Sub-pixel-level image matching navigation positioning method

Abstract

The invention provides a sub-pixel-level image matching navigation positioning method, which comprises the following steps of: compressing an image by using Haar wavelet, representing the image by using affine invariance characteristics, extracting edge characteristics, and performing sub-pixel-level matching positioning on characteristic points by combining Gaussian sub-pixel fitting. The methodhas high robustness, the problem that the matching performance is reduced due to large image viewing angle deviation can be solved, detail information such as texture of a real environment can be reconstructed with high precision, and the accuracy of unmanned aerial vehicle positioning and environment construction is improved; the calculation amount can be greatly reduced, and the matching efficiency is improved.

Description

technical field [0001] The invention relates to the field of visual navigation, in particular to an image matching navigation positioning method. The image is represented by the feature points with affine invariance, combined with the Gaussian sub-pixel fitting principle and the feature descriptor simplification method to achieve fast sub-pixel positioning, and to solve the need for high-precision image matching algorithms for the visual navigation of micro-miniature UAVs . Background technique [0002] UAV visual navigation means that the UAV perceives the environment with the help of visual sensors, uses the computer to process, analyze and recognize the image, and estimates the UAV's own pose by matching with the prior image to achieve distance measurement. , hovering, autonomous landing and planning path obstacle avoidance and other tasks. Image matching is one of the core technologies in the visual navigation process of UAVs, and its matching speed and accuracy have a...

AI Technical Summary

Problems solved by technology

[0004]But no matter which method has its own advantages and disadvantages: 1) The extended phase correlation method can only deal with fixed forms of image changes, for more complex nonlinear transformations will not be able to achieve the matching effect

2) The disadvantage of the interpolation method is that the amount of calculation is large, and the matching accuracy depends on the quality of the interpolation algorithm

[0005] Based on the above research and analysis, various methods have large differences in matching accuracy, running time, and robustness caused by differences between images

Therefore, it is still a big challenge to improve the image matching accuracy to the sub-pixel level, reduce the complexity of the algorithm, reduce the false matching rate and enhance the robustness of the algorithm to image differences.

Images