Improving Dataset Handling for Sentinel-1 and Sentinel-2 Images (#1)
Browse files- Improving Dataset Handling for Sentinel-1 and Sentinel-2 Images (e9615edd5bd54edb0850902cdd4ef517f6969d7b)
Co-authored-by: Raiden Williams <rwilliams@users.noreply.huggingface.co>
README.md
CHANGED
@@ -97,22 +97,53 @@ Ready to start using **[CloudSEN12](https://cloudsen12.github.io/)**?
|
|
97 |
|
98 |
<br>
|
99 |
|
100 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
101 |
|
102 |
<br>
|
103 |
|
|
|
104 |
**train shape: (8490, 512, 512)**
|
105 |
<br>
|
106 |
**val shape: (535, 512, 512)**
|
107 |
<br>
|
108 |
**test shape: (975, 512, 512)**
|
109 |
-
|
110 |
<br>
|
111 |
-
|
112 |
-
### **Example**
|
113 |
-
|
114 |
-
<br>
|
115 |
-
|
116 |
```py
|
117 |
import numpy as np
|
118 |
|
@@ -135,7 +166,6 @@ y = np.memmap('test/manual_hq.dat', dtype='int8', mode='r', shape=test_shape)
|
|
135 |
<br>
|
136 |
|
137 |
|
138 |
-
|
139 |
This work has been partially supported by the Spanish Ministry of Science and Innovation project
|
140 |
PID2019-109026RB-I00 (MINECO-ERDF) and the Austrian Space Applications Programme within the
|
141 |
**[SemantiX project](https://austria-in-space.at/en/projects/2019/semantix.php)**.
|
|
|
97 |
|
98 |
<br>
|
99 |
|
100 |
+
<be>
|
101 |
+
|
102 |
+
# **Dataset information, working with np.memmap:**
|
103 |
+
|
104 |
+
Sentinel-1 and Sentinel-2 collect images that span an area of 5090 x 5090 meters at 10 meters per pixel.
|
105 |
+
This results in 509 x 509 pixel images, presenting a challenge.
|
106 |
+
|
107 |
+
**Given each layer is a two-dimensional matrix, true image data is held from pixel (1,1) to (509,509)**
|
108 |
+
|
109 |
+
The subsequent images have been padded with three pixels around the image to make the images 512 x 512, a size that most models accept.
|
110 |
+
|
111 |
+
To give a visual representation of where the padding has been added:
|
112 |
+
x marks blank pixels stored as black (255)
|
113 |
+
|
114 |
+
xxxxxxxxxxxxxx
|
115 |
+
x xx
|
116 |
+
x xx
|
117 |
+
x xx
|
118 |
+
x xx
|
119 |
+
x xx
|
120 |
+
xxxxxxxxxxxxxx
|
121 |
+
xxxxxxxxxxxxxx
|
122 |
+
|
123 |
+
The effects of the padding can be mitigated by adding a random crop within (1,1) to (509, 509)
|
124 |
+
or completing a center crop to the desired size for network architecture.
|
125 |
+
|
126 |
+
### The current split of image data is into three categories:
|
127 |
+
|
128 |
+
- Training: 84.90 % of total
|
129 |
+
- Validation: 5.35 % of total
|
130 |
+
- Testing: 9.75 % of total
|
131 |
+
|
132 |
+
For the recomposition of the data to take random samples of all 10,000 available images,
|
133 |
+
we can combine the np.memmap objects and take random selections at the beginning of each trial,
|
134 |
+
selecting random samples of the 10,000 images based on the desired percentage of the total data available.
|
135 |
+
|
136 |
+
This approach ensures the mitigation of training bias based on the original selection of images for each category.
|
137 |
|
138 |
<br>
|
139 |
|
140 |
+
### **Example**
|
141 |
**train shape: (8490, 512, 512)**
|
142 |
<br>
|
143 |
**val shape: (535, 512, 512)**
|
144 |
<br>
|
145 |
**test shape: (975, 512, 512)**
|
|
|
146 |
<br>
|
|
|
|
|
|
|
|
|
|
|
147 |
```py
|
148 |
import numpy as np
|
149 |
|
|
|
166 |
<br>
|
167 |
|
168 |
|
|
|
169 |
This work has been partially supported by the Spanish Ministry of Science and Innovation project
|
170 |
PID2019-109026RB-I00 (MINECO-ERDF) and the Austrian Space Applications Programme within the
|
171 |
**[SemantiX project](https://austria-in-space.at/en/projects/2019/semantix.php)**.
|