A Bayesian Approach to Active Self-Paced Deep Learning for SAR Automatic Target Recognition

Ren, Haohao; Yu, Xiaoqing; Bruzzone, Lorenzo; Zhang, Yukun; Zou, Lin; Wang, Xuegang

doi:10.1109/lgrs.2020.3036585

Cited by 21 publications

(5 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…As summarized in Table 1, there are 2747 images for training and 2425 images for testing. As a matter of routine, 3 , 4 , 33 the data measured at the 15-deg depression angle are used as the test set, and the data with the 17-deg depression angle are used as the training set.…”

Section: Methodsmentioning

confidence: 99%

Synthetic aperture radar target recognition via deep attention convolutional network assisted by multiscale residual despeckling network

Zou

Wang

et al. 2023

J. Appl. Rem. Sens.

Self Cite

View full text Add to dashboard Cite

Owing to the inherent characteristics of synthetic aperture radar (SAR) imaging system based on the coherent imaging modality, SAR images are inevitably corrupted by speckle noise, thereby affecting SAR target recognition accuracy. We explore the impact of speckle noise on SAR target recognition performance. Subsequently, we propose a deep attention convolutional network assisted by a multiscale residual despeckling network to improve SAR target recognition accuracy under speckle corruption. Noise information is first learned via a despeckling subnetwork consisting of dual-branch multiscale feature extraction, feature fusion, and adaptive feature channel selection. Then a classification subnetwork with a cross-dimension interaction attention mechanism is designed to realize feature extraction and identity reasoning of SAR targets. Experimental results on the benchmark moving and stationary target acquisition and recognition dataset demonstrate the effectiveness and superiority of the proposed method.

show abstract

Section: Methodsmentioning

confidence: 99%

Synthetic aperture radar target recognition via deep attention convolutional network assisted by multiscale residual despeckling network

Zou

Wang

et al. 2023

J. Appl. Rem. Sens.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Recently, there have been some studies combining SPL with AL in other fields, such as in computer vision, Lin et al [41] first proposed to combine AL with SPL for face recognition. Subsequently, Ren et al [42] applied AL and SPL with DL to synthetic aperture radar automatic target recognition. In each of these fields, the combination of SPL and AL has yielded more desirable results.…”

Section: B Self-paced Learningmentioning

confidence: 99%

Pyramidal Dilation Attention Convolutional Network With Active and Self-Paced Learning for Hyperspectral Image Classification

Hou

Chen

Peng

et al. 2023

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

In recent years, deep neural networks have been widely used for hyperspectral image (HSI) classification and have shown excellent performance using numerous labeled samples. The acquisition of HSI labels is usually based on the field investigation, which is expensive and time consuming. Hence, the available labels are usually limited, which affects the efficiency of deep HSI classification methods. To improve the classification performance while reducing the labeling cost, this article proposes a semisupervised deep learning (DL) method for HSI classification, named pyramidal dilation attention convolutional network with active and self-paced learning (PDAC-ASPL), which integrates active learning (AL), self-paced learning (SPL), and DL into a unified framework. First, a densely connected pyramidal dilation attention convolutional network is trained with a limited number of labeled samples. Then, the most informative samples from the unlabeled set are selected by AL and queried real labels, and the highest confidence samples with corresponding pseudo labels are extracted by SPL. Finally, the samples from AL and SPL are added to the training set to retrain the network. Compared with some DL-and AL-based HSI classification methods, our PDAC-ASPL achieves better performance on four HSI datasets.

show abstract

“…Также широкое применение находит предварительное преобразование с помощью метода главных компонент [136][137][138] и выделение частотных признаков с использованием вейвлет-преобразования, позволяющее достигнуть точности распознавания до 95 % и выше на примере реальных радиолокационных данных ДЗЗ [8,9,139], Достаточно часто встречаются различные гибридные архитектуры, например, использование принципов байесовской классификации на этапе нахождения оптимальных «атомов» в методе разреженных представлений [140][141][142], демонстрируя точность, превышающую 97 % на примере распознавания объектов базы MSTAR. Байесовский классификатор успешно применяется на этапе обучения свёрточных нейронных сетей, позволяя достигнуть точности распознавания целей по реальным РЛИ до 98 % и выше без значительного углубления сети [125,143,144].…”

Section: байесовский классификаторunclassified

Обработка Сигналов От Пространственно Распределённых Целей

Доросинский,

Виноградова

2023

View full text Add to dashboard Cite

Книга посвящена решению теоретических и практических проблем обнаружения, измерения параметров и классификации пространственно-распределённых целей (ПРЦ) по отражённым сигналам и формируемым на их основе их радиолокационным изображениям (РЛИ). Подобные проблемы возникают при радиолокационном обзоре воздушного и космического пространства, земной и водной поверхности с использованием различных радиотехнических систем: РЛС дальнего обнаружения, РЛС с синтезированием апертуры, размещённых на воздушных и космических носителях и многих других подобных систем. В книге подробно рассмотрены методы синтеза и анализа алгоритмов обнаружениря, оценки параметров и классификации ПРЦ, алгоритмы формирования РЛИ движущихся объектов. Книга представляет интерес для специалистов, студентов и аспирантов, работающих в области разработки современных радиотехнических систем военного и гражданского назначения.

show abstract

A Bayesian Approach to Active Self-Paced Deep Learning for SAR Automatic Target Recognition

Cited by 21 publications

References 17 publications

Synthetic aperture radar target recognition via deep attention convolutional network assisted by multiscale residual despeckling network

Synthetic aperture radar target recognition via deep attention convolutional network assisted by multiscale residual despeckling network

Pyramidal Dilation Attention Convolutional Network With Active and Self-Paced Learning for Hyperspectral Image Classification

Обработка Сигналов От Пространственно Распределённых Целей

Contact Info

Product

Resources

About