结合全卷积神经网络与条件随机场的资源3号遥感影像云检测 
下载： 1118次
Cloud Detectionof ZY-3 Remote Sensing Images Based on Fully Convolutional Neural Network and Conditional Random Field
辽宁工程技术大学测绘与地理科学学院, 辽宁 阜新 123000
图 & 表
图 1. 所提算法流程图
Fig. 1. Flow chart of proposed algorithm
下载图片 查看原文
图 2. 全卷积神经网络原理示意图
Fig. 2. Principle diagram of fully convolutional network
下载图片 查看原文
图 3. 输入影像与标签图片。(a)输入影像1;(b)标签图片1;(c)输入影像2;(d)标签图片2
Fig. 3. Input images and label pictures. (a) Input image 1; (b) label picture 1; (c) input image 2; (d) label picture 2
下载图片 查看原文
图 4. 不同方法云检测结果对比。(a)输入影像;(b)标签影像;(c) FCN+CRF;(d) MyFCN;(e) FCN-8s;(f) FCM+SVM
Fig. 4. Comparison of cloud detection results with different methods. (a) Input image; (b) label picture; (c) FCN+CRF; (d) MyFCN; (e) FCN-8s; (f) FCM+SVM
下载图片 查看原文
表 1改进的全卷积神经网络算法结构代码对应表
Table1. Structural code mapping in improved fully convolutional neural network algorithm
| From top to bottom | From bottom to top |
|---|
| Conv1_1-64n-3×3k-‘relu’ | | | Conv1_2-64n-3×3k-‘relu’ | | | Pool1-2×2 max pooling | Deconv9-64n-3×3k-‘relu’ | | Conv2_1-128n-3×3k-‘relu’ | Conv8_1-128n-3×3k-‘relu’ | | Conv2_2-128n-3×3k-‘relu’ | Conv8_2-128n-3×3k-‘relu’ | | Pool2-2×2 max pooling | Merge8 (conv2_2, Deconv8) | | Conv3_1-256n-3×3k-‘relu’ | Deconv8-128n-3×3k-‘relu’ | | Conv3_2-256n-3×3k-‘relu’ | Conv7_2-256n-3×3k-‘relu’ | | Pool3-2×2 max pooling | Conv7_1-256n-3×3k-‘relu’ | | Conv4_1-512n-3×3k-‘relu’ | Merge7 (conv3_2, Deconv7) | | Conv4_2-512n-3×3k-‘relu’ | Conv7-256n-3×3k-‘relu’ | | Drop4-dropout (0.5) | Conv6_2-512n-3×3k-‘relu’ | | Pool4-2×2 max pooling | Conv6_1-512n-3×3k-‘relu’ | | Conv5_1-1024n-3×3k-‘relu’ | Merge6 (drop4, Deconv6) | | Conv5_2-1024n-3×3k-‘relu’ | Deconv6-512n-3×3k-‘relu’ |
|
查看原文
表 2不同方法的云检测精度与检测时间比较
Table2. Comparison of accuracy and detection time of cloud area detection by different methods
| Item | FCN+CRF | MyFCN | FCN-8s | FCM+SVM |
|---|
| Raccuracy /% | 97.38 | 90.11 | 83.96 | 83.72 | | Rprecision /% | 99.70 | 96.92 | 94.15 | 84.89 | | Rrecall /% | 96.77 | 90.36 | 85.35 | 84.61 | | /% | 98.22 | 93.53 | 89.53 | 84.75 | | Detection time /s | - | 0.46 | 2.24 | >60 |
|
查看原文
裴亮, 刘阳, 高琳. 结合全卷积神经网络与条件随机场的资源3号遥感影像云检测[J]. 激光与光电子学进展, 2019, 56(10): 102802. Liang Pei, Yang Liu, Lin Gao. Cloud Detectionof ZY-3 Remote Sensing Images Based on Fully Convolutional Neural Network and Conditional Random Field[J]. Laser & Optoelectronics Progress, 2019, 56(10): 102802.
PDF全文
