SAR Target Recognition With Modified Convolutional Random Vector Functional Link Network

Qijun Dai

doi:10.1109/LGRS.2021.3132020

SAR Target Recognition With Modified Convolutional Random Vector Functional Link Network

2022-01-01

4

PRE

AI

DOI：10.1109/LGRS.2021.3132020

Original

Original request for help

Save

Abstract

En 中文

Deep learning models have achieved remarkable performance in synthetic aperture radar (SAR) target recognition. However, the accuracy of these methods is sensitive to the hyper-parameters and the traditional backpropagation is time consuming. In this letter, we proposed a modified convolutional random vector functional link (IntCRVFL) network for SAR target recognition, which can simplify the SAR target recognition system. The CRVFL network consists of a convolutional neural network and an RVFL network. First, the fixed convolutional layers with randomly initialized parameters extract SAR image features and then the RVFL network performs target recognition. Especially, inspired by hyperdimensional computing, the activations of the hidden layer are obtained through a new encoding manner. Besides, only the connections between hidden and output layers need to train by a closed-form solution for the ultimately precise target recognition. The experimental results on the moving and stationary target acquisition and recognition (MSTAR) dataset demonstrate the proposed IntCRVFL network can obtain a satisfying accuracy with a faster speed.

Keywords:

Synthetic aperture radar

Target recognition

Convolutional neural networks

Feature extraction

Radar polarimetry

Computer architecture

Training

Convolutional neural network (CNN)

hyperdimensional computing (HDC)

random vector functional link (RVFL)

synthetic aperture radar (SAR)

target recognition

SAR Target Recognition With Modified Convolutional Random Vector Functional Link Network

Abstract

Journal

Organization