Picture of boy being examining by doctor at a tuberculosis sanatorium

Understanding our future through Open Access research about our past...

Strathprints makes available scholarly Open Access content by researchers in the Centre for the Social History of Health & Healthcare (CSHHH), based within the School of Humanities, and considered Scotland's leading centre for the history of health and medicine.

Research at CSHHH explores the modern world since 1800 in locations as diverse as the UK, Asia, Africa, North America, and Europe. Areas of specialism include contraception and sexuality; family health and medical services; occupational health and medicine; disability; the history of psychiatry; conflict and warfare; and, drugs, pharmaceuticals and intoxicants.

Explore the Open Access research of the Centre for the Social History of Health and Healthcare. Or explore all of Strathclyde's Open Access research...

Image: Heart of England NHS Foundation Trust. Wellcome Collection - CC-BY.

A deep convolutional generative adversarial networks (DCGANs)-based semi-supervised method for object recognition in synthetic aperture radar (SAR) images

Gao, Fei and Yang, Yue and Wang, Jun and Sun, Jinping and Yang, Erfu and Zhou, Huiyu (2018) A deep convolutional generative adversarial networks (DCGANs)-based semi-supervised method for object recognition in synthetic aperture radar (SAR) images. Remote Sensing, 10 (6). ISSN 2072-4292

Text (Gao-etal-RS-2018-A-deep-convolutional-generative-adversarial-networks-based-semi-supervised-method-for-object-recognition)
Final Published Version
License: Creative Commons Attribution 4.0 logo

Download (3MB) | Preview


Synthetic aperture radar automatic target recognition (SAR-ATR) has made great progress in recent years. Most of the established recognition methods are supervised, which have strong dependence on image labels. However, obtaining the labels of radar images is expensive and time-consuming. In this paper, we present a semi-supervised learning method that is based on the standard deep convolutional generative adversarial networks (DCGANs). We double the discriminator that is used in DCGANs and utilize the two discriminators for joint training. In this process, we introduce a noisy data learning theory to reduce the negative impact of the incorrectly labeled samples on the performance of the networks. We replace the last layer of the classic discriminators with the standard softmax function to output a vector of class probabilities so that we can recognize multiple objects. We subsequently modify the loss function in order to adapt to the revised network structure. In our model, the two discriminators share the same generator, and we take the average value of them when computing the loss function of the generator, which can improve the training stability of DCGANs to some extent. We also utilize images of higher quality from the generated images for training in order to improve the performance of the networks. Our method has achieved state-of-the-art results on the Moving and Stationary Target Acquisition and Recognition (MSTAR) dataset, and we have proved that using the generated images to train the networks can improve the recognition accuracy with a small number of labeled samples.