Errors and quality metrics in super-resolution and beyond by Siân Culley
Автор: GloBIAS
Загружено: 2025-01-31
Просмотров: 46
Описание:
Presentation at the first in-person workshop of GloBIAS in November 2024 in Gothenburg, Sweden. Part of the session on Precision and errors in bioimage analysis (practical) - Co-Chairs: Rocco D'Antuono & Ana Stojiljkovic
Fluorescence microscopy allows biologists to access quantitative information about the spatiotemporal organisation of samples. This requires analysis of good-quality data, be that raw images from a microscope or processed images such as the results of super-resolution or denoising image processing. However, it can be very difficult to tell if whether an image does in fact contain high quality information that will support this downstream analysis. Here we discuss our previous work developing the SQUIRREL image quality metrics for super-resolution microscopy, and our current work in assessing commonly used quality metrics for image processing methods such as deep learning-based denoising. We demonstrate metric performance for a range of different tasks, and how well this assessment correlates with biological information present in images.
Повторяем попытку...
Доступные форматы для скачивания:
Скачать видео
-
Информация по загрузке: