avatar
Articles
165
Tags
34
Categories
0

Home
Archives
Tags
Categories
List
  • Music
  • Movie
Link
About
ALTNT's Hexo Blog
Home
Archives
Tags
Categories
List
  • Music
  • Movie
Link
About

Reconstructing NDVI time series in cloud-prone regions: A fusion-and-fit approach with deep learning residual constraint

Created2024-10-13|Updated2024-10-13
|Post Views:
Author: ALTNT
Link: http://blog.705553939.xyz/2024/10/13/crop_classification/Cloud_Removal/ISPRS-2024-Reconstructing-NDVI-time/
Copyright Notice: All articles in this blog are licensed under CC BY-NC-SA 4.0 unless stating additionally.
crop classificationcloud removal
Previous
周报10月12日
Next
Subfield-level crop yield mapping without ground truth data: A scale transfer framework
Related Articles
2024-09-12
A Deep Learning Framework under Attention Mechanism for Wheat Yield Estimation Using Remotely Sensed Indices in the Guanzhong Plain
2024-10-08
Temporal Convolutional Neural Network for the Classification of Satellite Image
2024-10-27
Self-Supervised Learning of Remote Sensing Scene Representations Using Contrastive Multiview Coding
2025-02-17
Climate-adaptive emergency crop monitoring in inaccessible regions with satellite imagery
2024-10-07
记录如何生产 CACM 数据集过程
2024-09-24
Towards interpreting multi-temporal deep learning models in crop mapping
avatar
ALTNT
Articles
165
Tags
34
Categories
0
Follow Me
Announcement
This is my Blog
Contents
  1. 1. Reconstructing NDVI time series in cloud-prone regions: A fusion-and-fit approach with deep learning residual constraint
    1. 1.0.1. 1. 空间分辨率
    2. 1.0.2. 2. 光谱波段
    3. 1.0.3. 3. 重访周期
    4. 1.0.4. 4. 数据应用领域
  2. 1.1. 引言
    1. 1.1.1. 填充的方法(依赖空间的相关性)
    2. 1.1.2. 拟合的方法(依赖时间的相关性)
    3. 1.1.3. 这两种方法的挑战
    4. 1.1.4. 解决方法:
    5. 1.1.5. 融合策略
      1. 1.1.5.1. (传统融合方法)
      2. 1.1.5.2. (基于深度学习的方法)
    6. 1.1.6. 本论文
  3. 1.2. 方法
    1. 1.2.1. 2.1. 残差约束融合模型 - ReCoF
      1. 1.2.1.1. 2.1.1. 网络架构 - 多维信息整合
      2. 1.2.1.2. 2.1.2. 残差约束 - 变化信息提取
        1. 1.2.1.2.1. (公式推导)
        2. 1.2.1.2.2. (方法所使用公式的最终推导结果形式)
        3. 1.2.1.2.3. (参数 a 的获取——可以直接预先设置)
        4. 1.2.1.2.4. (残差 的计算————使用深度学习模型来计算他)
        5. 1.2.1.2.5. (方程(11)的解释)
    2. 1.2.2. 2.2. 萨维茨基-高莱(Savitzky–Golay,SG)滤波器——拟合。
    3. 1.2.3. 2.3. Evaluation of fusion results
      1. 1.2.3.1. (均方根误差(RMSE))
      2. 1.2.3.2. (平均绝对误差(MAE))
      3. 1.2.3.3. (相对平均绝对误差(rMAE))
      4. 1.2.3.4. (Correlation Coefficient相关系数(CC)指数)
      5. 1.2.3.5. (边缘度量)
  4. 1.3. 3. Experimental design and dataset
    1. 1.3.1. 3.1. 研究地点和数据集。
      1. 1.3.1.1. (地点 1)
      2. 1.3.1.2. (地点 2 和地点 3)
      3. 1.3.1.3. (标签的获取)
      4. 1.3.1.4. (如何选择多云地区的参考图像)------这些地区都没有没云的时间段的图像作为参考图像吗?
    2. 1.3.2. 3.2. 实验设置
      1. 1.3.2.1. 3.2.1. 实验 I
      2. 1.3.2.2. 3.2.2. 实验二
      3. 1.3.2.3. 3.2.3. 实验 III
        1. 1.3.2.3.1. (验证方案)
      4. 1.3.2.4. 3.2.4. 实验 IV
    3. 1.3.3. 3.3. 与其他方法的比较
  5. 1.4. 4、结果
    1. 1.4.1. 4.1. 实验 I:理想观测条件下不同融合方法的比较
    2. 1.4.2. 4.2. 实验二:土地覆盖变化情况下时空融合的结果
    3. 1.4.3. 4.3. 实验 III:实际云区 NDVI 时间序列重建结果
    4. 1.4.4. 4.4. 实验四:多云地区大面积长时间序列数据重建结果
  6. 1.5. 5、讨论
    1. 1.5.1. 5.1. ReCoff 方法的优越性及其原理
    2. 1.5.2. 5.2. 消融实验
      1. 1.5.2.1. (进行是否应用残差约束的结果对比)
    3. 1.5.3. 5.3. 时间序列数据评估中相关系数(CC)的可靠性
    4. 1.5.4. 5.4. 不同分辨率的粗糙图像对融合结果的影响
    5. 1.5.5. 5.5. 重构时间序列数据的应用
    6. 1.5.6. 5.6. ReCoff 方法的局限性
  7. 1.6. 6、结论
Recent Post
周报2025年6月8日2025-06-09
The 20 m Africa rice distribution map of 20232025-06-04
DCM————DeepCropMapping: A multi-temporal deep learning approach with improved spatial generalizability for dynamic corn and soybean mapping2025-06-04
重新复现CACM记录2025-06-04
机器学习相关概念2025-05-26
©2020 - 2025 By ALTNT
Framework Hexo|Theme Butterfly