Image classification is one of the classical image processing problems. There are various approaches such as Support Vector Machine, Artificial Neural Networks, Convolutional Neural Networks, K-Nearest Neighbors and Decision Tree for solving this problem. In this project, Support Vector Machine (SVM) is used to classify Images and we are trying to understand SVM and then understand how to draw a decision boundary and try to make it optimal and use it for classification. It’s a supervised learning algorithm that is mainly used to classify data into different classes. The main advantage of SVM is that it can be used for both classification and regression problems. SVM draws a decision boundary which is a hyperplane between any two classes in order to separate them or classify them.
The main objectives of developing this project were:
- To develop a machine learning model which classifies images using a support vector machine algorithm.
- To predict the categories of the input image using the features of the model.
- To analyze feature selection methods and understand their working principle.
- To find a maximum marginal hyperplane(MMH) that best divides the dataset into classes.
The dataset which we took is publicly available on the Kaggle Website. The dataset contains panda and bear images generated by DALL·E Mini, an AI model that draws images from any prompt. The task for this dataset is binary classification of images. It provides numerous photos of pandas and bears which are different from each other. The dataset consists of 100 images of each animal. The dataset is in .jpeg (Joint Photographic Experts Group) format which is further prepared to data frame as supported by pandas library in python.
kaggle dataset : https://www.kaggle.com/datasets/mattop/panda-or-bear-image-classification.