Remote sensing image road extraction method based on D-LinkNet

Abstract

The invention provides a remote sensing image road extraction method based on D-LinkNet, and the method comprises the following steps: S1, inputting a feature map into a D-LinkNet network, and completing the processing in an encoder sub-network based on a residual network and transfer learning; S2, inputting the feature map output in the step S1 into a feature extraction sub-network based on an expansion convolution and convolution block attention module for feature extraction; S3, enabling the feature map obtained after processing of the first two sub-networks to enter a decoder sub-network based on transposed convolution to realize image recovery. According to the invention, downsampling can be carried out on the road features in the remote sensing image, the problem of network degradation is well avoided, and the extraction of the road features is enhanced; the receptive field can be amplified by using the expansion convolution, the road characteristics in a larger range are sensedand the characteristics are extracted without increasing down-sampling, and the problem that the proportion of the road part in the remote sensing image is too small can be well solved.

Description

technical field [0001] The invention relates to a method for extracting a road from a remote sensing image, in particular to a method for extracting a road from a remote sensing image based on D-LinkNet, and belongs to the field of remote sensing image processing. Background technique [0002] Remote Sensing (RS) refers to the non-contact, long-distance real-time collection of earth resources and other targets through remote sensors or sensors, and then extracts, analyzes and processes relevant data information. In recent decades, many scholars at home and abroad have carried out extensive and profound research on the complex road information in remote sensing images, and have proposed various road extraction algorithms one after another. At present, according to the three relatively mature and commonly used strategies, the road extraction methods of remote sensing images at home and abroad can be divided into three categories, namely, the three road extraction methods based...

AI Technical Summary

Problems solved by technology

However, when the model is affected by complex terrain, the extracted roads are still not sufficient, and it cannot respond sensitively to changes in road width.

[0006] The road extraction method based on edge features can effectively extract simple road information, but its anti-interference ability is not strong; when there is less information similar to the road in the image, the extraction effect of the object-based road extraction method is good, but in other When the ground features are too similar to the road and the space is closely adjacent, most of these methods have adhesion phenomenon. For a few methods that improve this problem, the design of the extraction process is too complicated or the extraction scale is difficult to grasp, which makes road extraction difficult. Bottleneck; the road extraction method based on deep learning has a strong learning ability, which can well solve the problems of changing road types, complex and diverse backgrounds, and similar road and non-road features. However, such methods still have insufficient road extraction, It is easy to lose spatial information or not universal enough for scenes with large changes

Images