A method for extracting road characteristics in a high-resolution image

Abstract

The invention discloses a method for extracting road characteristics in a high-resolution image. A generator G capable of accurately identifying road characteristics is obtained by optimizing structures of a generator G, a discriminator D and a generative adversarial network V (D, G) and training parameters. In the training process, the difference between the generated sample Pg (x) and the real sample PData (x) is evaluated by using the Waserstein distance, so that the problem of gradient non-convergence of a loss function in the existing KL divergence or JS divergence evaluation can be effectively avoided. The road characteristics obtained through identification of the generator G obtained through the method are closer to the actual situation, and the accuracy is higher.

Description

technical field [0001] The invention relates to the field of image processing, in particular to a method for extracting road features in high-resolution images. Background technique [0002] The continuous and rapid development of high-resolution remote sensing technology and aerial photography acquisition means has led to the wide application of high-resolution images. The remote sensing observation images with a ground resolution of less than 10m are also promising for urban traffic planning, environmental monitoring, and other civilian fields. [0003] By 2018, several high-resolution satellites have been put into use. The transportation industry data center of the high-resolution earth observation system is also simultaneously open to the public for high-score data support in the transportation industry. It is foreseeable that the combination of "high score + traffic" will further enhance the refined management of road resources. [0004] Road is one of the most basic...

AI Technical Summary

Problems solved by technology

[0015] 1) For specific areas or types of roads, although various traditional methods can achieve good results, they lack universality and require special parameter design for this area or type of roads, and the effect of migration application is not good;

[0016] 2) In the face of low- and medium-resolution remote sensing images, although various methods can achieve certain results, but in the face of the refined extraction requirements of high-resolution remote sensing images, the general application effect is not good;

[0017] 3) Traditional methods cannot make full use of various features in remote sensing images to make comprehensive judgments, and traditional methods often only focus on the processing of some features and levels in remote sensing images;

[0018] 4) The degree of automation is low, and manual preprocessing and manual postprocessing are frequently involved

Images