Filtering Pixel Latent Variables for Unmixing Noisy and Undersampled Volumetric Images

arXiv:2312.05357v2 Announce Type: replace-cross
Abstract: The development of robust signal unmixing algorithms is essential for leveraging multimodal datasets acquired through a wide array of scientific imaging technologies, including hyperspectral or time-resolved acquisitions. In experimental physics, enhancing the spatio-temporal resolution or expanding the number of detection channels often leads to diminished sampling rate and signal-to-noise ratio, significantly affecting the efficacy of signal unmixing algorithms. We propose applying band-pass filters to the latent space of a multi-dimensional convolutional neural network to disentangle …

abstract acquired acquisitions algorithms array arxiv channels cs.cv datasets detection development eess.iv experimental filtering images imaging leads multimodal physics pixel resolution robust scientific signal technologies temporal through type variables