Skip to content

Robust NDWI Calculation with Edge-Protected Gaussian Filteringprotected blurring #73

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
Vishesh0305 wants to merge 1 commit into
base: master
Choose a base branch
from
Open

Robust NDWI Calculation with Edge-Protected Gaussian Filteringprotected blurring #73

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
13 changes: 7 additions & 6 deletions ndwi_labels.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -69,17 +69,18 @@ def get_ndwi_label(image_path, points_path, ksize=100, blurring=True):
green = src_raster.read(2).astype(np.float32) # Get the green band
nir_num = src_raster.count # Adjusting NIR band to 4 or 5 band images
nir = src_raster.read(nir_num).astype(np.float32) # Get NIR band

ndwi_profile =src_raster.profile
np.seterr(divide='ignore', invalid='ignore')

ndwi = (green - nir) / (green + nir) # NDWI equation
ndwi[np.isnan(ndwi)] = 0 # Sets any NaN values in the NDWI array to 0. (Dividing by zero => NaN pixels)
ndwi_profile = src_raster.profile # Copies the image profile (metadata).
with np.errstate(divide= 'ignore', invalid='ignore'):
ndwi = (green - nir) / (green + nir) # NDWI equation(Calculation with scientific robustness)
nodata_mask = np.isnan(ndwi) | (green== 0 ) # Identifying NoData areas (where band are 0 or NDWI is NaN )
ndwi= np.nan_to_num(ndwi, nan=0.0) # Clean the NaN values Before blurring.

# Apply Gaussian blur
if blurring:
print("Gaussian Filtering Applied")
print("Gaussian Filtering Applied with Edge Protecto")
ndwi = cv2.GaussianBlur(ndwi, KSIZE_BLUR, SIGMA_X, SIGMA_Y)
ndwi[nodata_mask] = -1.0 #Set NoData to -1 (distinct from 0/water)

# Blank label layer
label = np.zeros((src_raster.height, src_raster.width)).astype(np.uint8)
Expand Down