Hi Kalhara,
The Image Labeler app is indeed a good choice for labeling unlabeled data. It provides an easy way to interactively create different shapes to mark as regions of interest (ROI) labels, which is required for training algorithms like CNNs.
For medical images specifically, you can use the Medical Image Labeler app which is specifically designed to label pixels in 2D and 3D medical images. This app allows you to export labeled data as a groundTruthMedical object to train semantic segmentation algorithms.
You can refer the following examples to learn more about using the Image Labeler app:
- Medical Image Labeler: https://www.mathworks.com/help/medical-imaging/ug/get-started-with-medical-image-labeler.html
- 3D Image Labeling: https://www.mathworks.com/help/medical-imaging/ug/Label-3-d-medical-image-using-medical-image-labeler.html
- Image Labeler: https://www.mathworks.com/help/vision/ug/get-started-with-the-image-labeler.html
For using previously labeled data to label the unlabeled data, you will need to train a basic model and apply the model to the unlabeled data using the Image Labeler app. You can check out this example to automate the labeling task:
Hope this helps!
