#!/usr/bin/env python
# coding: utf-8

# <a href="https://githubtocolab.com/gee-community/geemap/blob/master/examples/notebooks/141_image_array_viz.ipynb" target="_parent"><img src="https://colab.research.google.com/assets/colab-badge.svg" alt="Open in Colab"/></a>
# 
# **Visualizing in-memory raster datasets and image arrays**

# In[ ]:


# !pip install -U geemap


# In[ ]:


import ee
import geemap


# Use a Landsat image covering San Francisco Bay Area.

# In[ ]:


m = geemap.Map()
image = ee.Image("LANDSAT/LC08/C02/T1_TOA/LC08_044034_20140318").select(
    ['B5', 'B4', 'B3']
)
vis_params = {
    'min': 0.0,
    'max': 0.4,
}
m.add_layer(image, vis_params, 'False Color (543)')
m.centerObject(image)
m


# Retrieve the projection information from the image.

# In[ ]:


projection = image.select(0).projection().getInfo()
crs = projection['crs']
crs


# Read the Earth Engine image into an Xarray DataArray.

# In[ ]:


ds = geemap.ee_to_xarray(
    image, 
    crs="EPSG:32610", 
    scale=300,
    geometry=image.geometry(),
    ee_mask_value=0
)
ds


# Save the DataArray as a Cloud Optimized GeoTIFF.

# In[ ]:


geemap.xee_to_image(ds, filenames=['landsat.tif'])


# Calculate the NDVI and save it to the DataArray.

# In[ ]:


ndvi = (ds['B5'] - ds['B4']) / (ds['B5'] + ds['B4'])


# Create an in-memory raster dataset from the DataArray.

# In[ ]:


ndvi_image = geemap.array_to_image(ndvi, source="landsat.tif")


# Visualize the in-memory raster dataset.

# In[ ]:


m.add_raster('landsat.tif', indexes=[1, 2, 3], nodata=-1, layer_name="Landsat 7")
m.add_raster(ndvi_image, colormap="Greens", layer_name="NDVI")
m


# Save the in-memory raster dataset to a GeoTIFF file.

# In[ ]:


geemap.array_to_image(ndvi, output="ndvi.tif", source="landsat.tif")