In recent years

wild fire occurs frequently

which seriously threatens the security and stability of power grid. Consequently

on the basis of binocular stereo visions

we studied a positioning technology of wild fire nearby transmission lines by using the visible and infrared camera of the distributed wild fire monitoring device. A flame identification model for visible images was established based on HSV（Hue

Saturation

Value）color space. And a method of positioning for wild fire nearby transmission lines based on visible and infrared stereo vision was proposed. Taking the center of flame as the feature point of stereo matching

the method overcomes the problem of heterogeneous asynchronous binocular flame image matching. And the automatic measurement of fire to camera distance is realized.Moreover

we established a coordinate system based on the rotation center of the head

and put forward an accurate calculation method of the shortest distance from the fire to the transmission lines. This omits the problem of calibrating the external parameters of the camera outdoors

and the accurate positioning of the wildfire relative to the transmission line is achieved. Finally

through the experimental study

the feasibility and validity of the proposed method are verified. In the range of 500 meters

the measurement error of the distance from the fire to the camera is less than 6.6%

which can meet the practical application requirements. The research is of great significance to realize the automatic judgment of fire hazard level of transmission lines.