Optical fiber detection model training method, optical fiber detection method and related equipment

The invention relates to the field of optical fiber detection, in particular to a training method of an optical fiber detection model, an optical fiber detection method and related equipment. The training method of the optical fiber detection model comprises the steps of collecting an optical signal transmitted by a damaged optical fiber, wherein the damaged optical fiber is an optical fiber subjected to external stimulation; performing feature extraction on the optical signal to obtain a plurality of feature parameters of the optical signal; creating a sample data set based on the plurality of feature parameters of the optical signal; training an optical fiber detection model based on the sample data set; wherein the optical fiber detection model is used for determining a detection result of the to-be-detected optical fiber based on the characteristic parameters of the optical signal transmitted by the to-be-detected optical fiber, and the detection result is used for describing whether the to-be-detected op