Accession number BBBC009 · Version 1
This image set consists of five differential interference contrast (DIC) images of red bood cells.
There are two options when using this data set to quantify an algorithm's effectiveness. The simplest method is simply to compute the percentage of pixels that your algorithm and the ground truth have in common. However, a wide class of algorithms for handling DIC images do not explicity segment the image, but rather transform the image so that it can later be thresholded. In this case, to ensure that the algorithms are judged on their accuracy of transformation and not the separate problem of thresholding, it is best to find the area under the ROC curve generated by trying every possible thresholding value.
The images were acquired by Jeffrey Skerker and annotated by Tom Morgan.
To the extent possible under law, Roger Wiegand has waived all copyright and related or neighboring rights to Human Red Blood Cells DIC images. This work is published from: United States.
To the extent possible under law, Anne Carpenter has waived all copyright and related or neighboring rights to Human Red Blood Cells DIC ground truth. This work is published from: United States.