection method, in particular to a method for fire detection using a flame color template. Background technique [0002] We know that traditional smoke, temperature, and light detectors have been widely used in building fire protection. Smoke detectors usually detect whether there are smoke particles produced by fire in the surveillance area. Smoke is a precursor to fire. The suspended solid and liquid particles contained in the flue gas become a basis for fire detection, but the size of the particles, environmental dust, electromagnetic, airflow and other factors often cause wrong judgments. Another feature of a fire is accompanied by a rise in temperature. The turbulent smoke also has a certain temperature. When the fire temperature parameter exceeds a certain value within a specified period of time, it triggers a point or line type temperature detector. In some cases, the temperature is low. If the fire is in the smolder...

AI Technical Summary

Problems solved by technology

[0002] We know that traditional smoke, temperature, and light detectors have been widely used in building fire protection. Smoke detectors usually detect whether there are smoke particles produced by fire in the monitoring area. Smoke is a precursor to fire, so The suspended solid and liquid particles contained in the smoke become a basis for fire detection, but the size of the particles, dust in the environment, electromagnetic, airflow and other factors often cause wrong judgments

Another feature of the fire is that the temperature rises, and the turbulent smoke also has a certain temperature. When the fire temperature parameter exceeds a certain value within a specified period of time, it triggers a point-type or linear temperature-sensitive detector. In some occasions or the fire is in the smoldering stage, the alarm time is often delayed because the temperature rise still does not reach the limit of triggering the alarm, and the drastic change in temperature, such as the process of using air conditioning or heating equipment, the airflow directly blows through the alarm containing temperature sensor The detectors of the components may cause false alarms; the photosensitive detectors have ultraviolet flame type and infrared flame type, which use the ultraviolet radiation and infrared radiation generated by the flame to detect the flame and respond respectively. The photosensitive detector has similar defects and is not Only fire emits ultraviolet and infrared light, there are X-rays in the environment, welding arcs, or the detector is directly or indirectly irradiated by sunlight or other light sources, which may cause false alarms

Images