Home // DATA ANALYTICS 2020, The Ninth International Conference on Data Analytics // View article

Online Feature Selection for Semantic Image Segmentation

Authors:
Rishav Rajendra
Chris J. Michael
Elias Ioup
Md Tamjidul Hoque
Mahdi Abdelguerfi

Keywords: machine learning; semantic segmentation; streaming images; feature selection

Abstract:
In this project, we classify each pixel from the incoming stream of aerial imagery of water bodies as either “land” or “water” in real-time. Traditional batch feature processing techniques can be too slow to adapt to real-time changes. This paper proposes an online distributed framework for Semantic Segmentation using conditional independence to discard irrelevant and redundant features to train a fast and lightweight but accurate machine learning model. Through extensive experimental results using aerial imagery of water bodies, we demonstrate that our approach is faster than existing online feature selection methods while maintaining high accuracy.

Pages: 41 to 46

Copyright: Copyright (c) IARIA, 2020

Publication date: October 25, 2020

Published in: conference

ISSN: 2308-4464

ISBN: 978-1-61208-816-7

Location: Nice, France

Dates: from October 25, 2020 to October 29, 2020