Tricycle manned identification method and system

The invention provides a tricycle manned identification method and system, and the method comprises the steps: obtaining a traffic snapshot image in real time, and carrying out the image recognition; wherein the traffic snapshot image is acquired by a front-end image acquisition device and uploaded to a first message queue; the far-end server receives a traffic snapshot image based on the first message queue, identifies the image based on a built-in preset artificial intelligence algorithm, and uploads an identification result to a second message queue; based on an image recognition result, performing preliminary screening on data in the second message queue according to a first preset rule; on the basis of the preliminary screening result, amplifying the image data in the second message queue and then carrying out secondary identification on the image, and on the basis of the identification result, carrying out secondary screening on the data in the second message queue according to a second preset rule; and