Remote sensing image semantic segmentation method based on attention multi-scale feature fusion

Abstract

The invention discloses a remote sensing image semantic segmentation method based on attention multi-scale feature fusion, and the method comprises the steps: establishing a semantic segmentation network based on attention multi-scale feature fusion, building a training data set, and carrying out the network parameter training through employing the training data set. And performing semantic segmentation on to-be-tested data by using the trained network during testing. And the network is a lightweight encoder-decoder structure. Wherein the idea of an image cascade network is introduced; meanwhile, coding features and decoding features are optimized by using an attention mechanism, a multi-scale attention optimization module, a multi-scale feature fusion module and a boundary enhancement module are constructed, feature maps of different scales are extracted and fused, training is guided by using multi-scale semantic tags and boundary tags, and semantic segmentation of remote sensing images can be effectively carried out.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a remote sensing image semantic segmentation method based on multi-scale feature fusion of attention. Background technique [0002] At present, semantic segmentation has become one of the key problems in the field of computer vision. From a macro perspective, semantic segmentation is a high-level task that lays the foundation for scene understanding. In reality, there are a large number of application scenarios that need to infer relevant knowledge or semantics from image data (that is, the process from concrete to abstract). These applications include unmanned driving, medical health, image search engine, augmented reality, etc. These problems have been well solved by applying various traditional methods of computer vision and machine learning techniques. However, the rapid development of deep learning makes these methods gradually lose people's favor. In recent yea...

AI Technical Summary

Problems solved by technology

[0007] (3) Different target boundaries overlap, such as the shadows of trees and buildings will cause occlusion problems

[0008] All of the above characteristics pose new challenges to the existing deep learning models to learn robust feature representations, which is the key to improving the accuracy of semantic segmentation of remote sensing images

Moreover, there are very few remote sensing data sets with labels, which limits the development of deep learning networks in semantic segmentation applications, and makes the model prone to overfitting

Images