Hi
user
Admin Login:
Username:
Password:
Name:
nD image segmentation using learned region agglomeration with the Ray Python library
--client
enthought
--show
scipy_2012
--room Ballroom_CDE 1206 --force
Next: 11 PythonTeX: Fast Access to Python from within LaTeX
show more...
Marks
Author(s):
Juan Nunez-Iglesias
Location
Ballroom (CDE)
Date
jul Thu 19
Days Raw Files
Start
11:45
First Raw Start
11:48
Duration
20:00
Offset
0:03:45
End
12:05
Last Raw End
12:09
Chapters
00:00
Total cuts_time
20 min.
raw-playlist
raw-mp4-playlist
encoded-files-playlist
host
archive
public
tweet
mp4
svg
png
assets
release.pdf
nD_image_segmentation_using_learned_region_agglomeration_with_the_Ray_Python_library.json
logs
Admin:
episode
episode list
cut list
raw files day
marks day
marks day
image_files
State:
---------
borked
edit
encode
push to queue
post
richard
review 1
email
review 2
make public
tweet
to-miror
conf
done
Locked:
clear this to unlock
Locked by:
user/process that locked.
Start:
initially scheduled time from master, adjusted to match reality
Duration:
length in hh:mm:ss
Name:
Video Title (shows in video search results)
Emails:
email(s) of the presenter(s)
Released:
Unknown
Yes
No
has someone authorised pubication
Normalise:
Channelcopy:
m=mono, 01=copy left to right, 10=right to left, 00=ignore.
Thumbnail:
filename.png
Description:
One of the principal goals of the Janelia Farm Research Campus is the reconstruction of complete neuronal circuits. This involves 3D electron-microscopy (EM) volumes many microns across with better than 10nm resolution, resulting in gigavoxel scale images. From these, individual neurons must be segmented out. Although image segmentation is a well-studied problem, these data present unique challenges in addition to scale: neurons have an elongated, irregular branching structure, with processes up to 50nm thin but hundreds of micrometers long); one neuron looks much like the next, with only a thin cellular boundary separating densely packed neurons; and internal neuronal structures can look similar to the cellular boundary. The first problem in particular means that small errors in segment boundary predictions can lead to large errors in neuron shape and neuronal network connectivity. Our segmentation workflow has three main steps: a voxelwise edge classification, a fine-grained segmentation into supervoxels (which can reasonably be assumed to be atomic groups of voxels), and hierarchical region agglomeration. For the first step, we use Ilastik, a pixel-level interactive learning program. Ilastik uses the output of various image filters as features to classify voxels as labeled by the user. We then use the watershed algorithm on the resulting edge probability map to obtain supervoxels. For the last step, we developed a new machine learning algorithm (Nunez-Iglesias et al, in preparation). Prior work has used the mean voxel-level edge-probability along the boundaries between regions to agglomerate them. This strategy works extremely well because boundaries get longer as agglomeration proceeds, resulting in ever-improving estimates of the mean probability. We hypothesized that we could improve agglomeration accuracy by using a classifier (which can use many more features than the mean). However, a classifier can perform poorly because throughout agglomeration we may visit a part of the feature space that has not yet been sampled. In our approach, we use active learning to ensure that we have examples from all parts of the space we are likely to encounter. We implemented our algorithm in arbitrary dimensions in an open-source, MIT-licensed Python library, Ray (https://github.com/jni/ray). Ray combines leading scientific computing Python libraries, including NumPy, SciPy, NetworkX, and scikits-learn to deliver state of the art segmentation accuracy in Python.
markdown
Comment:
http://s3.us.archive.org/nextdayvideo/enthought/scipy_2012/nD_image_segmentation _using_learned_region_agglomeration_with_the_Ray_Python_library.mp4?Signature=yQyA0Z5vMnl24AUVitkHYs6EaHY%3D&Expires=1346382467&AWSAccessKeyId=FEWGReWX3QbNk0h3 <?xml version='1.0' encoding='UTF-8'?