Skip to main content
17 events
when toggle format what by license comment
Feb 14, 2012 at 13:57 comment added vini Neural networks why not ! however formulating how it will be done and finding a solution would be tough
Feb 14, 2012 at 13:21 comment added MSalters I'm just thinking what this would do to strawberries. This might be very domain-specific. You could probably use a "similarity to known good pears/bananas/strawberries" measure that would be more robust, but that would deserve its own question. (Neural Network?)
Feb 10, 2012 at 6:44 vote accept vini
Feb 9, 2012 at 11:30 answer added Stocastico timeline score: 4
Feb 9, 2012 at 5:29 history tweeted twitter.com/#!/StackSignals/status/167480167168606208
Feb 9, 2012 at 3:25 comment added vini Yes its just a discoloration however it gets seperated from the rest of the image which i have already achieved
Feb 8, 2012 at 20:34 comment added Spacey @vini Hmm, would it be correct to say that (so far) a defect is defined as simply a discoloration? That seems to be the only feature here ya?
Feb 8, 2012 at 17:54 history edited vini CC BY-SA 3.0
added 218 characters in body
Feb 8, 2012 at 16:58 comment added vini i think ill need to use connected component labeling to solve my problem
Feb 8, 2012 at 16:21 comment added vini These are fruit defects like bruises, rot, scald , hail etc
Feb 8, 2012 at 15:45 comment added Jean-Yves @vini: checked, but still wondering: what is a defect?
Feb 8, 2012 at 14:24 comment added vini Check the edit :)
Feb 8, 2012 at 14:23 history edited vini CC BY-SA 3.0
added 94 characters in body
Feb 8, 2012 at 14:18 comment added Phonon Yes, original images would help. Also, what qualifies as a defect?
Feb 8, 2012 at 14:18 history edited vini CC BY-SA 3.0
added 94 characters in body
Feb 8, 2012 at 14:09 comment added Geerten You should add more detail, such as, what are the original two images, what operations have you executed on them to obtain this end result, and what is the defect you are talking about?
Feb 8, 2012 at 13:46 history asked vini CC BY-SA 3.0