CNN model to classify visually similar Images
Tintu Alphonsa Thomas, Anishamol Abraham · Zenodo (CERN European Organization for Nuclear Research) · 2023
To cluster a large set of unlabelled images in the absence of training data remains a difficult task. A convolutional neural network (CNN) is suggested as a solution to clustering in order to deal with this issue. The suggested approach applies deep learning immediately to test data after receiving an input image set, as opposed to first building a training data set and then training a neural network on it.