Next Article in Journal
Ground-Based Remote Sensing of Volcanic CO2 Fluxes at Solfatara (Italy)—Direct Versus Inverse Bayesian Retrieval
Next Article in Special Issue
Total Variation Regularization Term-Based Low-Rank and Sparse Matrix Representation Model for Infrared Moving Target Tracking
Previous Article in Journal
Geospatial Object Detection in High Resolution Satellite Images Based on Multi-Scale Convolutional Neural Network
Previous Article in Special Issue
SparkCloud: A Cloud-Based Elastic Bushfire Simulation Service
Article Menu
Issue 1 (January) cover image

Export Article

Open AccessArticle
Remote Sens. 2018, 10(1), 124; https://doi.org/10.3390/rs10010124

A CNN-Based Method of Vehicle Detection from Aerial Images Using Hard Example Mining

Center for Spatial Information Science (CSIS), University of Tokyo, 5-1-5 Kashiwanoha, Kashiwa-shi, Chiba 2778568, Japan
*
Author to whom correspondence should be addressed.
Received: 31 October 2017 / Revised: 29 December 2017 / Accepted: 16 January 2018 / Published: 18 January 2018
Full-Text   |   PDF [8116 KB, uploaded 18 January 2018]   |  

Abstract

Recently, deep learning techniques have had a practical role in vehicle detection. While much effort has been spent on applying deep learning to vehicle detection, the effective use of training data has not been thoroughly studied, although it has great potential for improving training results, especially in cases where the training data are sparse. In this paper, we proposed using hard example mining (HEM) in the training process of a convolutional neural network (CNN) for vehicle detection in aerial images. We applied HEM to stochastic gradient descent (SGD) to choose the most informative training data by calculating the loss values in each batch and employing the examples with the largest losses. We picked 100 out of both 500 and 1000 examples for training in one iteration, and we tested different ratios of positive to negative examples in the training data to evaluate how the balance of positive and negative examples would affect the performance. In any case, our method always outperformed the plain SGD. The experimental results for images from New York showed improved performance over a CNN trained in plain SGD where the F1 score of our method was 0.02 higher. View Full-Text
Keywords: vehicle detection; hard example mining; high-resolution; aerial image; satellite image; convolutional neural network (CNN) vehicle detection; hard example mining; high-resolution; aerial image; satellite image; convolutional neural network (CNN)
Figures

Graphical abstract

This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).
SciFeed

Share & Cite This Article

MDPI and ACS Style

Koga, Y.; Miyazaki, H.; Shibasaki, R. A CNN-Based Method of Vehicle Detection from Aerial Images Using Hard Example Mining. Remote Sens. 2018, 10, 124.

Show more citation formats Show less citations formats

Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.

Related Articles

Article Metrics

Article Access Statistics

1

Comments

[Return to top]
Remote Sens. EISSN 2072-4292 Published by MDPI AG, Basel, Switzerland RSS E-Mail Table of Contents Alert
Back to Top