Aircraft flight attitude visual image recognition method

Abstract

The invention provides an aircraft flight attitude visual image recognition method comprising the steps that an aircraft flight real-time image or an offline single frame / sequence image is acquired toact as an input image, and the image is denoised by using adaptive median filtering; the denoised image is binarized by using by using a moving mean adaptive threshold method; the foreground is processed by using morphological operation and the holes of the area are filled so as to acquire each connected area, and preliminary screening is performed according to the aircraft feature similarity index so as to acquire a suspected target set; the concave angle points and the convex angle points of each target in the suspected target set are detected by using the multi-scale interior angle point operator, and the concave and convex angle points of each target are connected so as to form the contour of each target; the multi-feature fusion matching index is established according to the contourfeatures of the target to be detected, and the contour of the aircraft to be detected is recognized from the suspected target set by using the index; the angle code template image most similar to thecontour of the target aircraft is acquired from a simulation template library by using the local predictive searching and template matching strategy; and the flight attitude parameters of the currentframe of aircraft are solved from the matched angle code image.

Description

technical field [0001] The invention relates to a target recognition technology, in particular to a method for recognition of aircraft flight attitude. Background technique [0002] The three-dimensional attitude angle of the aircraft when it is flying in the air, that is, pitch angle, yaw angle, and roll angle, is an important parameter to characterize the flight state. Its accurate measurement is of great value in the fields of aircraft test experiments and accident analysis. More and more researchers and domestic and foreign scholars are paying attention to this research field. Since different aircraft targets have their own structural characteristics, they are in different positions in space at different times, and have different attitude angles, so the attitude angle is actually an important indicator reflecting the flight status and performance of the aircraft in the air. Angle parameters, you can understand the flight status of the aircraft, and quantitatively measur...

AI Technical Summary

Problems solved by technology

For visible light images, the difficulty of aircraft recognition is that the target itself does not have obvious common grayscale features, and the shape, size, and grayscale of different types of aircraft vary greatly, so it is difficult to obtain a complete and accurate image of the aircraft through the grayscale method. Shape, low recognition rate, difficult to distinguish models

In actual images, due to the influence of occlusion, shadow, background interference and low image quality, it is often difficult to accurately extract the closed contour of the target through the contour

Images