> <ns0:entry xmlns:ns0="http://www.w3.org/2005/Atom"><ns0:category scheme="http://schemas.google.com/g/2005#kind" term="http://gdata.youtube.com/schemas/2007#video" /><ns0:category label="Education" scheme="http://gdata.youtube.com/schemas/2007/categories.cat" term="Education" /><ns0:category scheme="http://gdata.youtube.com/schemas/2007/keywords.cat" term="enthought" /><ns0:category scheme="http://gdata.youtube.com/schemas/2007/keywords.cat" term="scipy_2012" /><ns0:category scheme="http://gdata.youtube.com/schemas/2007/keywords.cat" term="General" /><ns0:category scheme="http://gdata.youtube.com/schemas/2007/keywords.cat" term="JuanNunez-Iglesias" /><ns0:id>http://gdata.youtube.com/feeds/api/videos/-cQpExBrh74</ns0:id><ns0:author><ns0:name>NextDayVideo</ns0:name><ns0:uri>https://gdata.youtube.com/feeds/api/users/NextDayVideo</ns0:uri></ns0:author><ns0:content type="text">Juan Nunez-Iglesias</ns0:content><ns0:updated>2012-08-02T22:53:20.000Z</ns0:updated><ns0:published>2012-08-02T22:53:20.000Z</ns0:published><ns1:comments xmlns:ns1="http://schemas.google.com/g/2005"><ns1:feedLink countHint="0" href="https://gdata.youtube.com/feeds/api/videos/-cQpExBrh74/comments?client=ndv" rel="http://gdata.youtube.com/schemas/2007#comments" /></ns1:comments><ns1:group xmlns:ns1="http://search.yahoo.com/mrss/"><ns1:keywords>enthought, scipy_2012, General, JuanNunez-Iglesias</ns1:keywords><ns1:description type="plain">Juan Nunez-Iglesias</ns1:description><ns1:title type="plain">nD image segmentation using learned region agglomeration with the Ray Python library</ns1:title><ns2:duration seconds="0" xmlns:ns2="http://gdata.youtube.com/schemas/2007" /><ns1:content duration="0" expression="full" isDefault="true" medium="video" type="application/x-shockwave-flash" url="https://www.youtube.com/v/-cQpExBrh74?version=3&f=user_uploads&c=ndv&d=Aarb2r5skm2_yNTuKEinXdAO88HsQjpE1a8d1GxQnGDm&app=youtube_gdata" ns2:format="5" xmlns:ns2="http://gdata.youtube.com/schemas/2007" /><ns1:thumbnail height="360" time="00:00:00" url="http://i.ytimg.com/vi/-cQpExBrh74/0.jpg" width="480" /><ns1:thumbnail height="90" time="00:00:00" url="http://i.ytimg.com/vi/-cQpExBrh74/1.jpg" width="120" /><ns1:thumbnail height="90" time="00:00:00" url="http://i.ytimg.com/vi/-cQpExBrh74/2.jpg" width="120" /><ns1:thumbnail height="90" time="00:00:00" url="http://i.ytimg.com/vi/-cQpExBrh74/3.jpg" width="120" /><ns1:category label="Education" scheme="http://gdata.youtube.com/schemas/2007/categories.cat">Education</ns1:category><ns1:category scheme="http://gdata.youtube.com/schemas/2007/developertags.cat">enthought</ns1:category><ns1:category scheme="http://gdata.youtube.com/schemas/2007/developertags.cat">scipy_2012</ns1:category><ns1:category scheme="http://gdata.youtube.com/schemas/2007/developertags.cat">General</ns1:category><ns1:category scheme="http://gdata.youtube.com/schemas/2007/developertags.cat">JuanNunez-Iglesias</ns1:category><ns1:player url="https://www.youtube.com/watch?v=-cQpExBrh74&feature=youtube_gdata_player" /></ns1:group><ns0:title type="text">nD image segmentation using learned region agglomeration with the Ray Python library</ns0:title><ns1:control xmlns:ns1="http://purl.org/atom/app#"><ns1:draft>yes</ns1:draft><ns2:state name="processing" xmlns:ns2="http://gdata.youtube.com/schemas/2007" /></ns1:control><ns0:link href="https://www.youtube.com/watch?v=-cQpExBrh74&feature=youtube_gdata" rel="alternate" type="text/html" /><ns0:link href="https://gdata.youtube.com/feeds/api/videos/-cQpExBrh74/responses?client=ndv" rel="http://gdata.youtube.com/schemas/2007#video.responses" type="application/atom+xml" /><ns0:link href="https://gdata.youtube.com/feeds/api/videos/-cQpExBrh74/ratings?client=ndv" rel="http://gdata.youtube.com/schemas/2007#video.ratings" type="application/atom+xml" /><ns0:link href="https://gdata.youtube.com/feeds/api/videos/-cQpExBrh74/complaints?client=ndv" rel="http://gdata.youtube.com/schemas/2007#video.complaints" type="application/atom+xml" /><ns0:link href="https://gdata.youtube.com/feeds/api/videos/-cQpExBrh74/related?client=ndv" rel="http://gdata.youtube.com/schemas/2007#video.related" type="application/atom+xml" /><ns0:link href="https://gdata.youtube.com/feeds/api/users/nextdayvideo/uploads/-cQpExBrh74?client=ndv" rel="self" type="application/atom+xml" /><ns0:link href="https://gdata.youtube.com/feeds/api/users/nextdayvideo/uploads/-cQpExBrh74?client=ndv" rel="edit" type="application/atom+xml" /></ns0:entry>
production notes
2012-07-19/09_48_45.dv
Apply:
11:48:45 - 11:48:54 ( 00:00:09 )
S:
11:48:45 -
E:
11:48:54
D:
00:00:09
show more...
vlc ~/Videos/veyepar/enthought/scipy_2012/dv/Ballroom_CDE/2012-07-19/09_48_45.dv :start-time=00.0 --audio-desync=0
Raw File
Cut List
11:48:45
seconds: 0.0
Wall: 11:48:45
Duration
00:00:09
11:48:54
seconds: 0.0
Wall: 11:48:45
Comments:
mp4
mp4.m3u
dv.m3u
Split:
Sequence:
:
delete
2012-07-19/09_48_55.dv
Apply:
11:48:55 - 12:09:44 ( 00:20:49 )
S:
11:48:55 -
E:
12:09:44
D:
00:20:49
show more...
vlc ~/Videos/veyepar/enthought/scipy_2012/dv/Ballroom_CDE/2012-07-19/09_48_55.dv :start-time=00.0 --audio-desync=0
Raw File
Cut List
11:48:55
seconds: 0.0
Wall: 11:48:55
Duration
00:20:49
12:09:44
seconds: 0.0
Wall: 11:48:55
Comments:
mp4
mp4.m3u
dv.m3u
Split:
Sequence:
:
delete
Rf filename:
root is .../show/dv/location/, example: 2013-03-13/13:13:30.dv
Sequence:
get this:
check and save to add this
2012-07-19/09_48_45.dv
2012-07-19/09_48_55.dv
Veyepar
Video Eyeball Processor and Review
