Hyperspectral Image Classification using Machine Learning

Singh, Anup (2018) Hyperspectral Image Classification using Machine Learning. Masters thesis, Indian Institute of Science Education and Research Kolkata.

PDF (MS dissertation of Anup Singh (13MS046)) 13MS046.pdf - Submitted Version Restricted to Repository staff only Download (1MB)

Abstract

Machine learning or ML has emerged as a new and effective tool for making predictions. The focus of advanced ML-techniques is to gain a useful insight from complex data. With the advent of new technology and optimised algorithms, it is now possible to build complex models using these techniques and test them on a particular research problem involving prediction or a classification task. Deep learning is a particular ML-technique which has been quite successful in tasks related to image classification, handwritten digit recognition and others. To handle very high dimensional data in prediction and classification tasks convolutional neural networks have shown good results. This thesis work explores the use of deep learning and other conventional machine learning methods such as Support Vector Machine(SVM) and Random Forest(RF) for hyperspectral image classification task. Hyperspectral imaging provides spectrum information along narrow spectral bands which helps to detect different land cover types and thus it can be considered as a classification task. Conventional machine learning algorithms involve cumbersome task of construction of handcrafted features. However, it becomes difficult to know what features are important to the particular classification problem at hand. In contrast, this thesis work introduces a deep learning algorithm that hierarchically learns relevant features from data. In summary, this work focuses on, A) exploration of ML & various ML techniques (Chapter 2-4) and B) possible applications of these techniques (Chapter 5-6), in the hyperspectral image classification. vii

Actions (login required)

View Item