IEEE Transactions on Geoscience and Remote Sensing· 2016Q1
Target Classification Using the Deep Convolutional Networks for SAR Images
- 1,315citations
- Q1SCImago
- 2016year
Short summary
An all-convolutional network (A-ConvNets) achieves 99% average accuracy classifying ten-class targets in SAR images, outperforming traditional ConvNets by eliminating fully connected layers to prevent overfitting on limited data.
AI-generated from the title and abstract; the full text is not read.
Key points
- A new all-convolutional network (A-ConvNets) architecture is proposed for SAR target classification.
- A-ConvNets removes fully connected layers, using only sparsely connected layers to reduce parameters and prevent overfitting.
- The A-ConvNets achieved 99% average accuracy on the MSTAR benchmark for ten-class target classification.
- A-ConvNets significantly outperforms traditional ConvNets, particularly in classifying target configuration and version variants.
AI-generated from the title and abstract; the full text is not read.
Abstract
The algorithm of synthetic aperture radar automatic target recognition (SAR-ATR) is generally composed of the extraction of a set of features that transform the raw input into a representation, followed by a trainable classifier. The feature extractor is often hand designed with domain knowledge and can significantly impact the classification accuracy. By automatically learning hierarchies of features from massive training data, deep convolutional networks (ConvNets) recently have obtained state-of-the-art results in many computer vision and speech recognition tasks. However, when ConvNets was directly applied to SAR-ATR, it yielded severe overfitting due to limited training images. To reduce the number of free parameters, we present a new all-convolutional networks (A-ConvNets), which only consists of sparsely connected layers, without fully connected layers being used. Experimental results on the Moving and Stationary Target Acquisition and Recognition (MSTAR) benchmark data set illustrate that A-ConvNets can achieve an average accuracy of 99% on classification of ten-class targets and is significantly superior to the traditional ConvNets on the classification of target configuration and version variants.
The authors' abstract, as published at the source. IEEE Transactions on Geoscience and Remote Sensing, 2016 · DOI ↗
Continue with a free account
Ask the paper: 3 free questions a day about this paper; save it, get its citation, new summaries every day for your field. Takeaways are Premium.
Continue free on the webSign in with Google or Apple; no card needed. You come back to this paper.
On your phone:
Field: Aerospace Engineering
Aerospace EngineeringEngineering