Inception-Residual Block based Neural Network for Thermal Image Denoising
This addresses noise reduction in thermal imaging for applications like surveillance or medical diagnostics, but it appears incremental as it builds on existing inception and residual concepts.
The paper tackled thermal image denoising by proposing a neural network with denoising inception-residual blocks, achieving the best SQNR performance and reasonable processing time compared to state-of-the-art methods.
Thermal cameras show noisy images due to their limited thermal resolution, especially for the scenes of a low temperature difference. In order to deal with a noise problem, this paper proposes a novel neural network architecture with repeatable denoising inception-residual blocks(DnIRB) for noise learning. Each DnIRB has two sub-blocks with difference receptive fields and one shortcut connection to prevent a vanishing gradient problem. The proposed approach is tested for thermal images. The experimental results indicate that the proposed approach shows the best SQNR performance and reasonable processing time compared with state-of-the-art denoising methods.