A traffic information statistics method and system based on UAV aerial photography video

Abstract

The embodiment of the present invention discloses a traffic information statistics method and system based on UAV aerial photography video. The statistical method includes: collecting high resolution aerial photography video of UAV; The starting frame is used for multi-target detection, and the detection targets in multiple subdivided objects are merged as a whole according to the merging rules; the target detection result is used as the initial tracking target for video frame target tracking, and the target displacement between two frames is calculated. Tracking: update the target through the results of target detection and target tracking, and track accurately in real time; perform post-processing according to the tracking results and obtain the actual space coordinates of each pixel through the conversion between the image space of the frame image and the object space. The spatial trajectory in the time series calculates traffic information and obtains road traffic flow information.

Description

technical field [0001] The invention relates to the field of intelligent analysis of aerial video, in particular to a traffic information statistics method and system based on unmanned aerial vehicle video. Background technique [0002] With the rapid increase in the number of cars in cities, traffic jams have become a difficult problem in the field of urban transportation. Especially in recent years, due to insufficient long-term consideration of urban planning, neglect of the importance of urban infrastructure construction, and backward road traffic management technology, road congestion and frequent traffic accidents have caused road traffic congestion to become the most difficult solution for large and medium-sized cities in China. one of the problems. Problems such as road congestion, traffic congestion, difficulty in riding, difficulty in driving, and difficulty in traveling affect people's travel and urban development. How to achieve efficient traffic scheduling, gu...

AI Technical Summary

Problems solved by technology

The time difference method is to use pixel-based inter-frame difference between two or three adjacent frames in a continuous image sequence, but it generally cannot fully extract all relevant feature pixels, and it is easy to generate holes in moving entities.

Motion detection based on the optical flow method is limited in its application due to its large amount of calculation, poor anti-noise performance, and dependence on special hardware devices.

At present, the automatic statistics method of traffic flow based on the surveillance video obtained by fixed surveillance cameras cannot achieve ideal results.

Images