Synthetic aperture radar (SAR) is an earth observation technology that can obtain high-resolution image in all-weather and all-time conditions, and hence, has been widely used in civil and military applications. SAR target detection and classification are the key processes for the detailed feature information extraction of the interested target. Compared with traditional matched filtering (MF) recovered result, sparse SAR image has lower sidelobes, noise, and clutter. Thus, it will theoretically has better performance in target detection and classification. In this article, we propose a novel sparse SAR image based target detection and classification framework. This novel framework first obtains the sparse SAR image dataset by complex approximate message passing (CAMP), which is an $L_1$ -norm regularization sparse imaging method. Different from other regularization recovery algorithms, CAMP can output not only a sparse solution, but also a nonsparse estimation of considered scene that well preserves the statistical characteristic of the image when protruding the target. Then, we detect and classify the targets by using the convolutional neural network based technologies from the sparse SAR image datasets constructed by the sparse and nonsparse solutions of CAMP, respectively. For clarify, these two kinds of sparse SAR image datasets are named as $\mathcal {D}_{\rm Sp}$ and $\mathcal {D}_{\rm Nsp}$ . Experimental results show that under standard operating conditions, the proposed framework can obtain 92.60% and 99.29% mAP on Faster RCNN and YOLOv3 by using the $\mathcal {D}_{\rm Nsp}$ sparse SAR image dataset. Under extended operating conditions, the mAP value of Faster RCNN and YOLOv3 are 95.69% and 89.91% mAP, respectively. These values based on the $\mathcal {D}_{\rm Nsp}$ dataset are much higher than the classified result based on the corresponding MF dataset.

中文翻译：

---

当稀疏 SAR 图像数据集可用时基于 CNN 的目标检测和分类

合成孔径雷达（SAR）是一种能够在全天候、全天候条件下获取高分辨率图像的对地观测技术，因此在民用和军用领域得到了广泛的应用。SAR目标检测和分类是感兴趣目标详细特征信息提取的关键过程。与传统的匹配滤波(MF)恢复结果相比，稀疏SAR图像具有更低的旁瓣、噪声和杂波。因此，理论上它将在目标检测和分类方面具有更好的性能。在本文中，我们提出了一种新颖的基于稀疏 SAR 图像的目标检测和分类框架。这个新颖的框架首先通过复杂的近似消息传递（CAMP）获得稀疏的 SAR 图像数据集，这是一个$L_1$ -范数正则化稀疏成像方法。与其他正则化恢复算法不同，CAMP 不仅可以输出稀疏解，还可以输出所考虑场景的非稀疏估计，在突出目标时很好地保留了图像的统计特征。然后，我们使用基于卷积神经网络的技术从分别由 CAMP 的稀疏和非稀疏解构建的稀疏 SAR 图像数据集中对目标进行检测和分类。为澄清起见，将这两种稀疏 SAR 图像数据集命名为$\mathcal {D}_{\rm Sp}$ 和 $\mathcal {D}_{\rm Nsp}$ . 实验结果表明，在标准操作条件下，所提出的框架可以在 Faster RCNN 和 YOLOv3 上获得 92.60% 和 99.29% 的 mAP。$\mathcal {D}_{\rm Nsp}$稀疏 SAR 图像数据集。在扩展运行条件下，Faster RCNN 和 YOLOv3 的 mAP 值分别为 95.69% 和 89.91% mAP。这些值基于$\mathcal {D}_{\rm Nsp}$ 数据集远高于基于相应 MF 数据集的分类结果。