TY - GEN
T1 - Bm3d Vs 2-Layer Onn
AU - Malik, Junaid
AU - Kiranyaz, Serkan
AU - Yamac, Mehmet
AU - Gabbouj, Moncef
N1 - jufoid=57423
PY - 2021
Y1 - 2021
N2 - Despite their recent success on image denoising, the need for deep and complex architectures still hinders the practical usage of CNNs. Older but computationally more efficient methods such as BM3D remain a popular choice, especially in resource-constrained scenarios. In this study, we aim to find out whether compact neural networks can learn to produce competitive results as compared to BM3D for AWGN image denoising. To this end, we conFigure networks with only two hidden layers and employ different neuron models and layer widths for comparing the performance with BM3D across different AWGN noise levels. Our results conclusively show that the recently proposed self-organized variant of operational neural networks based on a generative neuron model (Self-ONNs) is not only a better choice as compared to CNNs, but also provide competitive results as compared to BM3D and even significantly surpass it for high noise levels.
AB - Despite their recent success on image denoising, the need for deep and complex architectures still hinders the practical usage of CNNs. Older but computationally more efficient methods such as BM3D remain a popular choice, especially in resource-constrained scenarios. In this study, we aim to find out whether compact neural networks can learn to produce competitive results as compared to BM3D for AWGN image denoising. To this end, we conFigure networks with only two hidden layers and employ different neuron models and layer widths for comparing the performance with BM3D across different AWGN noise levels. Our results conclusively show that the recently proposed self-organized variant of operational neural networks based on a generative neuron model (Self-ONNs) is not only a better choice as compared to CNNs, but also provide competitive results as compared to BM3D and even significantly surpass it for high noise levels.
KW - AWGN
KW - Conferences
KW - Neurons
KW - Computer architecture
KW - Computational efficiency
KW - Convolutional neural networks
KW - Biological neural networks
KW - Image denoising
KW - operational neural networks
KW - self-organized operational neural networks
KW - discriminative learning
U2 - 10.1109/ICIP42928.2021.9506240
DO - 10.1109/ICIP42928.2021.9506240
M3 - Conference contribution
T3 - Proceedings : International Conference on Image Processing
SP - 1994
EP - 1998
BT - 2021 IEEE International Conference on Image Processing (ICIP)
PB - IEEE
T2 - IEEE International Conference on Image Processing
Y2 - 19 September 2021 through 22 September 2021
ER -