1
$\begingroup$

I am training a Unet model for cell image segmentation from microscopy images. In order to help the model generalize better to different microscopes, I attempted to apply brightness augmentation to the training data, randomly multiplying images by a value chosen between (0.5, 1.5), then normalizing each image.

However, adding this augmentation always caused the model to get stuck at a very high loss (using jaccard loss). Interestingly, in the absence of this augmentation, the model converged to a jaccard score of 0.50. A similar effect was observed when applying random Gaussian blurrin: the model failed to learn. On the contrary, spatial augmentations such as flipping, rotating, etc did not have this effect.

I performed extensive testing on my code verified that the image augmentation function worked correctly and that the cells were still recognizable from blurred/darkened/brightened images.

I thought that the model cannot escape some local minima in the loss landscape. So, I experimented with a variety of weight initialization including HeNormal and HeUniform. I also tried first training first without augmentation and then using the weights to train with the brightness and blurring augmentations. Nothing worked. I am using Relu activation function and Adam optimizer. I tried to switch to SGD to see if it would help, but to no avail.

$\endgroup$

1 Answer 1

0
$\begingroup$

Random levels of brightness could make the results worse because brightness is often related to sharpness as far as I know.

A low brightness would generally have a low sharpness, and vice-versa. If you have a high brightness with a low sharpness, this would make more confusion during the learning process, and this would alter generalization.

Consequently, if you improve the sharpness, you would improve the results.

Therefore, you can measure it to confirm that images with a good sharpness are better than the other ones.

Then, you can use simple solutions like OpenCV to improve it, but if you have enough quality data, you can use VAE/GANs to improve sharpness greatly.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.