RT Journal Article
SR Electronic
T1 QuickStitch for seamless stitching of confocal mosaics through high-pass filtering and recursive normalization
JF bioRxiv
FD Cold Spring Harbor Laboratory
SP 075440
DO 10.1101/075440
A1 Pavel A. Brodskiy
A1 Paulina M. Eberts
A1 Cody Narciso
A1 Jochen Kursawe
A1 Alexander Fletcher
A1 Jeremiah J. Zartman
YR 2016
UL http://biorxiv.org/content/early/2016/10/24/075440.abstract
AB Fluorescence micrographs naturally exhibit darkening around their edges (vignetting), which makes seamless stitching challenging. If vignetting is not corrected for, a stitched image will have visible seams where the individual images (tiles) overlap, introducing a systematic error into any quantitative analysis of the image. Although multiple vignetting correction methods exist, there remains no open-source tool that robustly handles large 2D immunofluorescence-based mosaic images. Here, we develop and validate QuickStitch, a tool that applies a recursive normalization algorithm to stitch large-scale immunofluorescence-based mosaics without incurring vignetting seams. We demonstrate how the tool works successfully for tissues of differing size, morphology, and fluorescence intensity. QuickStitch requires no specific information about the imaging system. It is provided as an open-source tool that is both user friendly and extensible, allowing straightforward incorporation into existing image processing pipelines. This enables studies that require accurate segmentation and analysis of high-resolution datasets when parameters of interest include both cellular-level phenomena and larger tissue-level regions of interest.