GridBasedImageResampling

Resamples an image according to a resampling grid

Description

This application allows performing image resampling from an input resampling grid.

Parameters

Input and output data

This group of parameters allows setting the input and output images.

Input image -io.in image Mandatory
The input image to resample

Output Image -io.out image [dtype] Mandatory
The resampled output image

Resampling grid parameters

Input resampling grid -grid.in image Mandatory
The resampling grid

Grid Type -grid.type [def|loc] Default value: def
allows one to choose between two grid types

  • Displacement grid: $G(x_out,y_out) = (x_in-x_out, y_in-y_out)$
    A deformation grid contains at each grid position the offset to apply to this position in order to get to the corresponding point in the input image to resample

  • Localisation grid: $G(x_out,y_out) = (x_in, y_in)$
    A localisation grid contains at each grid position the corresponding position in the input image to resample

Output Image parameters

Parameters of the output image

Upper Left X -out.ulx float Default value: 0
X Coordinate of the upper-left pixel of the output resampled image

Upper Left Y -out.uly float Default value: 0
Y Coordinate of the upper-left pixel of the output resampled image

Size X -out.sizex int Mandatory
Size of the output resampled image along X (in pixels)

Size Y -out.sizey int Mandatory
Size of the output resampled image along Y (in pixels)

Pixel Size X -out.spacingx float Default value: 1
Size of each pixel along X axis

Pixel Size Y -out.spacingy float Default value: 1
Size of each pixel along Y axis

Default value -out.default float Default value: 0
The default value to give to pixel that falls outside of the input image.


Interpolation -interpolator [nn|linear|bco] Default value: bco
This group of parameters allows one to define how the input image will be interpolated during resampling.

  • Nearest Neighbor interpolation
    Nearest neighbor interpolation leads to poor image quality, but it is very fast.

  • Linear interpolation
    Linear interpolation leads to average image quality but is quite fast

  • Bicubic interpolation

Bicubic interpolation options

Radius for bicubic interpolation -interpolator.bco.radius int Default value: 2
This parameter allows controlling the size of the bicubic interpolation filter. If the target pixel size is higher than the input pixel size, increasing this parameter will reduce aliasing artifacts.


Available RAM (MB) -ram int Default value: 256
Available memory for processing (in MB).

Examples

From the command-line:

otbcli_GridBasedImageResampling -io.in ROI_IKO_PAN_LesHalles_sub.tif -io.out ROI_IKO_PAN_LesHalles_sub_resampled.tif uint8 -grid.in ROI_IKO_PAN_LesHalles_sub_deformation_field.tif -out.sizex 256 -out.sizey 256 -grid.type def

From Python:

import otbApplication

app = otbApplication.Registry.CreateApplication("GridBasedImageResampling")

app.SetParameterString("io.in", "ROI_IKO_PAN_LesHalles_sub.tif")
app.SetParameterString("io.out", "ROI_IKO_PAN_LesHalles_sub_resampled.tif")
app.SetParameterOutputImagePixelType("io.out", 1)
app.SetParameterString("grid.in", "ROI_IKO_PAN_LesHalles_sub_deformation_field.tif")
app.SetParameterInt("out.sizex", 256)
app.SetParameterInt("out.sizey", 256)
app.SetParameterString("grid.type","def")

app.ExecuteAndWriteOutput()

See also

otbStereorecificationGridGeneration