Deep learning-based semantic segmentation of remote sensing images: a review

Abstract

Semantic segmentation is a fundamental but challenging problem of pixel-level remote sensing (RS) data analysis. Semantic segmentation tasks based on aerial and satellite images play an important role in a wide range of applications. Recently, with the successful applications of deep learning (DL) in the computer vision (CV) field, more and more researchers have introduced and improved DL methods to the task of RS data semantic segmentation and achieved excellent results. Although there are a large number of DL methods, there remains a deficiency in the evaluation and advancement of semantic segmentation techniques for RS data. To solve the problem, this paper surveys more than 100 papers in this field in the past 5 years and elaborates in detail on the aspects of technical framework classification discussion, datasets, experimental evaluation, research challenges, and future research directions. Different from several previously published surveys, this paper first focuses on comprehensively summarizing the advantages and disadvantages of techniques and models based on the important and difficult points. This research will help beginners quickly establish research ideas and processes in this field, allowing them to focus on algorithm innovation without paying too much attention to datasets, evaluation indicators, and research frameworks.

A Review on Recent Deep Learning-Based Semantic Segmentation for Urban Greenness Measurement

“…There is another important component that appears in the review called the Feed Forward Network (FFN) or Multilayer Perceptron (MLP) structure, which is frequently used for deep learning architectures to transform a feature space into another feature space or decision space. Thus, almost all the DL architectures are basically constructed from these three components followed by nonlinear activation functions in different configurations with subsidiary operational components [65][66][67][68][69].…”

Hybrid Attention Fusion Embedded in Transformer for Remote Sensing Image Semantic Segmentation

“…Unlike conventional methods, deep learning models do not require manual feature design. They automatically learn feature representations from data, thereby increasing automation and possessing strong nonlinear modeling capabilities [9]. By employing deep learning architectures such as CNNs, it becomes possible to capture more finegrained local contextual information, effectively extracting the complex advanced features of objects on the ground and achieving more accurate semantic segmentation.…”

Comparison of Network Architectures for Semantic Segmentation

“…Semantic segmentation has been an important part of computer vision applications, such as autonomous driving [1], remote sensing [2,3], and medical image processing [4], which gave motivation for a dynamic development of the field in recent years.…”