Road network extraction method based on adaptive cluster learning

Abstract

The invention provides a road network extraction method based on adaptive cluster learning. The method comprises steps of connection network constructing of an extracted road vector; new road detection and extraction. An inference of a road extraction result verifies three aspects of contents. A road connection network constructing process comprehensively considers a geometrical characteristic of the road and a detected road intersection structure constraint so as to guarantee rationality of a road connection result. New-road automatic extraction is a difficult point in a road extraction research field. Under the condition of an existing road-vector semantic mark, problems of high resolution remote sensing image road characteristic heterogeneity and diversity still need to be solved. Fusion of adaptive sample cluster and a multi-classifier road sample classification result is taken as a research thought of the new road extraction. Finally, based on an accuracy requirement of a road extraction result, a D-S evidence theory is introduced to verify a road extraction result, and according to relation between each kind of characteristic and the road, a corresponding verification probability distribution function is defined.

Description

technical field [0001] The invention relates to the technical field of remote sensing image applications, in particular to a road network extraction method based on adaptive clustering learning. Background technique [0002] The road network is an interconnected network object, and the completeness and correctness of its extraction are the basic requirements for road network update. The road segment extraction and intersection extraction guided by navigation data are independent extraction processes for each road segment, which leads to breaks between road segments. From the perspective of the integrity of the road network, the existing road segment extraction results do not cover the new road segment, and it is necessary to detect the new road object based on the known road features. Accurate extraction of new roads requires correct and comprehensive sample feature support. The workload of manual sample labeling is heavy and it is difficult to cover all road features. Usi...

AI Technical Summary

Problems solved by technology

[0003] The road network extraction task of high-resolution remote sensing images has its particularity. The extraction method based on sample learning has two main problems when dealing with high-resolution image data: (1) roads are diverse in different scenes; There are feature differences, and it is difficult to use fixed features and rules to achieve universal road extraction tasks; (2) Based on supervised road extraction, it depends on the absolute randomness of sampling, but it is very difficult to achieve randomness in sampling. Once the sampling process Any bias in the extraction results will vary

Images