Research on multi feature fusion perception technology of mine fire based on inspection robot

In order to improve the real-time and accuracy of mine fire identification, based on summary existing monitoring methods, a early warning model patrol robot multi-sensor image recognition is proposed. The video obtained by using high-definition camera inspection robot, flame suspected area combining frame difference background after filtering. improved three scale analytic hierarchy process used fuse circularity feature, sharp angle centroid movement color feature growth obtain result. three-frame method binarize smoke image, texture motion are extracted recognize image. three-scale (AHP) judge in combination infrared thermal imager temperature measurement sensor. Inspection uses unified detection standards intelligent analysis technology sense flame. This can real-time, robustness perception.