# PoissonNoise

### *class* deepinv.physics.PoissonNoise(gain=1.0, normalize=True, clip_positive=False, rng=None)

Bases: [`NoiseModel`](https://deepinv.org/api/stubs/deepinv.physics.NoiseModel.html.md#deepinv.physics.NoiseModel)

Poisson noise $y = \mathcal{P}(\frac{x}{\gamma})$
with gain $\gamma>0$.

If `normalize=True`, the output is multiplied by the gain, i.e., $\tilde{y} = \gamma y$.

* **Parameters:**
  * **gain** (*Union* *[*[*float*](https://docs.python.org/3.12/library/functions.html#float) *,* [*torch.Tensor*](https://docs.pytorch.org/docs/stable/tensors.html#torch.Tensor) *]*) – gain of the noise.
  * **normalize** ([*bool*](https://docs.python.org/3.12/library/functions.html#bool)) – normalize the output.
  * **clip_positive** ([*bool*](https://docs.python.org/3.12/library/functions.html#bool)) – clip the input to be positive before adding noise.
    This may be needed when a NN outputs negative values e.g. when using leaky ReLU.
  * **rng** ([*torch.Generator*](https://docs.pytorch.org/docs/stable/generated/torch.Generator.html#torch.Generator) *,* *None*) – (optional) a pseudorandom random number generator for the parameter generation.

#### NOTE
Poisson noise is only defined for non-negative inputs.
When used in combination with physics operators that can produce negative outputs (such as [`deepinv.physics.BlurFFT`](https://deepinv.org/api/stubs/deepinv.physics.BlurFFT.html.md#deepinv.physics.BlurFFT)), it is recommended to set `clip_positive=True` to avoid runtime errors.

#### TIP
All [pretrained denoisers](https://deepinv.org/user_guide/reconstruction/denoisers.html.md#denoisers) in the library can be re-used for Poisson denoising
using the [`Anscombe transform`](https://deepinv.org/api/stubs/deepinv.models.AnscombeDenoiser.html.md#deepinv.models.AnscombeDenoiser).

<hr />

* **Examples:**
  Adding Poisson noise to a physics operator by setting the `noise_model`
  attribute of the physics operator:
  ```pycon
  >>> from deepinv.physics import Denoising, PoissonNoise
  >>> import torch
  >>> physics = Denoising()
  >>> physics.noise_model = PoissonNoise()
  >>> x = torch.rand(1, 1, 2, 2)
  >>> y = physics(x)
  ```

#### forward(x, gain=None, seed=None, \*\*kwargs)

Adds the noise to measurements x

* **Parameters:**
  * **x** ([*torch.Tensor*](https://docs.pytorch.org/docs/stable/tensors.html#torch.Tensor)) – measurements
  * **gain** (*None* *,* [*float*](https://docs.python.org/3.12/library/functions.html#float) *,* [*torch.Tensor*](https://docs.pytorch.org/docs/stable/tensors.html#torch.Tensor)) – gain of the noise. If not None, it will overwrite the current noise level.
  * **seed** ([*int*](https://docs.python.org/3.12/library/functions.html#int)) – the seed for the random number generator, if `rng` is provided.
* **Returns:**
  noisy measurements

<a id="sphx-glr-backref-deepinv-physics-poissonnoise"></a>

## Examples using `PoissonNoise`:

<div class="sphx-glr-thumbnails">
<!-- thumbnail-parent-div-open --><div class="sphx-glr-thumbcontainer" tooltip="This example introduces the Richardson-Lucy algorithm for deconvolution in the blind setting, where both the underlying clean image and the blur kernel are unknown.">![](auto_examples/blind-inverse-problems/images/thumb/sphx_glr_demo_blind_richardsonlucy_thumb.png)

[Blind Richardson-Lucy deconvolution](https://deepinv.org/auto_examples/blind-inverse-problems/demo_blind_richardsonlucy.html.md)

  <div class="sphx-glr-thumbnail-title">Blind Richardson-Lucy deconvolution</div>
</div><div class="sphx-glr-thumbcontainer" tooltip="This example shows how to perform inference on and fine-tune the Reconstruct Anything Model (RAM) foundation model terris2025reconstruct to solve inverse problems.">![](auto_examples/models/images/thumb/sphx_glr_demo_foundation_model_thumb.png)

[Inference and fine-tune a foundation model](https://deepinv.org/auto_examples/models/demo_foundation_model.html.md)

  <div class="sphx-glr-thumbnail-title">Inference and fine-tune a foundation model</div>
</div><div class="sphx-glr-thumbcontainer" tooltip="This example demonstrates how to solve Poisson inverse problems using the Maximum-Likelihood Expectation-Maximization (MLEM) algorithm sheppMaximumLikelihoodReconstruction1982, also known as the Richardson-Lucy algorithm in the deconvolution setting richardsonBayesianBasedIterativeMethod1972,lucyIterativeTechniqueRectification1974.">![](auto_examples/optimization/images/thumb/sphx_glr_demo_poisson_mlem_thumb.png)

