# Normalization of MRI images to 1

**URL:** <https://discourse.slicer.org/t/normalization-of-mri-images-to-1/20232>\
**Category:** Support\
**Tags:** normalisation\
**Created:** [October 19, 2021, 11:49am UTC](https://discourse.slicer.org/t/normalization-of-mri-images-to-1/20232 "2021-10-19T11:49:21Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![Hadi](https://avatars.discourse-cdn.com/v4/letter/h/e99b99/32.png) [@Hadi](https://discourse.slicer.org/u/Hadi)\
**Post date:** [October 19, 2021, 11:49am UTC](https://discourse.slicer.org/t/normalization-of-mri-images-to-1/20232/1 "2021-10-19T11:49:22Z")

</div>

Hi,  
I am working on MRI images of BraTs2020 image dataset. I want to normalized them to 1. I use the NormalizedImageFilter, But this does not happen and it is often seen in numbers greater than 1 (rage 2-3).  
Please guide me to this.  
Thanks.

---

<div class="post-metadata">

**Author:** ![lassoan](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/lassoan/32/13_2.png) [@lassoan](https://discourse.slicer.org/u/lassoan)\
**Post date:** [October 20, 2021, 2:51am UTC](https://discourse.slicer.org/t/normalization-of-mri-images-to-1/20232/2 "2021-10-20T02:51:57Z")

</div>

You can change the intensity range of an image easily. For example, you can cast the image to float or double using Cast Scalar Volume module (an integer valued volume normalized to 0-1 would end up all black), and then you can use Simple Filters module → RescaleIntensityImageFilter).

However, this kind of image normalization and augmentation is typically done as part of your neural network training and not as a preprocessing step - see TorchIO and MONAI examples.

---

<div class="post-metadata">

**Author:** ![Hadi](https://avatars.discourse-cdn.com/v4/letter/h/e99b99/32.png) [@Hadi](https://discourse.slicer.org/u/Hadi)\
**Post date:** [October 20, 2021, 6:07pm UTC](https://discourse.slicer.org/t/normalization-of-mri-images-to-1/20232/3 "2021-10-20T18:07:49Z")

</div>

Thank you for your answer.  
It is works very well.