School of Electronics Contermeasure, National University of Defense Technology, No. 460, Huangshan Road, Shushan District, Hefei 230037, China.
Sensors (Basel). 2019 Mar 5;19(5):1124. doi: 10.3390/s19051124.
Synthetic aperture radar (SAR) ship detection is a heated and challenging problem. Traditional methods are based on hand-crafted feature extraction or limited shallow-learning features representation. Recently, with the excellent ability of feature representation, deep neural networks such as faster region based convolution neural network (FRCN) have shown great performance in object detection tasks. However, several challenges limit the applications of FRCN in SAR ship detection: (1) FRCN with a fixed receptive field cannot match the scale variability of multiscale SAR ship objects, and the performance degrade when the objects are small; (2) as a two-stage detector, FRCN performs an intensive computation and leads to low-speed detection; (3) when the background is complex, the imbalance of easy and hard examples will lead to a high false detection. To tackle the above issues, we design a multilayer fusion light-head detector (MFLHD) for SAR ship detection. Instead of using a single feature map, shallow high-resolution and deep semantic feature are combined to produce region proposal. In detection subnetwork, we propose a light-head detector with large-kernel separable convolution and position sensitive pooling to improve the detection speed. In addition, we adapt focal loss to loss function and training more hard examples to reduce the false alarm. Extensive experiments on SAR ship detection dataset (SSDD) show that the proposed method achieves superior performance in SAR ship detection both in accuracy and speed.
合成孔径雷达（SAR）舰船检测是一个热门且具有挑战性的问题。传统方法基于手工特征提取或有限的浅层学习特征表示。最近，随着深度神经网络（如快速区域卷积神经网络（FRCN））出色的特征表示能力，在目标检测任务中展现出了优异的性能。然而，一些挑战限制了 FRCN 在 SAR 舰船检测中的应用：（1）具有固定感受野的 FRCN 无法匹配多尺度 SAR 舰船目标的尺度可变性，当目标较小时性能会下降；（2）作为一种两阶段检测器，FRCN 需要进行密集的计算，导致检测速度较慢；（3）当背景复杂时，简单和困难样本的不平衡会导致高误检率。为了解决上述问题，我们设计了一种用于 SAR 舰船检测的多层融合轻量级检测器（MFLHD）。它不是使用单个特征图，而是结合浅层高分辨率和深层语义特征来生成候选区域。在检测子网络中，我们提出了一种具有大核可分离卷积和位置敏感池化的轻量级检测器，以提高检测速度。此外，我们还将焦点损失应用于损失函数中，并对更多困难样本进行训练，以减少误报。在 SAR 舰船检测数据集（SSDD）上的广泛实验表明，所提出的方法在 SAR 舰船检测的准确性和速度方面都取得了卓越的性能。
