欢迎访问中国科学院大学学报,今天是

中国科学院大学学报 ›› 2026, Vol. 43 ›› Issue (1): 93-103.DOI: 10.7523/j.ucas.2024.019

• 电子信息与计算机科学 • 上一篇    下一篇

分块尺寸对大场景遥感图像空间超分应用的影响分析

孙瑞奇1,2, 张文娟1(), 李震1, 马雪松3, 梅君林1,2   

  1. 1.中国科学院空天信息创新研究院,北京 100094
    2.中国科学院大学,北京 100049
    3.中国矿业大学(北京),北京 100083
  • 收稿日期:2024-01-19 修回日期:2024-04-03 发布日期:2024-05-09
  • 通讯作者: 张文娟
  • 基金资助:
    国家自然科学基金(42201503)

Effect of patch size on super-resolution of large scene remote sensing images

Ruiqi SUN1,2, Wenjuan ZHANG1(), Zhen LI1, Xuesong MA3, Junlin MEI1,2   

  1. 1.Aerospace Information Research Institute,Chinese Academy of Sciences,Beijing 100094,China
    2.University of Chinese Academy of Sciences,Beijing 100049,China
    3.China University of Mining and Technology-Beijing,Beijing 100083,China
  • Received:2024-01-19 Revised:2024-04-03 Published:2024-05-09
  • Contact: Wenjuan ZHANG

摘要:

以高空间分辨率(高分)遥感图像为对象,采用3种典型超分模型,开展9组不同分块尺寸实验,研究其对整图超分精度和效率的影响。综合分析结果如下:1)分块超分导致拼接缝,当分块较小时,拼接缝呈现斑块效应,不一致性更为显著;2)随着分块尺寸增大,模型的超分精度和计算效率均有所提高,当分块比大于1时,耗时与精度趋于稳定;3)整图输入的计算可行性、精度和模型关系密切。ESPCN模型在整图输入时精度最优,RDBPN模型由于图像非方阵而导致精度下降,HSENET模型对算力要求较高,无法进行整图计算。综上,该研究为遥感超分工程化应用的分块尺寸选取提供了实验依据。

关键词: 深度学习, 超分辨率, 大场景遥感图像, 空间分辨率, 输入尺寸, 分块尺寸

Abstract:

Single image super-resolution (SISR) can improve the resolution of remote sensing images (RSIs), thereby improving the application value of data. At present, the number of pixels of RSIs generally reaches hundreds of millions, and it is usually necessary to divide the image into patches when performing SISR. However, there is a lack of relevant research on how to effectively determine the patch size and whether different sizes affect the results. In this paper, taking a large-scale high-resolution RSIs as the experiment data, 3 typical SISR models are selected, 9 groups of SR experiments under different patch sizes are carried out, and the super-resolution (SR) results for the whole of the large-scaled RSI are analyzed comprehensively both qualitatively and quantitatively. The results show that: 1) Cutting of the patches results in stitching seams at the stitching place. In particular, when the patch size is small, a large number of stitching seams show a block effect and the inconsistency is more obvious. 2) With the increase of the patch size, the SR accuracy of the three models is improved, and the overall computational efficiency is also improved. When the test patches are larger than the training patches, the elapsed time and accuracy stabilize. 3) The feasibility and accuracy of the whole RSI input are closely related to the model. The ESPCN model has the best accuracy when inputting the whole RSI, the RDBPN model may cause the accuracy to decrease due to the non-square matrix of the RSI, and the HSENET model has high requirements for computing power and cannot calculate the whole RSI. In conclusion, this paper provides an experimental basis for the selection of patch size for RSI SR engineering applications.

Key words: deep learning, super-resolution, large scene remote sensing images, spatial resolution, input size, patch size

中图分类号: