This forum is disabled, please visit https://forum.opencv.org
 | 1 | initial version |
It seems you have to start from scratch with your project, right ?
If you want to use hough you could group your lines by distance and choose your best
Anyway before coding you have to set up the right environment like light colour and direction, background colour or pattern and some consideration like: can the labels appear in front of the tube ? which accuracy is needed... measure or just localization ?
Good settings will improve results and simplify algorithms... for example you could start trying dark or reflective background than use some threshold in HSV, fill holes than blobs
You could proceed with feasibility using some tools like Fiji or Cassandra
don't ask for full code please :) specific question about OpenCV are welcome !
