xmipp_transform_normalize (v3.0)


Change the range of intensity values of pixels. In general, most of the methods requires a background to separate particles from noise


-i, --input <input_file>
Input file: metadata, stack, volume or image.
--mode <mode=overwrite>
Metadata writing mode.
where <mode> can be:
  • overwrite Replace the content of the file with the Metadata
  • append Write the Metadata as a new block, removing the old one
--label <image_label=image>
Label to be used to read/write images.
-o, --output <output_file=>
Output file: metadata, stack, volume or image.
--oroot <root=>
Rootname of output individual images. Output image format can be set adding extension after rootname as ":ext".
for 2D-images: do not apply transformation stored in the header
--method <mth=NewXmipp>
Normalizing method.
where <mth> can be:
  • OldXmipp I=(I-m(I))/stddev(I) Avg(I)=0, Stddev(I)=1, does not need background
  • Near_OldXmipp I=(I-m(I))/stddev(bg)
  • NewXmipp I=(I-m(bg))/stddev(bg) Avg(bg)=0, Stddev(I)=1 Positivity constraints can be added in reconstruction
  • Tomography I=(I-mean(I))/(stddev(I)*cos(tilt)) does not need background, it assumes the tilt series is vertically aligned Similar to OldXmipp but with an extra division by the cos(tilt)
  • Tomography0 I=(I-mean(I(0 degrees)))/(stddev(I)*cos(tilt)) does not need background Similar to Tomography but the average at 0 degrees is used for all images
  • NewXmipp2 I=(I-m(bg))/(m(I)-m(bg))
  • Michael I=(I-m(bg))/stddev(bg)
  • None Used for removing only dust
  • Random I=aI+b
  • Ramp Subtract ramp and then NewXmipp
  • Neighbour Replace pixels in the background with random noise
Invert contrast.
--thr_black_dust <sblack=-3.5>
Remove black dust particles with sigma threshold sblack.
--thr_white_dust <swhite=3.5>
Remove white dust particles with sigma threshold swhite.
--thr_neigh <value=1.2>
Sigma threshold for neighbour removal.
--prm <a0> <aF> <b0> <bF>
Requires --method Random. I=aI+b.
Apply a mask depending on the tilt requires --method Tomography or Tomography0
--background <mode>

where <mode> can be:
    • frame <r> Rectangular background of r pixels.
    • circle <r> Circular background outside radius r.
or --mask <mask_type=circular>
Use an alternative type of background mask.
where <mask_type> can be:
  • circular <R> circle/sphere mask if R>0 => outside R if R<0 => inside R
  • DWT_circular <R> <smin> <smax> circle/sphere mask smin and smax define the scales to be kept
  • rectangular <Xrect> <Yrect> <Zrect=-1> 2D or 3D rectangle if X,Y,Z > 0 => outside rectangle if X,Y,Z < 0 => inside rectangle
  • crown <R1> <R2> 2D or 3D crown if R1,R2 > 0 => outside crown if R1,R2 < 0 => inside crown
  • cylinder <R> <H> 2D circle or 3D cylinder if R,H > 0 => outside cylinder if R,H < 0 => inside cylinder
  • cone <theta> 3D cone (parallel to Z) if theta > 0 => outside cone if theta < 0 => inside cone
  • wedge <th0> <thF> 3D missing-wedge mask for data collected between tilting angles th0 and thF (around the Y-axis)
  • binary_file <binary_file> Read from file
--center <x0=0> <y0=0> <z0=0>
mask center

Examples and notes

Normalize using OldXmipp method
xmipp_transform_normalize -i images.sel --method OldXmipp
Normalize 64x64 images using NewXmipp method
xmipp_transform_normalize -i images.sel --method NewXmipp --background circle 29
Normalize 64x64 images using NewXmipp method and a crown mask
xmipp_transform_normalize -i images.sel --method NewXmipp --mask crown 29 32
Normalize a volume to have zero mean and unit variance
xmipp_transform_normalize -i volume.vol

User's comments