Gee: remote sensing image composite and mosaic

---

To learn, you must be rigid and constant , If it's not rigid, it's easy , If it doesn't last, it will fall back .—— The song dynasty . Fengzixian

---

Preface

GEE, How to realize the synthesis of remote sensing images （composite） And inlay （mosaic）？ What's the difference between them ？

The analysis area is as follows ：

---

One 、 Analysis steps

Main analysis steps ：

Two 、python Code

1、 Image synthesis Composite

python The code is as follows
Be careful ： synthesis composite It mainly reflects the synthesis of multi-phase images in the same area

Map = geemap.Map()
#  Load image data 
naip2004_2012 = ee.ImageCollection('USDA/NAIP/DOQQ') \
    .filterBounds(ee.Geometry.Point(-71.09, 42.4)) \
    .filterDate('2004-07-01', '2012-12-31') \
    .select(['R', 'G', 'B'])
#  Maximum value synthesis 
composite = naip2004_2012.max()
Map.addLayer(composite, {
    }, 'max value composite')
Map.setCenter(-71.09, 42.4, 14)
Map

give the result as follows ：

2、 Image mosaic Mosaic

• ee.ImageCollection.mosaic()
  
  Be careful ： set mosaic It mainly reflects the synthesis operation of images in different regions
  You need to splice four different DOQQ, But in different places

python The code is as follows

#  Load span 4 Data of images 
naip2012 = ee.ImageCollection('USDA/NAIP/DOQQ') \
    .filterBounds(ee.Geometry.Rectangle([-71.18, 42.35, -71.09, 42.41])) \
    .filterDate('2012-01-01', '2012-12-31')
#  Image mosaic mosaic
mosaic = naip2012.mosaic()
Map.addLayer(naip2012, {
    }, 'naip2012')
Map.setCenter(-71.12532, 42.3712, 12)

give the result as follows ：

3、 Dig further Mosaic Characteristics of

a key ：mosaic() Methods according to their order in the set （ Finally at the top ） Composite overlapping image . To control tessellation （ Or synthesis ） The source of pixels in , Image masks can be used mask.
for example , The following uses spectral threshold to mosaic new image data

python The code is as follows ：

#  Load image , Show 
naip = ee.Image('USDA/NAIP/DOQQ/m_4207148_nw_19_1_20120710')
Map.setCenter(-71.0915, 42.3443, 12)
Map.addLayer(naip, {
    }, 'naip')

#  Generate NDVI、NDWI Index 
ndvi = naip.normalizedDifference(['N', 'R']).rename('ndvi')
ndwi = naip.normalizedDifference(['G', 'N']).rename('ndwi')

#  Use the above 2 Indices , Use different threshold conditions , The detection area is naked 
#  Naked land 1, Vegetation index is lower than 0.2, At the same time, the water index is lower than 0.3 Region 
bare1 = ndvi.lt(0.2).And(ndwi.lt(0.3))
#  Naked land 2, Vegetation index is lower than 0.2, At the same time, the water index is lower than 0.8 Region . It can be seen that its sensitivity is lower than the previous threshold 
bare2 = ndvi.lt(0.2).And(ndwi.lt(0.8))

#  Display parameters 
vis_ndvi = {
    'min':-1, 'max':1, 'palette':['red', 'green']}
vis_ndwi = {
    'min':0.5, 'max':1, 'palette':['gray', 'blue']}

#  Load display 
Map.addLayer(ndvi, vis_ndvi, 'ndvi')
Map.addLayer(ndwi, vis_ndwi, 'ndwi')
Map.addLayer(bare1, {
    }, 'bare1')
Map.addLayer(bare2, {
    }, 'bare2')

Show results ：

#  Mask and embed visual images , You can see that the last layer is on the top （ This is a Mosaic Characteristics ）
mosaic = ee.ImageCollection([
    ndwi.updateMask(ndwi.gte(0.5)).visualize(**vis_ndwi), 
    ndvi.updateMask(ndvi.gte(0.2)).visualize(**vis_ndvi),
    bare2.updateMask(bare2.And(bare1.Not())).visualize(**{
    'palette':['gray']}),
    bare1.updateMask(bare1).visualize(**{
    'palette':['white']})
]).mosaic()
Map.addLayer(mosaic, {
    }, 'visualization mosaic')

give the result as follows ：

---

---

3、 ... and 、 Summary

1. Two different influence synthesis methods of image synthesis and image mosaic are analyzed , among Image synthesis composite It is more manifested in the combination of images in the same area at different times , The most commonly used synthesis methods are maximization synthesis （MVC,max value composite）、 Median synthesis （median）、 Mean synthesis （mean） And other synthetic methods , In specific work, different synthesis methods can be adopted according to actual needs ; set （mosaic） More performance in the synthesis of multiple images in different areas .
2. The feature of mosaic is that the last layer is on the top , This is also Mosaic One of the characteristics of .

---

Reference resources ：

• https://developers.google.com/earth-engine/apidocs/ee-image-visualize?hl=en
• https://github.com/giswqs/earthengine-py-notebooks/blob/master/ImageCollection/mosaicking.ipynb
• https://developers.google.com/earth-engine/guides/ic_composite_mosaic?hl=en