scispace - formally typeset
Patent

Fuzzy matching method for photogrammetry of port crane

Reads0
Chats0
TLDR
In this article, a fuzzy matching method for the photogrammetry of a port crane is presented, which includes the steps of calculating coordinates of an image plane coordinate system of each to-be-measured point as a key point according to an estimated parametric model and left and right exterior orientation elements.
Abstract
The invention discloses a fuzzy matching method for photogrammetry of a port crane. The fuzzy matching method for the photogrammetry of the port crane includes the steps of: calculating coordinates ofan image plane coordinate system of each to-be-measured point as a key point according to an estimated parametric model and left and right exterior orientation elements; respectively getting a simulated left photo to-be-measured point distribution map and a simulated right photo to-be-measured point distribution map according to the calculated plane coordinates of each to-be-measured point in theimage plane coordinate system; conducting a point screening process based on geometric relationship matching, which includes first adopting a feature recognition operator to obtain automatic recognition point maps of left and right photos, first filtering range according to preset matching point fuzzy recognition radius, and then using geometric constraint to select points to obtain a projectionpoint of an object point in the image plane coordinate system; determining whether a match exists; changing the matching point fuzzy recognition radius to expand the selection range when no result output occurs, and returning to iteration until matching results are acquired. The fuzzy matching method for the photogrammetry of the port crane has the advantage of solving special and complex environmental problems of port crane photography.

read more

Citations
More filters
Patent

Photogrammetry method for port hoisting machinery

TL;DR: In this paper, a photogrammetry method for port hoisting machinery is presented, where preliminary detection is made for real-time focal length by introducing a redundant observation point as a pointerpoint according to the method, and a pointer point weighting algorithm based on a lens optical distortion model is proposed to improve the accuracy of a focal length detection result.
References
More filters
Patent

Detection and removal of image occlusion errors

TL;DR: In this article, the authors proposed a method for detecting and eliminating correspondence errors associated with image occlusions by applying traditional correspondence methods for matching points in two images, a left image (FIG. 1A) and a right image (Fig. 1B), taken of the same scene.
Patent

Heterological remote-sensing image registration method based on geometric structure similarity

TL;DR: In this paper, a heterological remote-sensing image registration method based on geometric structure similarity is proposed, which belongs to the technical field of satellite image processing and is mainly used for HRS image registration.
Patent

Method for recognising not gauged star map

TL;DR: In this paper, a non-labeled identifying method of star atlas includes picking up character rector of navigation star, setting up navigation star mode library according to grille storage.
Patent

Spatial light clustering-based image surface feature point matching method

TL;DR: In this article, a spatial light clustering-based image surface feature point matching method is proposed, in which the uniqueness of a measured space is utilized to regress two-dimensional matching problems to a 3D space to solve the problems.
Patent

Image point matching method for industrial digital photogrammetry

TL;DR: In this paper, the distance from image points to corresponding epipolar planes are utilized as constraint conditions, and images are grouped according to known points, and the method comprises the following steps of sorting the images into a plurality of groups according to whether containing the same object points; then, combining and matching parts of each group of images with better calculated geometrical quality.