Early fire detection method, computer readable medium and system

A detection method and fire technology, applied in the field of computer vision, can solve the problems of slow flame development and easy missed detection due to the difference between before and after frames, so as to improve the accuracy, prevent the missed detection and increase the accuracy.
CN110378265APending Publication Date: 2019-10-25创新奇智(成都)科技有限公司

Patent Information

Authority / Receiving Office
CN · China
Current Assignee / Owner
创新奇智(成都)科技有限公司
Publication Date
2019-10-25

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

The invention relates to an early fire detection method which is used for detecting a fire and comprises the following steps: S1, acquiring a real-time monitoring video; S2, providing a single-frame detection network model, and using the single-frame detection network model to detect whether flame and / or smoke exists in the image at the current moment or not so as to obtain a first detection result; S3, providing a time sequence detection network model, and detecting a flame level and / or a smoke level in the video sequence before the current moment by using the time sequence detection networkmodel to obtain a second detection result; and S4, calculating a fire risk level by combining the first detection result and the second detection result. Thus, missed detection of fire disasters is prevented, and the accuracy of fire detection is improved. The invention also provides a computer readable medium. The invention also provides a detection system.
Need to check novelty before this filing date? Find Prior Art

Description

【Technical field】

[0001] The invention relates to the field of computer vision, in particular to an early fire detection method, a computer-readable medium and a system. 【Background technique】

[0002] Most of the existing fires are detected based on smoke sensors or infrared sensors. However, it takes a certain amount of time for smoke and heat to diffuse to the corresponding sensors, so these sensors cannot detect fires in time;

[0003] With the improvement of image analysis technology, researchers consider using video as input and using multi-frame information for flame detection. This type of method generally detects the motion area by calculating the frame difference between the two frames of images before and after, and then judges whether the motion area is flame or smoke, that is, uses the color feature to detect the suspected flame area on the frame difference image, and detects it based on the background difference method. Suspected smoke areas, and then extract ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More