summary |
shortlog | log |
commit |
commitdiff |
tree
first ⋅ prev ⋅ next
Greg Burri [Tue, 28 Mar 2017 20:53:42 +0000 (22:53 +0200)]
Cleaning syntax.
Greg Burri [Thu, 3 Mar 2016 12:11:28 +0000 (13:11 +0100)]
Piaz files wasn't associated during setup.
Greg Burri [Fri, 19 Feb 2016 18:36:51 +0000 (19:36 +0100)]
Replace "for [..] in [..]" by "for [..] to [..]" for consistency.
Greg Burri [Fri, 5 Feb 2016 08:54:44 +0000 (09:54 +0100)]
Script modification for testing.
Greg Burri [Fri, 5 Feb 2016 00:41:55 +0000 (01:41 +0100)]
Update report.
Greg Burri [Wed, 3 Feb 2016 13:18:27 +0000 (14:18 +0100)]
Save predefined PPI and sensor sizes in JSON files.
Greg Burri [Tue, 2 Feb 2016 23:01:45 +0000 (00:01 +0100)]
Peculiar elements overlap RBC now.
Greg Burri [Mon, 1 Feb 2016 16:07:06 +0000 (17:07 +0100)]
Remove the removing of cells bases on standard deviation.
Greg Burri [Mon, 1 Feb 2016 09:02:08 +0000 (10:02 +0100)]
Change the way the dark stain is marked.
Greg Burri [Sat, 30 Jan 2016 22:03:10 +0000 (23:03 +0100)]
Doc + cleaning.
Greg Burri [Sat, 30 Jan 2016 18:06:03 +0000 (19:06 +0100)]
Update report.
Greg Burri [Sat, 30 Jan 2016 17:11:17 +0000 (18:11 +0100)]
Adjustment of some parameters.
Greg Burri [Sat, 30 Jan 2016 00:15:39 +0000 (01:15 +0100)]
Increase a bit the number of built ellipses.
Greg Burri [Fri, 29 Jan 2016 23:44:49 +0000 (00:44 +0100)]
More homogeneous ellipse density.
Greg Burri [Fri, 29 Jan 2016 21:58:21 +0000 (22:58 +0100)]
Cleaning and some little tweaks.
Greg Burri [Fri, 29 Jan 2016 21:51:48 +0000 (22:51 +0100)]
Update report.
Greg Burri [Thu, 28 Jan 2016 22:40:40 +0000 (23:40 +0100)]
Ajout de la section sur la granulometrie.
Greg Burri [Thu, 28 Jan 2016 08:20:51 +0000 (09:20 +0100)]
Add a DPI calculator to help to find the correct image resolution.
Greg Burri [Tue, 26 Jan 2016 23:28:56 +0000 (00:28 +0100)]
* Modify a bit the parasite nucleus detection sensibility.
* Add a setup.
Greg Burri [Tue, 26 Jan 2016 13:55:21 +0000 (14:55 +0100)]
Split the module 'ImgTools' in many modules.
Greg Burri [Mon, 25 Jan 2016 23:15:47 +0000 (00:15 +0100)]
Cleaning + renaming.
Greg Burri [Mon, 25 Jan 2016 23:14:15 +0000 (00:14 +0100)]
Add report.
Greg Burri [Mon, 25 Jan 2016 16:12:09 +0000 (17:12 +0100)]
Remove the parasite detection function from Ma.
Greg Burri [Mon, 25 Jan 2016 15:01:36 +0000 (16:01 +0100)]
Project the colors to have the best contrast for RBCs and parasites analyze.
Greg Burri [Mon, 25 Jan 2016 00:11:13 +0000 (01:11 +0100)]
Add a way to detect the membrane of a parasite in the ring stage.
Greg Burri [Fri, 22 Jan 2016 11:38:55 +0000 (12:38 +0100)]
Add some GUI elements :
- A warning if not enough erytrocytes
- The list of manually modifier erytrocytes
- Image names
- Exportation des résultats
Greg Burri [Thu, 21 Jan 2016 16:34:55 +0000 (17:34 +0100)]
Cleaning.
