News for hackers. Stuff that matters.
Re:Sorting 2 Tons of Lego, Many Questions, Results by settings on Wednesday June 28, 2017 @12:20PM (#14655448)
You may be able to normalize the labeled images by removing the background. I found some OCR examples of doing that with bluring and subtraction. The same normalization steps could be run on smartphone images taken with a consistent background.
Re:Sorting 2 Tons of Lego, Many Questions, Results by jacquesm on Wednesday June 28, 2017 @12:38PM (#14655620)
Hm. Ok, so 'white sheet of paper' would qualify I guess. Then only white parts would be hard :)