DeBCR: trained model weights for fluorescence microscopy image enhancement

Deep learning trained model weights deposition for the DeBCR framework for microscopy images restoration.

DeBCR is a deep learning framework for microscopy images restoration such as denoising and deconvolution (resolution enhancement). DeBCR implements the m-rBCR model - a multi-scale image restoration model, proposed by us earlier (Li, Kudryashev, Yakimovich, ECCV 2024, 2025).

How to cite

If you use our m-rBCR/DeBCR model (code/weights) or one of our DeBCR packages (API or GUI), please cite us as the following:

Li, R., Yushkevich, A., Chu, X. et al. DeBCR: a sparsity-efficient framework for image enhancement through a deep-learning-based solution to inverse problems. Commun Eng 5, 25 (2026). https://doi.org/10.1038/s44172-025-00582-4

Description

The previous version (v1) of this deposition contains the re-processed previously published datasets provided as training/test/validation data. In the current version (v2) we provide the respective weights of our deep learning model, trained on these data:

• Confocal data denoising (trained DeBCR model: LM_CARE_S.mediterranea.zip) - low/high exposure confocal dataset of Schmidtea mediterranea (Denoising_Planaria) from the publication of CARE network applied to fluorescent microscopy data (Weigert, Schmidt, Boothe et al., Nature Methods, 2018).
• Confocal data denoising (trained DeBCR model: LM_CARE_T.castaneum.zip) - low/high exposure confocal dataset of Tribolium castaneum (Denoising_Triboleum) from the publication of CARE network applied to fluorescent microscopy data (Weigert, Schmidt, Boothe et al., Nature Methods, 2018).
• Widefield data resolution enhancement (trained DeBCR model: WF_SIM_S.aureus.zip) - widefield/SIM dataset of Staphylococcus aureus from the deposition (Pereira & Pinho, Zenodo, 2021).
• Confocal data resolution enhancement (trained DeBCR model: CF_STED_FActin.zip) - confocal/STED dataset of F-actin from the deposition (Bouchard, Gagné & Lavoie-Cardinal, Zenodo, 2023).

Related records

The core code and Python API interface: debcr (PyPI), 10.5281/zenodo.17673859 (Zenodo).

The Napari GUI interface: napari-debcr (PyPI), 10.5281/zenodo.17674104 (Zenodo).