Pansharpen#

class deepinv.physics.Pansharpen(img_size, filter='bilinear', factor=4, srf='flat', noise_color=GaussianNoise(), noise_gray=GaussianNoise(), use_brovey=True, device='cpu', padding='circular', **kwargs)[source]#

Bases: StackedLinearPhysics

Pansharpening forward operator.

The measurements consist of a high resolution grayscale image and a low resolution RGB image, and are represented using deepinv.utils.TensorList, where the first element is the RGB image and the second element is the grayscale image.

By default, the downsampling is done with a gaussian filter with standard deviation equal to the downsampling, however, the user can provide a custom downsampling filter.

It is possible to assign a different noise model to the RGB and grayscale images.

Parameters:

  • img_size (tuple[int]) – size of the high-resolution multispectral input image, must be of shape (C, H, W).

  • filter (torch.Tensor, str, NoneType) – Downsampling filter. It can be ‘gaussian’, ‘bilinear’ or ‘bicubic’ or a custom torch.Tensor filter. If None, no filtering is applied.

  • factor (int) – downsampling factor/ratio.

  • srf (str, tuple, list) – spectral response function of the decolorize operator to produce grayscale from multispectral. See deepinv.physics.Decolorize for parameter options. Defaults to flat i.e. simply average the bands.

  • use_brovey (bool) – if True, use the Brovey method to compute the pansharpening, otherwise use the conjugate gradient method.

  • noise_color (torch.nn.Module) – noise model for the RGB image.

  • noise_gray (torch.nn.Module) – noise model for the grayscale image.

  • device (torch.device, str) – torch device.

  • padding (str) – options are 'valid', 'circular', 'replicate' and 'reflect'. If padding='valid' the blurred output is smaller than the image (no padding) otherwise the blurred output has the same size as the image.

Examples:

Pansharpen operator applied to a random 32x32 image:

>>> from deepinv.physics import Pansharpen
>>> import torch
>>> x = torch.randn(1, 3, 32, 32) # Define random 32x32 color image
>>> physics = Pansharpen(img_size=x.shape[1:], device=x.device)
>>> x.shape
torch.Size([1, 3, 32, 32])
>>> y = physics(x)
>>> y[0].shape
torch.Size([1, 3, 8, 8])
>>> y[1].shape
torch.Size([1, 1, 32, 32])
A_dagger(y: TensorList, **kwargs) → Tensor[source]#

If the Brovey method is used, compute the classical Brovey solution, otherwise compute the conjugate gradient solution.

See review paper for details.

Parameters:

y (TensorList) – input tensorlist of (MS, PAN)

Returns:

Tensor of image pan-sharpening using the Brovey method.

Examples using Pansharpen:#

Remote sensing with satellite images

Remote sensing with satellite images

A tour of forward sensing operators

A tour of forward sensing operators