Research on Binocular Forest Fire Source Location and Ranging System
Zhang Yaping, Shao Zhihang · 2020 IEEE International Conference on Power, Intelligent Computing and Systems (ICPICS) · 2020
The system under OpenCV environment was studied to realize the localization and ranging of forest fire sources effectively. The forest fire was identified by using the solar-blind ultraviolet-infrared binocular detector, and a gyroscope measured the azimuth. With the switching function of the binocular sensor, the ranging binocular part of the system can be activated. The equipment that has completed camera calibration and binocular image correction uses the SGBM algorithm to perform binocular matching and obtain a disparity map. Then the three-dimensional transformation of coordinates can be completed to get the ranging result under the OpenCV environment. The system adopts ultraviolet sun-blind imaging monitoring technology to eliminate the interference of sunlight and utilizes SGBM algorithm, which has the advantages of fast processing speed and distinct processing effect. The obtained parallax map provides a reliable basis for binocular ranging; thus, it has a minor error. This system has realized the binocular forest fire source localization and ranging, with advantages of self-sensing, self-identification, self-triggering, so it has a great automatic extent. It has advanced significantly in the forest fire location and range.