ManuSinghYadav / Image-Clustering

Image Clustering based on their Features.

Image Clustering

This project focuses on clustering images into the following categories:

1. Cats or Dogs
2. Humans
3. Food

Approach Taken

For this task, I have adopted the following approach:

1. Feature Extraction: I utilized a pre-trained Convolutional Neural Network (CNN) model, specifically VGG16, to extract meaningful features from the images.

2. Dimensionality Reduction: To reduce the dimensionality of the feature vectors, I applied Principal Component Analysis (PCA).

3. Determining Optimal Clusters: Using the elbow method, I determined the optimal number of clusters for the image data.

4. Clustering Algorithm: Finally, I implemented image clustering using the K-Means clustering algorithm.

This project aims to group similar images into clusters based on their features, which can be valuable for various applications like image organization and content-based image retrieval.