search

knime-ip / knip

KNIME Image Processing Extension
https://www.knime.com/community/image-processing
50 stars 10 forks source link

Feature Nodes: Return missing value instead of NaN #320

Open gab1one opened 8 years ago

gab1one commented 8 years ago

Suggestion by a user: https://tech.knime.org/forum/knime-image-processing/feature-calculator-geometric-features-2d-size#comment-42321

Discuss: Should we implement that change? As the node is still in beta we can still change the behavior without the need for a new version of the node.

imagejan commented 8 years ago

I support this suggestion. Missing values make post-processing within KNIME easier.

In addition, it would be helpful if e.g. "Size" could be calculated in any case, since it can be determined for objects as small as 1 pixel.

avallepa commented 8 years ago

Hello, when the Feature Calculator (Beta) encounters an image with no measurable features it skip it in the result table.

is is possible to configure the node as suggested by gab1one??

" Where features cannot be calculated, e.g. "LabelRegion was too small to calculate geometric features.", we get 'missing' values instead?

Was there a resolution to the issue reported: https://github.com/knime-ip/knip/issues/320

Thanks so much!!

tibuch commented 8 years ago

There is no resolution to this issue until now. But I am currently reimplementing the Feature Calculator and will take care of this issue in the new version.

avallepa commented 8 years ago

If the above issue is not resolved, is there a workaround?

I need to have a table listing the missing measurements... :)

any suggestions? I could join a table built from the original JPG list with the measurement table but HOW can I join them so that the row corresponding to a JPG with no measurement is correctly matched?

thanks!

tibuch commented 8 years ago

Here is a workaround-workflow which replaces all NaN with Missing Values. workaround.zip

avallepa commented 8 years ago

thanks for the workaround! however I do not get NaN lines, the Feature Calculator just skip the lines were no areas are found....why?

tibuch commented 8 years ago

I can't reproduce this error or maybe my configuration is different. Could you send me a small example workflow?

avallepa commented 8 years ago

these are the nodes I use: I need to create a single excel file where the path to the raw images in each line is matched with the measurements -> but the output table is shorter 4 lines are skipped. workflow example.pdf

tibuch commented 8 years ago

What are the settings of the "Label Segment Settings"-Tab in the Feature Calculator node?

avallepa commented 8 years ago

settings are all checked. v append labels of overlapping segments v Overlapping segments do NOT need to completely overlap v append segment info

Segment label Filter Rule 1:* 0* [OR]

thanks!!!

On Wed, Jul 6, 2016 at 5:58 PM, tibuch notifications@github.com wrote:

What are the settings of the "Label Segment Settings"-Tab in the Feature Calculator node?

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/knime-ip/knip/issues/320#issuecomment-230817777, or mute the thread https://github.com/notifications/unsubscribe/ATYJ-X3rn5aS3o-Ml13GIIqiQEUmSXw8ks5qS9CbgaJpZM4HxXYM .

tibuch commented 8 years ago

Could it be that the Segment label Filter Rule 1: is filtering the 4 labels?

avallepa commented 8 years ago

do not think so, I spent most of the day trying all the options, as well as all the options related to joining or filtering and then joining tables by rows and columns. seems like I really need a 'missing' values row.. any suggestion would be most welcome! :D

On Wed, Jul 6, 2016 at 6:07 PM, tibuch notifications@github.com wrote:

Could it be that the Segment label Filter Rule 1: is filtering the 4 labels?

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/knime-ip/knip/issues/320#issuecomment-230821018, or mute the thread https://github.com/notifications/unsubscribe/ATYJ-VvCUBdWSrbSmJnL9GQZLaFow8Vfks5qS9LegaJpZM4HxXYM .

tibuch commented 8 years ago

Oh! I probably misunderstood you all the time! Do you need a left (=top) outer join? The top table is the table with all rows and the bottom table has the missing rows. Left outer join will join the bottom table to the top table and if a join-id is missing a missing-row will be inserted. leftouterjoin

avallepa commented 8 years ago

thanks, the end result should be the table as you showed it. But the first Column (or set of columns) comes from 1 table the second column (or set of columns) comes from the Feature Calculator. The sequence of the rows matters: I cannot add all the missing columns at the end because I must merge two tables with multiple columns.

The tables have initially the same row-sequence, but the feature calculator skip rows instead of adding MISSING. Hence the problem... I'm new to this Feature Calculator.... and If you have any suggestion for a step-by-step workaround it would be gratefully appreciated. thanks!!! :) Paolo

On Wed, Jul 6, 2016 at 6:32 PM, tibuch notifications@github.com wrote:

Oh! I probably misunderstood you all the time! Do you need a left (=top) outer join? The top table is the table with all rows and the bottom table has the missing rows. Left outer join will join the bottom table to the top table and if a join-id is missing a missing-row will be inserted. [image: leftouterjoin] https://cloud.githubusercontent.com/assets/5450869/16626141/eb9a76da-43a7-11e6-81d9-f4b8d3926e32.png

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/knime-ip/knip/issues/320#issuecomment-230828796, or mute the thread https://github.com/notifications/unsubscribe/ATYJ-c976k-ByfjPD88CTjJ1KBOv8o80ks5qS9iBgaJpZM4HxXYM .