Steady and rapid unmanned aerial vehicle photogrammetry mark detection and positioning method

Abstract

The invention provides a steady and rapid unmanned aerial vehicle photogrammetry mark detection and positioning method. The method comprises the following steps: firstly, enhancing and graying a photographic image; comparing the centroid coordinates of each pair of black connected domain and white connected domain in the image; when the distance between the two does not exceed a set pixel threshold value, considering the region as a candidate target region and taking the centroid coordinate of the white connected domain as the initial detection coordinate of the candidate mark; detecting the number of zero-crossing points in the neighborhood of the candidate mark by using a zero-crossing detector; when and only when the number of the zero crossing points is 4, reserving the candidate targets, otherwise, deleting the candidate targets; finally, performing polynomial surface fitting on a sign image gray intensity response diagram generated based on local Radon transform, and solving theaccurate coordinates where the maximum value point of the fitting surface is located, namely the high-precision coordinates of the measurement sign, in an iterative mode. The method of the invention is strong in anti-interference capability, improves the calculation efficiency and the mark positioning precision, and is suitable for the processing of a large-resolution image.

Description

technical field [0001] The invention belongs to the technical field of engineering surveying, and in particular relates to a robust and fast method for detecting and locating marks of unmanned aerial vehicle photogrammetry. Background technique [0002] With the gradual maturity of UAV hardware technology and computer vision algorithms, photogrammetry methods based on UAV platforms have been widely used in the field of engineering surveying. Among them, the use of UAV photogrammetry marks to determine the coordinate reference frame of the photogrammetry 3D model is one of the main research contents of the photogrammetry technology based on the UAV platform, and the research has high accuracy, high precision positioning and no human intervention. The method of automatic detection and positioning of measurement marks is very important for improving the degree of automation and measurement accuracy of photogrammetry technology. However, due to the complex natural environment, ...

AI Technical Summary

Problems solved by technology

[0005] Harris corner detection is the most general corner detection method, but its defect is that it can not only detect cross-shaped corners, but also detect other types of corners such as sharp corners, and cannot achieve a single screening, so it will seriously affect Detection effectiveness and reliability of UAV survey marks

In addition, the computer vision library OpenCV provides a typical cross-shaped mark detection method, which detects corner coordinates by segmenting black and white rectangular grids and using their edge features; Wang et al. also proposed a fitting However, both methods are based on image edge features, which are easily affected by image noise, image blur and other factors, resulting in low accuracy of logo positioning

Geiger et al. proposed a method based on template matching for initial detection, and achieved good experimental results, but this method is a search method based on a sliding window, which has high computational complexity, so the computational efficiency is low, and it is not practical. sex

Images