Night airport foreign matter intrusion intelligent detection method and system

Abstract

The application provides a kind of night airport foreign matter intrusion intelligent detection method and system, it is related to deep learning and foreign matter intrusion detection technical field, obtains runway adjacent article type and adjacent article position, according to runway adjacent article type, data frequent item mining is carried out, runway foreign matter category set is matched, calibrates foreign matter intrusion high-frequency area and foreign matter intrusion low-frequency area, collects foreign matter detection data set, based on window conversion self-attention mechanism, trains night airport runway foreign matter intrusion detection model, based on model, night airport foreign matter intrusion detection is carried out, solve the technical problem that existing airport foreign matter intrusion detection technology is carried out by artificial investigation to foreign matter intrusion identification, detection method is lower in intelligent degree, low in efficiency and higher in cost, and the technical problem that detection accuracy is insufficient, based on machine vision instead of artificial investigation, intelligent accurate prediction analysis of collected information is carried out, accurate and efficient intrusion foreign matter investigation is realized.

Images