Data Cleansing For Training Vehicle Detection Models

Date

2020-04-30

Authors

Wang, Guanyu

Journal Title

Journal ISSN

Volume Title

Publisher

Abstract

In recent years, machine learning has become more and more popular. As a special type of machine learning, supervised learning uses various algorithms to learn from training data and makes predictions about future outcome. In this project, we present our study on data cleansing and its impact on vehicle detection. In the first half, we demonstrate how to collect a clean dataset for training. We also build a website which helps users to annotate and check results online. In the second half, we conduct experiments to find what factors may influence the results and how to get better test results.

Description

Keywords

Citation