[Poisson Inverse Problems with Maximum-Likelihood Expectation-Maximization (MLEM)](https://deepinv.org/auto_examples/optimization/demo_poisson_mlem.html.md)

  <div class="sphx-glr-thumbnail-title">Poisson Inverse Problems with Maximum-Likelihood Expectation-Maximization (MLEM)</div>
</div><div class="sphx-glr-thumbcontainer" tooltip="This example demonstrates how to denoise images corrupted by Poisson-Gaussian noise using the Generalized Anscombe Transform (GAT), which converts any Gaussian denoiser into a Poisson-Gaussian denoiser makitalo2012optimal.">![](auto_examples/physics/images/thumb/sphx_glr_demo_anscombe_thumb.png)

[Poisson-Gaussian Denoising with the Generalized Anscombe Transform](https://deepinv.org/auto_examples/physics/demo_anscombe.html.md)

  <div class="sphx-glr-thumbnail-title">Poisson-Gaussian Denoising with the Generalized Anscombe Transform</div>
</div><div class="sphx-glr-thumbcontainer" tooltip="This example provides a tour of some of the forward operators implemented in DeepInverse. We restrict ourselves to operators where the signal is a 2D image. The full list of operators can be found in here.">![](auto_examples/physics/images/thumb/sphx_glr_demo_physics_tour_thumb.png)

[Tour of forward sensing operators](https://deepinv.org/auto_examples/physics/demo_physics_tour.html.md)

  <div class="sphx-glr-thumbnail-title">Tour of forward sensing operators</div>
</div><div class="sphx-glr-thumbcontainer" tooltip="This is a simple example to show how to use a mirror descent algorithm for solving an inverse problem with Poisson noise. See hurault2023convergent for more details.">![](auto_examples/plug-and-play/images/thumb/sphx_glr_demo_PnP_mirror_descent_thumb.png)

[Plug-and-Play algorithm with Mirror Descent for Poisson noise inverse problems.](https://deepinv.org/auto_examples/plug-and-play/demo_PnP_mirror_descent.html.md)

  <div class="sphx-glr-thumbnail-title">Plug-and-Play algorithm with Mirror Descent for Poisson noise inverse problems.</div>
</div><div class="sphx-glr-thumbcontainer" tooltip="This example shows you how to train a denoiser network in a fully self-supervised way, i.e., using noisy images only via the Neighbor2Neighbor loss, which exploits the local correlation of natural images.">![](auto_examples/self-supervised-learning/images/thumb/sphx_glr_demo_n2n_denoising_thumb.png)

[Self-supervised denoising with the Neighbor2Neighbor loss.](https://deepinv.org/auto_examples/self-supervised-learning/demo_n2n_denoising.html.md)

  <div class="sphx-glr-thumbnail-title">Self-supervised denoising with the Neighbor2Neighbor loss.</div>
</div><div class="sphx-glr-thumbcontainer" tooltip="This code shows how to restore a single image corrupted by Poisson noise using Poisson2Sparse, without requiring external training or knowledge of the noise level.">![](auto_examples/self-supervised-learning/images/thumb/sphx_glr_demo_poisson2sparse_thumb.png)

[Poisson denoising using Poisson2Sparse](https://deepinv.org/auto_examples/self-supervised-learning/demo_poisson2sparse.html.md)

  <div class="sphx-glr-thumbnail-title">Poisson denoising using Poisson2Sparse</div>
</div><div class="sphx-glr-thumbcontainer" tooltip="This example shows you how to train a denoiser network in a fully self-supervised way, using noisy images only via the Generalized Recorrupted2Recorrupted (GR2R) loss monroy2025generalized, which exploits knowledge about the noise distribution. You can change the noise distribution by selecting from predefined noise models such as Gaussian, Poisson, and Gamma noise.">![](auto_examples/self-supervised-learning/images/thumb/sphx_glr_demo_r2r_denoising_thumb.png)

[Self-supervised denoising with the Generalized R2R loss.](https://deepinv.org/auto_examples/self-supervised-learning/demo_r2r_denoising.html.md)

  <div class="sphx-glr-thumbnail-title">Self-supervised denoising with the Generalized R2R loss.</div>
</div><div class="sphx-glr-thumbcontainer" tooltip="This example shows you how to train a denoiser network in a fully self-supervised way, i.e., using noisy images only via the SURE loss, which exploits knowledge about the noise distribution.">![](auto_examples/self-supervised-learning/images/thumb/sphx_glr_demo_sure_denoising_thumb.png)

[Self-supervised denoising with the SURE loss.](https://deepinv.org/auto_examples/self-supervised-learning/demo_sure_denoising.html.md)

  <div class="sphx-glr-thumbnail-title">Self-supervised denoising with the SURE loss.</div>
</div>
<!-- thumbnail-parent-div-close --></div>