A Method of Extracting Road Intersections from Remote Sensing Images Based on Structural Index Features

Abstract

A method for extracting road intersections from remote sensing images based on structural index features. The method first obtains the intersection location set to be detected according to the line intersection information in the road network vector; obtains the intersection image slices according to the intersection location and buffer radius, and The multi-directional morphological filtering is used to enhance the processing to eliminate the obvious small-sized disturbance objects on the road surface. Intersection structure detection is mainly implemented in two levels: level intersection structure detection and road grade intersection structure detection. When the detection result of the level intersection structure is inconsistent with the intersection structure in the navigation road network, it is considered that the false detection result may be caused by the separated grade intersection, and the road grade intersection structure detection method is used for further detection and verification. Considering the widespread existence of compound road intersection structures, the extracted adjacent road intersections are structured using the prior road intersection model as a constraint.

Description

technical field [0001] The invention relates to the technical field of remote sensing image applications, in particular to a method for extracting road intersections from remote sensing images based on structural index features. Background technique [0002] The topological structure of road network is one of the contents concerned in the update of basic geographic information road network, and it is an indispensable and important information for building a complete road. Road intersections are the centralized representation of the topological structure between road segments. By detecting the structural information of existing road vector intersections, the connection relationship between roads can be determined, and then the roads can be connected into a road network; and the road intersection information also provides powerful clues for new road detection and verification. Road intersection elements are mainly represented by point objects, which belong to road elements an...

AI Technical Summary

Problems solved by technology

The method based on edge detection is only suitable for the intersection detection task with significant edge features of the branch. Even for the road intersection extraction task with significant edge features of the branch, the interference of the edge features of the background objects will cause the stability of the intersection extraction method to decrease. , so its application is very limited

With the improvement of image resolution, mixed pixels are gradually replaced by relatively purer pixels, and the distribution of spectral characteristics of pixels inside the same kind of ground objects is also more complex, which makes the traditional corner detection and gradient statistical features Road intersection extraction method fails

Images