Cited 2 time in
Cited 14 time in
Micro-Doppler Based Classification of Human Aquatic Activities via Transfer Learning of Convolutional Neural Networks
- Micro-Doppler Based Classification of Human Aquatic Activities via Transfer Learning of Convolutional Neural Networks
- Park, Jinhee; Javier, Rios Jesus; Moon, Taesup; Kim, Youngwook
- DGIST Authors
- Moon, Taesup
- Issue Date
- Sensors, 16(12)
- Article Type
- Activity Classifications; Aquatic Activity Classification; Classification Accuracy; Classification of Activity; Convolution; Convolutional Neural Network; Convolutional Neural Networks; Micro-Doppler; Micro-Doppler Signatures; Neural Networks; Radar; Radar Cross Section; Real Measurements; Simulation Studies; Transfer Learning
- Accurate classification of human aquatic activities using radar has a variety of potential applications such as rescue operations and border patrols. Nevertheless, the classification of activities on water using radar has not been extensively studied, unlike the case on dry ground, due to its unique challenge. Namely, not only is the radar cross section of a human on water small, but the micro-Doppler signatures are much noisier due to water drops and waves. In this paper, we first investigate whether discriminative signatures could be obtained for activities on water through a simulation study. Then, we show how we can effectively achieve high classification accuracy by applying deep convolutional neural networks (DCNN) directly to the spectrogram of real measurement data. From the five-fold cross-validation on our dataset, which consists of five aquatic activities, we report that the conventional feature-based scheme only achieves an accuracy of 45.1%. In contrast, the DCNN trained using only the collected data attains 66.7%, and the transfer learned DCNN, which takes a DCNN pre-trained on a RGB image dataset and fine-tunes the parameters using the collected data, achieves a much higher 80.3%, which is a significant performance boost. © 2016 by the authors; licensee MDPI, Basel, Switzerland.
- MDPI AG
- Department of Information and Communication EngineeringMachine Learning and Data Science Lab1. Journal Articles
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.