Greg Burri [Thu, 21 Jan 2016 12:50:59 +0000 (13:50 +0100)]
Add an option to change the brightness of the highlight box color (healthy/infected) (Can't be changed with the UI).
Greg Burri [Wed, 20 Jan 2016 21:19:04 +0000 (22:19 +0100)]
* Add logos CHUV and HES-SO.
* Code documentation.
Greg Burri [Wed, 20 Jan 2016 15:47:30 +0000 (16:47 +0100)]
Use a maybe monad to be able abort an analysis.
Greg Burri [Wed, 20 Jan 2016 09:40:15 +0000 (10:40 +0100)]
Force the user to input a DPI value for new image.
Greg Burri [Wed, 20 Jan 2016 08:51:17 +0000 (09:51 +0100)]
Add documentation.
Greg Burri [Wed, 20 Jan 2016 06:59:38 +0000 (07:59 +0100)]
Fix a bug during area opening/closing when the area value may become negative.
Greg Burri [Tue, 19 Jan 2016 13:08:13 +0000 (14:08 +0100)]
The minimum area computation wasn't correct.
Greg Burri [Tue, 19 Jan 2016 09:56:46 +0000 (10:56 +0100)]
Fix some approximation issues.
Greg Burri [Mon, 18 Jan 2016 16:33:57 +0000 (17:33 +0100)]
Add a status bar.
Greg Burri [Mon, 18 Jan 2016 12:22:20 +0000 (13:22 +0100)]
Add a logger assembly and split the main assembly in two : the UI and the parasitemia evaluation.
Greg Burri [Sun, 17 Jan 2016 18:39:37 +0000 (19:39 +0100)]
Add an about window.
Greg Burri [Sun, 17 Jan 2016 17:00:46 +0000 (18:00 +0100)]
* Add an exact method to compute an ellipse from three points and two tangents.
* Simplification of the matching ellipses process.
Greg Burri [Fri, 15 Jan 2016 10:39:42 +0000 (11:39 +0100)]
Set an application icon.
Greg Burri [Fri, 15 Jan 2016 09:13:54 +0000 (10:13 +0100)]
Use two radius in the configuration, one computed with the image resolution and one by granulometry.
Greg Burri [Thu, 14 Jan 2016 18:00:23 +0000 (19:00 +0100)]
The images to be analyzed can be selected.
Greg Burri [Wed, 13 Jan 2016 21:49:40 +0000 (22:49 +0100)]
Use real unit (um and ppi) instead of pixel in the parameters.
Greg Burri [Wed, 13 Jan 2016 13:44:16 +0000 (14:44 +0100)]
Improve readability of highlighted RBCs.
Greg Burri [Wed, 13 Jan 2016 12:30:53 +0000 (13:30 +0100)]
Add a way to pan from the preview.
Greg Burri [Wed, 13 Jan 2016 10:45:14 +0000 (11:45 +0100)]
* Add the analysis window.
* Other improvements.
Greg Burri [Mon, 11 Jan 2016 17:01:39 +0000 (18:01 +0100)]
* Add the possibility to set an RBC as healthy or infected
Greg Burri [Sun, 10 Jan 2016 16:29:19 +0000 (17:29 +0100)]
GUI (work in progress..)
Greg Burri [Sun, 10 Jan 2016 11:50:39 +0000 (12:50 +0100)]
GUI (work in progress..)
Greg Burri [Fri, 8 Jan 2016 20:08:00 +0000 (21:08 +0100)]
GUI (work in progress..)
Greg Burri [Fri, 8 Jan 2016 12:25:17 +0000 (13:25 +0100)]
Add the panel to display image source previews.
Greg Burri [Thu, 7 Jan 2016 17:34:45 +0000 (18:34 +0100)]
* Add area granulometry (not used for the moment)
* beginning of a little GUI.
