Journal of Natural Resources, Volume. 35, Issue 4, 963(2020)
Detection of the construction land change in fine spatial resolution remote sensing imagery coupling spatial autocorrelation
Figures & Tables(19)
Fig. 2. True-color image and change of ground truth with 625 rows × 625 columns in the first area
Fig. 3. True-color image and change of ground truth with 625 rows × 625 columns in the second area
Fig. 4. True-color image and change of ground change truth with 1250 rows × 1250 columns in the third area
Fig. 6. Global Moran's
Fig. 7. Change distribution with the multi-scale local
Fig. 8. Change distribution with the multi-scale local
Fig. 9. Change distribution in the first area by different image feature combinations
Fig. 10. Change distribution in the second area by different image feature combinations
Fig. 11. Local autocorrelation statistics and change differences in the first and second areas
Table 1. Confusion matrix for change detection accuracy assessment
View tableView in ArticleTable 1. Confusion matrix for change detection accuracy assessment
真实情况 检测结果 正例 反例 正例 TP（真正例） FN（假反例） 反例 FP（假正例） TN（真反例）
Table 2. Global Moran's I, normalized Z value and P value of the change vector in the first area
View tableView in ArticleTable 2. Global Moran's I, normalized Z value and P value of the change vector in the first area
变化向量 全局Moran's I 原假设 标准化 Z 值P 值第一波段 0.87 随机分布 872.68 0.001 正态分布 872.68 0.001 第二波段 0.85 随机分布 853.73 0.001 正态分布 853.72 0.001 第三波段 0.87 随机分布 872.44 0.001 正态分布 872.45 0.001 第四波段 0.83 随机分布 831.66 0.001 正态分布 831.66 0.001
Table 3. Multi-scale parameters configuration
View tableView in ArticleTable 3. Multi-scale parameters configuration
特征组合 简称 特征组合 简称 Lag=1 多尺度1 Lag=1、2、3、4、5、6 多尺度6 Lag=1、2 多尺度2 Lag=1、2、3、4、5、6、7 多尺度7 Lag=1、2、3 多尺度3 Lag=1、2、3、4、5、6、7、8 多尺度8 Lag=1、2、3、4 多尺度4 Lag=1、2、3、4、5、6、7、8、9 多尺度9 Lag=1、2、3、4、5 多尺度5
Table 4. Accuracy assessment by the multi-scale local G statistics in the first and second areas (%)
View tableView in ArticleTable 4. Accuracy assessment by the multi-scale local G statistics in the first and second areas (%)
多尺度 区域1 区域2 查全率 查准率 F 1分数查全率 查准率 F 1分数1 68.6 82.9 75.1 72.6 76.2 74.4 2 72 85.1 78 77.6 78.6 78.1 3 76.3 85.9 80.1 81.2 79.9 80.6 4 79.4 87 83 84.1 81.6 82.9 5 81.9 87.7 84.7 85.7 84.2 84.9 6 84.2 87.5 85.8 86.3 86.7 86.5 7 87.6 88.9 88.3 88.6 90.1 89.4 8 84 87.4 85.7 86.9 88 87.4 9 84.3 84.8 84.6 86.1 87.7 86.9
Table 5. Feature combinations and codes
View tableView in ArticleTable 5. Feature combinations and codes
编号 分类用的图像特征 特征维度 A 图像四个光谱波段的变化向量 4 B 变化向量和局部Moran's I 指数32 C 变化向量和局部 G 指数32 D 变化向量和局部Geary's C 指数32 E 变化向量、局部Moran's I 指数和局部G 指数60 F 变化向量、局部Moran's I 指数和局部Geary'sC 指数60 G 变化向量、局部Geary's C 指数和局部G 指数60 H 变化向量和三种局部空间自相关指数 88
Table 6. Accuracy assessment by feature combinations in the first and second areas (%)
View tableView in ArticleTable 6. Accuracy assessment by feature combinations in the first and second areas (%)
编号 区域1 区域2 查全率 查准率 F 1分数查全率 查准率 F 1分数A 61.7 73.8 67.2 64.2 72.5 68.1 B 75.1 83.7 79.2 77.2 82.8 79.9 C 87.6 88.9 88.3 88.6 90.1 89.4 D 52.6 74.2 61.3 72.5 70.7 71.6 E 82.5 87.7 85 87.2 88.4 87.8 F 64.4 83.7 72.8 75.4 79.4 77.3 G 79.9 84 81.9 84.5 86.3 85.4 H 69.7 85.6 76.8 84.8 86.7 85.8
Table 7. Accuracy assessment of change detection in the third area (%)
View tableView in ArticleTable 7. Accuracy assessment of change detection in the third area (%)
特征 查全率 查准率 F 1分数变化向量 74.9 34.0 46.8 变化向量及多尺度局部 G 指数78.4 67.4 72.5
Tools
Get Citation
Copy Citation Text
Tao ZHANG, Hong FANG, Yu-chun WEI, Qi HU, Han-ze-yu XU. Detection of the construction land change in fine spatial resolution remote sensing imagery coupling spatial autocorrelation[J]. Journal of Natural Resources, 2020, 35(4): 963
Paper Information
Category:
Received: Feb. 27, 2019
Accepted: --
Published Online: Oct. 17, 2020
The Author Email: WEI Yu-chun (weiyuchun@njnu.edu.cn)
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20