How can I applay unsupervised images clustering

Hello,

I have data set of 1,000 images and I would like to sort them without labeling.

How can I implement it?

[KSSV]

Based on what criteria you want to sort them?

[Eden Dotan]

Based the object in the frame, more precisely,

some of the images contain cells, some tiny particles, and some empty.

Answer 1

[Vignesh Murugavel]

Hi,

As per my understanding you wish to apply a suitable unsupervised learning algorithm to classify the images/cluster the images into several groups.

You could try using an Autoencoder which can be used to get a representation of given input data which can be an image, text, or videos.

To train an autoencoder you can use trainAutoencoder.

The Autoencoder will generate meaningful features of the images, these features can be extracted from the last layer of encoder. The last layer of encoder is a lower dimensional representation of the image which can be used in k-means for clustering the features in K groups. This way the images will be classified into several groups.

You can refer the following link to know more about K-means Clustering and how to implement k-means

You must make sure that the clustering is based on some high-level similarities between the images.

Answer 2

[Image Analyst]

Try searching for "video scene classification"

https://paperswithcode.com/task/video-classification