Greg Burri [Wed, 6 Jan 2016 18:04:25 +0000 (19:04 +0100)]
* Try another approach to remove false ellipses without success (commented).
* Add otsu method in ImgTools.
Greg Burri [Tue, 5 Jan 2016 10:42:12 +0000 (11:42 +0100)]
Use float32 to reduce memory footprint.
Greg Burri [Mon, 4 Jan 2016 12:09:54 +0000 (13:09 +0100)]
Cleaning, micro-optimizations.
Greg Burri [Sun, 3 Jan 2016 21:21:55 +0000 (22:21 +0100)]
Cleaning.
Greg Burri [Sun, 3 Jan 2016 20:19:44 +0000 (21:19 +0100)]
Use float32 images instead of byte to improve the edge detection precision.
Greg Burri [Sat, 2 Jan 2016 20:42:40 +0000 (21:42 +0100)]
Use k-means instead of k-medians.
Greg Burri [Sat, 2 Jan 2016 16:20:22 +0000 (17:20 +0100)]
* Treat some special cases when ellipses intersecting.
* Some adjustments.
Greg Burri [Fri, 1 Jan 2016 09:39:27 +0000 (10:39 +0100)]
Simplification of the parasite marker.
Greg Burri [Wed, 23 Dec 2015 11:29:51 +0000 (12:29 +0100)]
Change the way the parasites are detected.
Greg Burri [Wed, 23 Dec 2015 00:00:39 +0000 (01:00 +0100)]
Little adjustments.
Greg Burri [Tue, 22 Dec 2015 11:30:21 +0000 (12:30 +0100)]
* Remove ellipses with a too high standard variation.
Greg Burri [Tue, 22 Dec 2015 00:38:04 +0000 (01:38 +0100)]
Cleaning.
Greg Burri [Mon, 21 Dec 2015 22:34:46 +0000 (23:34 +0100)]
Fix a bug in the ellipse overlap area computation.
Greg Burri [Mon, 21 Dec 2015 22:34:17 +0000 (23:34 +0100)]
Change the parasite detection method.
Greg Burri [Sun, 20 Dec 2015 13:06:49 +0000 (14:06 +0100)]
* Remove the 'DoG' filter.
* Fix the are opening.
Greg Burri [Fri, 18 Dec 2015 19:27:24 +0000 (20:27 +0100)]
Add functions to apply an area opening and to compute granulometry.
Greg Burri [Mon, 14 Dec 2015 21:06:31 +0000 (22:06 +0100)]
Fix an out-of-bound array access.
Greg Burri [Mon, 14 Dec 2015 18:13:10 +0000 (19:13 +0100)]
* Remove ellipses with too small area (improved).
* Draw all the ellipses in debug mode.
Greg Burri [Mon, 14 Dec 2015 13:44:51 +0000 (14:44 +0100)]
Improve the thinning process performance.
Greg Burri [Mon, 14 Dec 2015 01:29:31 +0000 (02:29 +0100)]
The main process is now complete.
Greg Burri [Fri, 11 Dec 2015 20:40:51 +0000 (21:40 +0100)]
Remove ellipses touching the edges.
Greg Burri [Fri, 11 Dec 2015 16:27:43 +0000 (17:27 +0100)]
beginning of the classifier.
Greg Burri [Fri, 11 Dec 2015 15:47:55 +0000 (16:47 +0100)]
Finding ellipses and parasites.
Greg Burri [Wed, 9 Dec 2015 20:02:02 +0000 (21:02 +0100)]
First commit of the f# source code.
Greg Burri [Mon, 23 Nov 2015 10:33:25 +0000 (11:33 +0100)]
Add some tests on different variations of Hough algorithm.
Greg Burri [Tue, 6 Oct 2015 06:00:11 +0000 (08:00 +0200)]
Logiciel de parasitémie lié au papier de Charles Ma.