Cheat sheet of Land cover land use classification using python and others.
Introduction
filepath and open tif file
1
2
3
4
5
6
7
8
9
10
| import rasterio
l8_path = '../data/landsat_image.tif'
labels_path = '../data/labels_image.tif'
full_dataset = rasterio.open(l8_path)
img_rows, img_cols = full_dataset.shape
img_bands = full_dataset.count
print(full_dataset.shape)
print(full_dataset.count)
|
Open our raster dataset
1
2
3
4
|
import rasterio
landsat_dataset = rasterio.open(l8_path)
|
How many bands does this image have?
1
2
| num_bands = landsat_dataset.count
print('Number of bands in image: {n}\n'.format(n=num_bands))
|
How many rows and columns?
1
2
| rows, cols = landsat_dataset.shape
print('Image size is: {r} rows x {c} columns\n'.format(r=rows, c=cols))
|
What driver was used to open the raster?
1
2
| driver = landsat_dataset.driver
print('Raster driver: {d}\n'.format(d=driver))
|
What is the raster’s projection?
1
2
3
| proj = landsat_dataset.crs
print('Image projection:')
print(proj)
|
Open up the dataset and read into memory
1
2
| landsat_image = landsat_dataset.read()
landsat_image.shape
|
NDVI Calculate
1
2
3
4
| bandNIR = landsat_image[4, :, :]
bandRed = landsat_image[3, :, :]
ndvi = np.clip((bandNIR.astype(float) - bandRed.astype(float)) / (bandNIR.astype(float) + bandRed.astype(float)), -1,1)
|
Histogram
1
2
3
4
5
6
7
| import matplotlib.pyplot as plt
plt.hist(x = '需要处理的数据',bins = '需要分割的精度')
plt.xlabel('x轴标签')
plt.ylabel('y轴标签')
plt.title('总标题')
plt.show()
|