Determine the benchmarking methods

devoworm / GSoC-2017

Repository for Google Summer of Code 2017 project

2 stars 1 forks source link

Determine the benchmarking methods #18

Open devoworm opened 7 years ago

devoworm commented 7 years ago

Possible approaches:

[x] code Euclidean distance based benchmark
- [ ] explore the use particle analysis algorithm in ImageJ
[x] explore more refined techniques.
[x] implement f-score based metrics

sedflix commented 7 years ago

We could do something like the following: We create a testing dataset by manually marking cell boundaries in some images(At least 10-20 different images). The marked cell boundaries will establish the ground truth.

Then, we form some kind of loss function(we could start with Euclidean Distance) to determine differences between the result given by our algorithms and the ground truth.

devoworm commented 7 years ago

Sid will get back with Bradly on issue. Feels comfortable with implementation.

sedflix commented 7 years ago

The current state of this issue is here

[ ] Documentation of it needs to be improved