Method and device for positioning ground fire by monocular camera

Abstract

The invention relates to a method and a device for positioning ground fire by a monocular camera. The method comprises the following steps of: arranging the monocular camera above a ground area to be monitored, driving the monocular camera to move and scan the ground area to be monitored to obtain a monitoring image, when a flame object appears in the monitoring image, appointing a point within the monitoring image by taking the center of the bottom edge of the image formed by a flame object as a reference point, and driving the monocular camera to move, so as to enable the reference point to be coincided with the appointed point to put out a fire by spraying water. Correspondingly, the device comprises a movement mechanism which is arranged above the center of the ground area to be monitored and can rotate along the horizontal surface and the vertical surface, a monocular camera which is arranged on the movement mechanism, and a controller which is connected with the monocular camera and the movement mechanism respectively. According to the method and the device, the positioning precision is high, the positioning speed is quick, the positioning method is simple, technical requirement is low, and installation and debugging are easy.

Description

technical field [0001] The invention relates to the technical field of camera-based fire location, in particular to a method and device for locating ground fires using a monocular camera. Background technique [0002] Fire has caused huge loss of life and property to people's production and life. How to quickly and effectively find the fire source and extinguish the fire, and minimize the loss caused by the fire is a key technical problem that needs to be solved urgently. For an intelligent automatic fire extinguishing system, one of the key technologies is to lock the target of the fire source in the shortest possible time, accurately locate the location of the fire source, and prepare for sprinkler fire extinguishing. [0003] The traditional infrared sensor-based positioning method is affected by the vibration caused by the movement of the cannon body, the external temperature change, electromagnetic interference and other factors, which is easy to cause false positives a...

AI Technical Summary

Problems solved by technology

[0004] The existing fire positioning method for images uses two cameras to form a stereoscopic vision to obtain the fire coordinates. The positioning method is cumbersome and complicated, with high technical requirements and difficult installation and debugging.

The binocular camera stereo vision system needs to calibrate and correct the camera considering various distortion factors of the camera before positioning. The calibration process is complicated and professional, and it is difficult to popularize in practical applications.

Binocular camera positioning requires fixed installation of the camera position in the site. For different occasions, it is necessary to recalculate and establish the coordinate system, calculate the positional relationship between each camera and the coordinate system, and extract the fire source position. The algorithm is complex, the environment adaptability is poor, and the real-time performance is not good. high

Patent 200910033034 "A method and system for intelligent fire detection and extinguishment based on computer vision technology" disclosed in patent 200910033034 uses binocular parallax method for fire location, but it does not calibrate the binocular camera and does not consider various distortion factors of the camera , poor positioning accuracy, poor adaptability

Patent 201110238242 discloses the "Binocular camera fire detection and positioning device and its fire positioning method", which locates the fire by adjusting the two cameras to overlap the image coordinates of the fire root and the coordinates of the image center to form a triangle relationship, because the flames collected from different angles The position of the root image is different, and the ideal triangle relationship cannot be satisfied to a certain extent, which makes the algorithm result deviate from the actual situation, and the positioning accuracy is not high

Images