image labeler_Label Pixels Image

Hello ,I would happy if you can help me,

I Labeled Pixels ImageLabeled Pixels Image and i want to train on them a rcnn net.

But importing the marked images is not a table so I can't continue the trainning. What is the way to train a network on Labeled Pixels Image?

Answer 1

[Sourav Bairagya]

You can use "object DetectorTrainingData" function to create table form the labelled ground truth objects which are returned from labeler and then you can train the object detector using that.

For more information you can leverage these links:

https://www.mathworks.com/help/vision/ref/objectdetectortrainingdata.html

https://www.mathworks.com/help/vision/ug/labeler-pixel-label-data-storage.html

https://www.mathworks.com/help/vision/ref/trainrcnnobjectdetector.html#bvcpk41-1-groundTruth

Answer 2

[tamar wase]

Thank you!

But I tried to do it on Labeled Pixels Image, it didn't work because it's not a table.

There is an option to convert the Labeled images so we can train on them.