Gee: use of common mask functions in remote sensing image processing [updatemask]

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 The mask correlation function in 3 individual , Respectively mask、updateMask and unmask, Need to differentiate and master 3 Usage of mask function

---

One 、mask() function

• ee.Image.mask()

  Get or set the mask of the image . The output image retains the metadata and footprints of the input image . Pixels whose mask changes from zero to another value will be filled with zero , Or the value closest to zero in the range of pixel types .

Mask image . If specified , Then copy the input image to the output , But the mask is given by the value of the image . If this is a single band , Then it is used for all bands in the input image . If not specified , Returns the image created from the mask of the input image , Zoom to range 0:1.

Be careful ： The version that sets the mask will be deprecated . To mask an image on a previously unmasked pixel , Please use Image.updateMask. To unshield previously masked pixels , Please use Image.unmask.

Usage method ：image.mask(), Will use the mask area and non mask area of the image 0 and 1 Distinguish , among 1 Represented by mask ,0 Represents unmasked . Within the frame of the image , Black represents the masked area , White represents the unmasked area .

Two 、updateMask()

ee.Image.updateMask()

Update the mask of the image in all positions where the existing mask is not zero . The output image retains the metadata and footprints of the input image .

New mask for image , As 0, 1 Floating point values in range . If this image has a band , It is used to input all bands in the image ; otherwise , It must have the same number of bands as the input image .

Usage method ：image.updateMask(),updateMask The image will be shown as 0 The area mask of , If you want one image to have the same mask as another , have access to image2.updateMask(image1.mask().not()).

3、 ... and 、unmask()

ee.Image.unmask()

  In all positions where the input mask is zero , Replace the mask and value of the input image with the mask and value of another image . The output image retains the metadata of the input image . By default , The output image also retains the imported footsteps , But will sameFootprint Set to false Allow extended footsteps .

Usage method ,image.unmask(number),unmask Its function is that it can replace the image value of the masked area with any value , The unmasked area remains the original value . for example image.unmask(-9999) The value in the masked area will be replaced by -9999, In this way, the image can be compared with arcgis Seamless compatibility .

Four 、 Case practice

python The code is as follows ：

（1） Image data

image = ee.Image('COPERNICUS/S2_SR/20210109T185751_20210109T185931_T10SEG')
trueColor = {
    'bands':['B4', 'B3', 'B2'], 
    'min':0, 'max':2800,
    'gamma':1.5
}

#  Load image 
Map = geemap.Map()
Map.addLayer(image, trueColor, 's2 image')
Map.centerObject(image)
Map

（2） Corresponding mask processing

# ********** Replace all mask areas with a certain value , Use unmask**********
# # ********** Mask out unqualified pixels , Use updateMask**********
#  utilize SWIR1 establish land Boolean band , among water by 0,land by 1
landMask = image.select('B11').gt(100)

#  Apply mask band to update the original image , It will make the mask on the image equal to 0 All pixels of become invalid values 
imageMasked = image.updateMask(landMask) #  Only reserved land

#  Load two images , landMask The image can be clearly divided into 2 Color ,0 For blue ,1 It's green 
Map.addLayer(landMask, {'palette': ['blue', 'lightgreen']}, 'Land mask')
Map.addLayer(imageMasked, trueColor, 'imageMasked')

# ********** Replace all mask areas with a certain value , Use unmask**********
#  Set invalid mask value , Replace the unmasked area with 32767 Null value appears 
imageUnmasked = imageMasked.unmask(32767)
Map.addLayer(imageUnmasked, trueColor, 'image unmasked')

#  Reset mask pixels to valid , Use the default value  0  fill , Enter footsteps 
maskResetFootprint = imageMasked.unmask()
Map.addLayer(maskResetFootprint, trueColor, 'maskResetFootprint')

#  Use something else 1 Image pixel value replacement 
fill = ee.Image('COPERNICUS/S2_SR/20200618T184919_20200618T190341_T10SEG')
imgFill = imageMasked.unmask(fill)
Map.addLayer(fill, trueColor, 'new s2')
Map.addLayer(imgFill, trueColor, 'image filled')

The results are illustrated below ：

 

---

5、 ... and 、 Summary

1. GEE in 3 A common mask function ,mask, updateMask and unmask, among mask As a function about to be discarded , It is not recommended to use too much . It can be used flexibly updateMask and Mask Realize data processing and analysis .
2. Using mask analysis , It can realize the magical function of transplanting flowers and trees , It is interesting to

---

Reference resources ：

1. https://developers.google.com/earth-engine/apidocs/ee-image-unmask?hl=en
2. https://blog.csdn.net/SunStrongInChina/article/details/110388401