Terahertz millimeter wave-based human face skeleton recognition method

The invention discloses a Terahertz millimeter wave-based human face skeleton recognition method. Millimeter waves are emitted to a to-be-detected target, reflection occurs on the interface of facialsoft tissues and facial bones of the to-be-detected target, a three-dimensional model is established for the facial bones of the to-be-detected target by collecting the reflected millimeter waves, comparison and search are performed in a facial bone model library after feature vectors of the model are extracted, and a recognition result is obtained. The method comprises five parts of millimeter wave emission, millimeter wave collection and emission, facial skeleton modeling, feature extraction and verification search. The method has the advantages that the human face can be recognized under the condition that part of the human face is shielded; the influence of external visible light factors is avoided; the method is not influenced by expression changes, age growth, face lifting and the like; the safety is improve