The present disclosure provide a method for detecting traffic accidents by using the helmet of an electric bicycle, and the method includes: acquiring real-time state information of the electric bicycle; controlling a camera mounted on a front of the helmet of the electric bicycle to enter a snapshot mode and acquiring environment image information captured by the camera according to a first preset time interval in response that the electric bicycle is in the started state; and in response that the environment image information comprises the preset traffic accident image, controlling the camera to enter a first video recording mode, and sending the environment image information of a first preset period of time recorded by the camera and/or a first warning command to a first target object after waiting for the first preset period of time.