Impact of different loss functions on denoising of microscopic images
- Resource Type
- Conference
- Authors
- Shah, Zafran Hussain; Muller, Marcel; Hammer, Barbara; Huser, Thomas; Schenck, Wolfram
- Source
- 2022 International Joint Conference on Neural Networks (IJCNN) Neural Networks (IJCNN), 2022 International Joint Conference on. :1-10 Jul, 2022
- Subject
- Bioengineering
Computing and Processing
General Topics for Engineers
Power, Energy and Industry Applications
Robotics and Control Systems
Signal Processing and Analysis
Frequency-domain analysis
Microscopy
Noise reduction
Superresolution
Neural networks
Computer architecture
Convolutional neural networks
- Language
- ISSN
- 2161-4407
The denoising of super-resolution structured illumination microscopy (SR-SIM) images using deep convolutional neural networks (CNNs) has re-ceived much attention in recent years. The contrast and fidelity of biological cell structures in denoised SR-SIM images critically depend on the CNN architecture and the chosen loss function. In this work, we propose two new combinations of loss functions for denoising SR-SIM images with low signal-to-noise ratio. The first combination consists of a loss component computed in the frequency domain by applying Fast Fourier Transform (FFT) to the image data and of pixel-wise loss. The second combination combines the loss in the frequency domain with a feature-based loss. In a series of experiments, these new loss functions are compared with various traditional loss functions for image de-noising. The results show that CNNs trained with the novel losses are superior in terms of PSNR and SSIM values. Moreover, critical structures of biological cells are better preserved in the denoised network output. Further results on the BSD dataset are also promising.