A smoke flame detection method, system and electronic device based on HNetMSFA

Abstract

The embodiment of the specification discloses a smoke flame detection method, system and electronic equipment based on HNetMSFA, wherein the smoke flame detection method comprises the following steps: collecting a smoke flame image data set, dividing the smoke flame image data set into a training set, a verification set and a test set and performing data preprocessing; constructing a detection network based on hierarchical multi-scale feature spectrum network modulation; the main network of the detection network comprises a multi-scale feature spectrum modulation visual transformer module; the training set, the verification set and the test set are used to train and test the constructed detection network, and a smoke flame detection model is obtained; smoke flame detection is performed on a to-be-detected image based on the smoke flame detection model; and the multi-scale feature spectrum modulation visual transformer module performs global multi-scale feature aggregation. The problems that the existing smoke flame target detection has low detection precision, is prone to missed detection and false detection in a complex environment due to great differences in scale, shape and texture are solved.

Images