This website requires JavaScript.

LEDCNet: A Lightweight and Efficient Semantic Segmentation Algorithm Using Dual Context Module for Extracting Ground Objects from UAV Aerial Remote Sensing Images

Xiaoxiang HanYiman LiuGang LiuQiaohong Liu
Dec 2022
摘要
Semantic segmentation of UAV aerial remote sensing images provides a moreefficient and convenient surveying and mapping method for traditional surveyingand mapping. In order to make the model lightweight and improve a certainaccuracy, this research developed a new lightweight and efficient network forthe extraction of ground features from UAV aerial remote sensing images, calledLDMCNet. Meanwhile, this research develops a powerful lightweight backbonenetwork for the proposed semantic segmentation model. It is called LDCNet, andit is hoped that it can become the backbone network of a new generation oflightweight semantic segmentation algorithms. The proposed model uses dualmulti-scale context modules, namely the Atrous Space Pyramid Pooling module(ASPP) and the Object Context Representation module (OCR). In addition, thisresearch constructs a private dataset for semantic segmentation of aerialremote sensing images from drones. This data set contains 2431 training sets,945 validation sets, and 475 test sets. The proposed model performs well onthis dataset, with only 1.4M parameters and 5.48G floating-point operations(FLOPs), achieving an average intersection-over-union ratio (mIoU) of 71.12%.7.88% higher than the baseline model. In order to verify the effectiveness ofthe proposed model, training on the public datasets "LoveDA" and "CITY-OSM"also achieved excellent results, achieving mIoU of 65.27% and 74.39%,respectively.
展开全部
图表提取

暂无人提供速读十问回答

论文十问由沈向洋博士提出,鼓励大家带着这十个问题去阅读论文,用有用的信息构建认知模型。写出自己的十问回答,还有机会在当前页面展示哦。

Q1论文试图解决什么问题?
Q2这是否是一个新的问题?
Q3这篇文章要验证一个什么科学假设?
0
被引用
笔记
问答