An Optimization Method of RS-Image Land Surface Classification Based on Unet-SVM Coupling Model

NIE Yan; JIANG Peng-fei; BIAN Fang; JIA Fang-yuan

doi:10.3969/j.issn.2096-6091.2023.18.002

您当前的位置：

首页 >

文章列表页 >

An Optimization Method of RS-Image Land Surface Classification Based on Unet-SVM Coupling Model

RESEARCH PAPER | 更新时间：2025-03-26

- An Optimization Method of RS-Image Land Surface Classification Based on Unet-SVM Coupling Model
- New Generation of Information Technology Vol. 6, Issue 18, Pages: 7-12(2023)
- 作者机构：
  
  1.中科光启空间信息技术有限公司遥感信息中心，河南郑州 450000
  2.中国联合网络通信有限公司郑州分公司，河南郑州 450000
- 作者简介：
- 基金信息：
- DOI：10.3969/j.issn.2096-6091.2023.18.002
  CLC： TP391
- Published：30 September 2023
- 稿件说明：
移动端阅览
聂岩, 蒋鹏飞, 边防, 等. 基于Unet和SVM耦合的遥感影像地物分类优化改进研究[J]. 新一代信息技术, 2023, 6(18): 07-12

NIE Yan, JIANG Peng-fei, BIAN Fang, et al. An Optimization Method of RS-Image Land Surface Classification Based on Unet-SVM Coupling Model[J]. New Generation of Information Technology, 2023, 6(18): 07-12
聂岩, 蒋鹏飞, 边防, 等. 基于Unet和SVM耦合的遥感影像地物分类优化改进研究[J]. 新一代信息技术, 2023, 6(18): 07-12 DOI： 10.3969/j.issn.2096-6091.2023.18.002.

NIE Yan, JIANG Peng-fei, BIAN Fang, et al. An Optimization Method of RS-Image Land Surface Classification Based on Unet-SVM Coupling Model[J]. New Generation of Information Technology, 2023, 6(18): 07-12 DOI： 10.3969/j.issn.2096-6091.2023.18.002.

摘要

基于语义分割网络的遥感地物分类是近年来遥感解译所采用的新兴手段之一，其中Unet网络因结构简单、训练高效和易于构建等而成为最经典的语义分割模型。本文以最大限度扩增遥感数据信息量为前提，通过纹理、光谱、语义三方面特征提升数据信息丰度，主要工作包括：（1）引入Resnet残差模块对Unet网络模型加以改进；（2）采用窗口滚动输入输出方法解决数据喂入和相邻瓦块间“缝隙”问题；（3）采用数据串联耦合方法实现数据的三个方面特征融合。实验表明，改进Unet模型能在防止模型退化和保留纹理特征信息的前提下进一步加深网络层，提高了网络提取丰富语义特征的能力。数据串联耦合保留了数据的高层语义特征、纹理高频信息和数据辐射光谱信息，提升了SVM模型的分类精度，达到了良好的地物识别分割效果。

Abstract

The classification of terrain objects in remote sensing data based on semantic segmentation model is becoming one of the newest methods adopted by RS interpretion task. Among those models

the Unet is likely to be the widely used one because of which of the succinct web structure

efficient training and easily construction. This paper aims to try hardest to strengthen the volume of RS data information through the way of promoting the volume of information in three aspects called texture

spectrum and semantics in data. The mainly jobs done among which contains as follows: (1) improve the classic Unet model by adopting the Resnet modules; (2) solve the problems of data-feed way and "chink" between adjacent data tiles; (3) adopt data series-coupled method to realize the integration of the three features mentioned above. The consequence of experiment indicates that the advanced Unet model can hold deeper data-transform layers but at the same time prevent model degeneration and keep texture information transmission

and thus imporve the medel's capability of recognizing abundant semantic features. and that the method of data series-coupled integrates all of three of the high-level semantic feature

high-frequency texture feature and primary spectral information feature

and thus to help imporve the SVM model's precision of classification to achieve a better effect in terrain objects segmentation task.

关键词

Keywords

references

张永宏 , 夏广浩 , 阚希 , 等 . 基于全卷积神经网络的多源高分辨率遥感道路提取 [J ] . 计算机应用 , 2018 , 38 ( 7 ): 2070 - 2075 .

LI W J , DONG R M , FU H H , et al . Large-scale oil palm tree detection from high-resolution satellite images using two-stage convolutional neural networks [J ] . Remote Sensing , 2018 , 11 ( 1 ): 11 - 31 .

刘尚旺 , 崔智勇 , 李道义 . 基于Unet网络多任务学习的遥感图像建筑地物语义分割 [J ] . 国土资源遥感 , 2020 , 32 ( 4 ): 74 - 83 .

施文灶 , 刘金清 . 基于邻域总变分和势直方图函数的高分辨率遥感影像建筑物提取 [J ] . 计算机应用 , 2017 , 37 ( 6 ): 1787 - 1792 .

SUN X F , LIN X G , SHEN S H , et al . High-resolution remote sensing data classification over urban areas using random forest ensemble and fully connected conditional random field [J ] . ISPRS International Journal of Geo-Information , 2017 , 6 ( 8 ): 245 - 271 .

李潇凡 , 王胜强 , 翁轩 , 等 . 基于UNet深度学习算法的东海大型漂浮藻类遥感监测 [J ] . 光学学报 , 2021 , 41 ( 2 ): 0201002 .

YUAN J . Automatic building extraction in aerial scenes using convolution networks [EB/OL ] . ( 2016-02-21 )[ 2023-12-15 ] . http://arxiv.org/abs/1602.06564 http://arxiv.org/abs/1602.06564 .

马广迪 , 杨为琛 . 基于CNN-CBIR的遥感图像分类检索方法‍ [J ] . 北京测绘 , 2021 , 35 ( 5 ): 634 - 639 .

王曦 , 于鸣 , 任洪娥 . UNET与FPN相结合的遥感图像语义分割 [J ] . 液晶与显示 , 2021 , 36 ( 3 ): 475 - 483 .

王庆涛 . 基于深度学习的遥感水体信息提取研究 [D ] . 开封 : 河南大学 , 2020 : 3 - 4 .

王曦 , 于鸣 , 任洪娥 . UNET与FPN相结合的遥感图像语义分割 [J ] . 液晶与显示 , 2021 , 36 ( 3 ): 475 - 483 .

张小芸 . 基于改进的Unet地表覆盖遥感图像语义分割算法研究 [D ] . 南昌 : 南昌大学 , 2022 : 30 - 36 .

邝辉宇 , 吴俊君 . 基于深度学习的图像语义分割技术研究综述 [J ] . 计算机工程与应用 , 2019 , 55 ( 19 ): 12 - 21, 42 .

刘钊 , 赵桐 , 廖斐凡 . 基于卷积神经网络预测结果的缝隙修复算法研究 [J ] . 遥感技术与应用 , 2021 , 36 ( 2 ): 247 - 255 .

周春江 , 徐丽华 . 遥感影像镶嵌中拼接缝消除方法研究 [J ] . 西南大学学报(自然科学版) , 2010 , 32 ( 9 ): 136 - 140 .

甘岚 , 沈鸿飞 , 王瑶 , 等 . 基于改进DCGAN的数据增强方法 [J ] . 计算机应用 , 2021 , 41 ( 5 ): 1305 - 1313 .

LI D C , WEN I H . A genetic algorithm-based virtual sample generation technique to improve small data set learning [J ] . Neurocomputing , 2014 , 143 ( 2 ): 222 - 230 .

Views

167

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

No data

Related Author

No data

Related Institution

No data

⁰