Fluorescence Microscopy Denoizing via Neighbor Linear Embedding

One of the difficulties in studying fluorescence imaging of biological structures is the presence of noise corruption. Even though hardware- and software-related technologies have undergone continual improvement, the unavoidable effect of a Poisson–Gaussian mixture type is generally encountered in fluorescence microscopy images. This noise should be mitigated to allow the extraction of valuable information from fluorescence images for various types of biological analysis. Thus, this study introduces a new and efficient learning-based denoising approach for fluorescence microscopy. The proposed approach is based mainly on linear transformations between noise-free and noisy submanifold structures of patch spaces, benefiting from linear neighbor embeddings of local image patches. According to visual and statistical results, the developed algorithm called the "neighbor linear-embedding denoising" algorithm has a highly competitive and generally superior performance in comparison with other algorithms used for fluorescence microscopy image denoizing in the literature. Cite this article as: C. Kirmiziay, B. Aydeniz, and M. Turkan, "Fluorescence microscopy denoising via neighbor linear embedding," Electrica, 24(1), 51-59, 2024.