A remote fire detection system and method based on video images

Abstract

The invention belongs to the technical field of remote fire detection, and discloses a remote fire detection system and method based on video images. The method includes: S1. periodically acquiring infrared images in the current scene; Image feature extraction, identify flames or smoke on the extracted image features through the recognition algorithm, and obtain the suspected fire area; S3. Obtain the visible light image in the suspected fire area, and identify whether there is visible flame or visible smoke in the visible light image; Yes , then a fire occurs, obtain visible light images of other areas except the suspected fire area, and perform step S4; otherwise, perform a deep learning based on the deep learning module, and re-execute step S1 after the learning is completed; S4. It is transmitted to the remote control center, and at the same time, a first-level fire alarm is issued in the suspected fire area, and a second-level indication alarm is issued in other areas except the suspected fire area.

Description

technical field [0001] The invention belongs to the technical field of remote fire detection, and in particular relates to a remote fire detection system and method based on video images. Background technique [0002] At present, with the rapid increase of population and the development of industrialization, electricity consumption has increased significantly. The potential threat of building fires; in order to reduce the losses caused by fires, people have developed various fire detection devices so as to give early warnings in the early stages of fires. [0003] In the prior art, common fire detection devices include smoke detectors for detecting smoke, temperature detectors for detecting temperature changes, flame detectors for detecting light intensity of flame burning and flickering frequency of flames, and flame detectors for detecting flammable or toxic gases. Special gas detectors, etc. [0004] However, most of the detection devices mentioned above will be affecte...

AI Technical Summary

Problems solved by technology

[0004] However, most of the detection devices mentioned above will be affected by the environment in which they are used in actual use, resulting in the problem of low detection accuracy

For example, for a temperature detector that detects temperature changes, generally only when the detected temperature is higher than a predetermined threshold, an alarm will be generated; in actual situations, the temperature detected by the temperature detector and its distance from the fire point If the distance is far away, it is difficult to detect the occurrence of fire in time

In addition, for some industrial equipment with high working temperature, the temperature may be higher than the predetermined threshold during normal operation. At this time, false alarms will occur, resulting in a waste of fire protection resources.

Images