BrainGenix-STS is the Scan Translation System division of the BrainGenix Department. The main goal of STS is the functional translation of architectural data from scans of brain tissue and the to-be-developed generic working model of the human brain.
GNU Affero General Public License v3.0
2
stars
2
forks
source link
Make a list of stitching technologies and compare their performance and accuracy on our dataset #11
After the DataLoader is done, we can then compare the stitching technologies. For this stage we'll only use one layer (and likely only part of one layer instead of the entire layer, since each layer has a lot of data and we don't need to use all of it for testing, at least not at this stage).
What we want to do is see which stitching technology is the most performant and accurate. We want accuracy over performance, but it may be useful to still retain the option for lower accuracy at higher performance (as long as the default option is for the highest accuracy possible even at the expense of performance).
Ultimately we want to produce a class or function which takes in two X, Y coordinate tuples (a start and end tuple), then produces an image within the specified 2D coordinate range. Or perhaps takes two X, Y, Z coordinate tuples and produces an array of layers within that area (which can then be stacked by another class).
After the DataLoader is done, we can then compare the stitching technologies. For this stage we'll only use one layer (and likely only part of one layer instead of the entire layer, since each layer has a lot of data and we don't need to use all of it for testing, at least not at this stage).
What we want to do is see which stitching technology is the most performant and accurate. We want accuracy over performance, but it may be useful to still retain the option for lower accuracy at higher performance (as long as the default option is for the highest accuracy possible even at the expense of performance).
Ultimately we want to produce a class or function which takes in two X, Y coordinate tuples (a start and end tuple), then produces an image within the specified 2D coordinate range. Or perhaps takes two X, Y, Z coordinate tuples and produces an array of layers within that area (which can then be stacked by another class).