# Faster brightness/contrast augmentation?
**URL:** https://discourse.itk.org/t/faster-brightness-contrast-augmentation/5145
**Category:** Algorithms
**Tags:** python, simpleitk
**Created:** [June 30, 2022, 9:06am UTC](https://discourse.itk.org/t/faster-brightness-contrast-augmentation/5145 "2022-06-30T09:06:53Z")
**Posts on this page:** 7
**Page:** 1
### Author:  [@Floowey](https://discourse.itk.org/u/Floowey)
#### Post date: [June 30, 2022, 9:06am UTC](https://discourse.itk.org/t/faster-brightness-contrast-augmentation/5145/1 "2022-06-30T09:06:53Z")
Hello,
I intend to alter the contrast/brightness of my images in my pytorch Dataset pipeline using the `sitk.AdaptiveHistogramEqualization` method to make my neural network more robust against such differences. However, I found that it decreases my step rate from 5 it/s to just 1 it/s, which in turn increases my training time from a few days to weeks which I can’t afford. Am I perhaps misusing the method or is there a different, faster alternative?
The dataset looks like this
```auto
class CTChangeContrastDataset(Dataset):
def __init__ (self, dataset: Dataset):
self.data = dataset
def __len__ (self):
return len(self.data)
def __getitem__ (self, idx):
im, label = self.data[idx]
rndAlpha = uniform(0,1)
rndBeta = uniform(0,1)
new_im = sitk.AdaptiveHistogramEqualization(im, alpha=rndAlpha, beta=rndBeta)
return new_im, label
```
---
### Author:  [@dzenanz](https://discourse.itk.org/u/dzenanz)
#### Post date: [June 30, 2022, 9:50am UTC](https://discourse.itk.org/t/faster-brightness-contrast-augmentation/5145/2 "2022-06-30T09:50:51Z")
You are re-doing the equalization each time an image is fetched. Of course it is slow. All fancy augmentations are slow. You might consider doing this augmentation with some probability, e.g. 10%. That should allow your NN to learn it and be resistant to it, but would not be overly computationally expensive.
---
### Author:  [@zivy](https://discourse.itk.org/u/zivy)
#### Post date: [June 30, 2022, 12:56pm UTC](https://discourse.itk.org/t/faster-brightness-contrast-augmentation/5145/3 "2022-06-30T12:56:22Z")
Hello @Floowey,
Deterministic, computationally complex augmentations are best done using a caching mechanism so that they are only performed once during training. For example, see the [MONAI implementations](https://github.com/Project-MONAI/MONAI/blob/dev/monai/data/dataset.py#L619-L809) for `PersistentDataset` and `CacheDataset`.
---
### Author:  [@blowekamp](https://discourse.itk.org/u/blowekamp)
#### Post date: [June 30, 2022, 1:20pm UTC](https://discourse.itk.org/t/faster-brightness-contrast-augmentation/5145/4 "2022-06-30T13:20:29Z")
For the size of images used for deep learning it should not be that time consuming. How long does just the AdaptiveHistogramEqualizationFilter take the filter to run? 4 seconds seems way too long.
What is your image size, pixel type and version of SimpleITK/ITK?
The current implementation using a moving histogram which should be fairly efficient.
---
### Author:  [@Floowey](https://discourse.itk.org/u/Floowey)
#### Post date: [June 30, 2022, 1:39pm UTC](https://discourse.itk.org/t/faster-brightness-contrast-augmentation/5145/5 "2022-06-30T13:39:40Z")
Thanks for your replies. Simply not doing it in every step is totally reasonable.
@zivy thank you, I will take a look at those. I’ll do a big part of my data preparation before training already, but maybe this could improve it too - gotta hope my machine can handle it.
@blowekamp My images are 35x35x35 floats and I’m using SITK version 2.1.1. One call takes in the order of 0.33-0.40 seconds. I’m sorry if I didn’t say it clearly, but the numbers I gave are the iterations/steps per second that decrease. One step loads 4 images.
---
### Author:  [@blowekamp](https://discourse.itk.org/u/blowekamp)
#### Post date: [June 30, 2022, 1:56pm UTC](https://discourse.itk.org/t/faster-brightness-contrast-augmentation/5145/6 "2022-06-30T13:56:51Z")
> [@Floowey](#):
>
> @blowekamp My images are 35x35x35 floats and I’m using SITK version 2.1.1. One call takes in the order of 0.33-0.40 seconds. I’m sorry if I didn’t say it clearly, but the numbers I gave are the iterations/steps per second that decrease. One step loads 4 images.
Not sure if that timing is just the Eq step, but this is what I am getting locally:
```auto
In [12]: img = sitk.Image([35,35,35], sitk.sitkFloat32)
In [13]: img = sitk.Noise(img)
In [14]: %timeit -n 1 out = sitk.AdaptiveHistogramEqualization(img, alpha=1, beta=1, radius=(3,3,3))
4.78 ms ± 693 µs per loop (mean ± std. dev. of 7 runs, 1 loop each)
```
---
### Author:  [@Floowey](https://discourse.itk.org/u/Floowey)
#### Post date: [June 30, 2022, 3:40pm UTC](https://discourse.itk.org/t/faster-brightness-contrast-augmentation/5145/7 "2022-06-30T15:40:46Z")
I couldn’t quite manage to reprodue that timeit in visual studio I timed it only around the AdaptiveHistogramEqualization call, but I noticed I get float64 from a previous normalization step. I doubt that it would make that much of a difference, but I’ll change it anyways. Thank you!
For now I’ll wait and see whether I get satisfying results just not doing it every time.