A Fuzzy Matching Method for Port Crane Photogrammetry

Abstract

The invention discloses a fuzzy matching method for port hoisting machinery photogrammetry, which includes calculating the image plane coordinate system coordinates of each point to be measured as a key point according to an estimated parameterized model and left and right outer orientation elements; The plane coordinates of each point to be measured obtained through the calculation in the image plane coordinate system are obtained respectively to obtain the distribution map of the points to be measured in the simulated left photo and the distribution map of the points to be measured in the simulated left photo; using the feature recognition operator, the left photo and The automatic recognition point map of the photo on the right; the point screening process based on geometric relationship matching, including firstly performing range screening according to the preset matching point fuzzy recognition radius, and then using geometric constraints to select points to obtain the object point in the image plane coordinate system Project the point to judge whether there is a matching result. When there is no result output, change the fuzzy recognition radius of the matching point, expand the range of selected points, and return to iteration until the matching result is obtained. The invention can solve the special and complicated environment problem of port hoisting machinery photography.

Description

technical field [0001] The invention belongs to the technical field of port hoisting machinery photogrammetry and is used for a fuzzy matching method of homonymous points in port hoisting machinery photogrammetry. Background technique [0002] In the process of photogrammetry, the identification and matching of points with the same name is a crucial step, which is a powerful guarantee for the efficiency of the measurement process and the accuracy of the measurement results. At present, in the fields of geodetic surveying, architectural surveying and cultural relics protection, there are relatively mature related algorithms. However, when these algorithms are applied to the measurement of port machinery, the effect is not satisfactory. The low recognition rate of feature points caused by its own structural characteristics and the particularity of the working environment is the main reason for this phenomenon. [0003] Different from the photogrammetry object of the tradition...

AI Technical Summary

Problems solved by technology

[0004] (1) The feature points have no obvious color characteristics, and the painting of port cranes is mostly a single blue or orange red, which makes the color characteristics of the points to be measured on port cranes less prominent than the background. A large amount of noise will be generated when the recognition operator performs automatic recognition;

[0005] (2) The shooting background of port cranes is complicated. In ports, port cranes of the same type or even the same model are often placed side by side. For example, container cranes on shore are often arranged side by side along the wharf, which makes There will be complex and indistinguishable backgrounds on the photos used for photogrammetry, which will further increase the number of noise points when using existing point recognition operators for automatic recognition;

[0006] (3) The photos used for photogrammetry of port hoisting machinery generally have large shooting distances and angles. The port environment is complex, and there are a large number of areas where measurement stations cannot be set up, such as seasides and logistics channels. Therefore, the photos used for solving The shooting position and shooting angle generally change greatly, which will undoubtedly greatly increase the difficulty of matching points with the same name

Images