Ship Detection Based on YOLOv2 for SAR Imagery

Abstract

Synthetic aperture radar (SAR) imagery has been used as a promising data source for monitoring maritime activities, and its application for oil and ship detection has been the focus of many previous research studies. Many object detection methods ranging from traditional to deep learning approaches have been proposed. However, majority of them are computationally intensive and have accuracy problems. The huge volume of the remote sensing data also brings a challenge for real time object detection. To mitigate this problem a high performance computing (HPC) method has been proposed to accelerate SAR imagery analysis, utilizing the GPU based computing methods. In this paper, we propose an enhanced GPU based deep learning method to detect ship from the SAR images. The You Only Look Once version 2 (YOLOv2) deep learning framework is proposed to model the architecture and training the model. YOLOv2 is a state-of-the-art real-time object detection system, which outperforms Faster Region-Based Convolutional Network (Faster R-CNN) and Single Shot Multibox Detector (SSD) methods. Additionally, in order to reduce computational time with relatively competitive detection accuracy, we develop a new architecture with less number of layers called YOLOv2-reduced. In the experiment, we use two types of datasets: A SAR ship detection dataset (SSDD) dataset and a Diversified SAR Ship Detection Dataset (DSSDD). These two datasets were used for training and testing purposes. YOLOv2 test results showed an increase in accuracy of ship detection as well as a noticeable reduction in computational time compared to Faster R-CNN. From the experimental results, the proposed YOLOv2 architecture achieves an accuracy of 90.05% and 89.13% on the SSDD and DSSDD datasets respectively. The proposed YOLOv2-reduced architecture has a similarly competent detection performance as YOLOv2, but with less computational time on a NVIDIA TITAN X GPU. The experimental results shows that the deep learning can make a big leap forward in improving the performance of SAR image ship detection.

Depthwise Separable Convolution Neural Network for High-Speed SAR Ship Detection

“…SAR images in this dataset possess different satellite sensors, various polarization modes, multiple resolutions, different scenes, and abundant ship sizes, so it can verify the robustness of methods. Therefore, many scholars [10,21,35,[55][56][57][58][59][60][61][62][63] conducted research based on it for a better comparison.…”

Deep Learning Based Fossil-Fuel Power Plant Monitoring in High Resolution Remote Sensing Images: A Comparative Study

“…Unsupervised-restricted CNN [35] was modified from DSSD for detecting different kinds of targets from the data by Geoeye and Quickbird sensors. In [36], YOLO is used for ship detection in synthetic aperture radar (SAR) data. RetinaNet is used to automatically detect ship in multi-resolution Gaofen-3 imagery [37].…”

Enhanced Feature Extraction for Ship Detection from Multi-Resolution and Multi-Scene Synthetic Aperture Radar (SAR) Images

“…The multi-step operation mode of the traditional methods leads to time-consuming and low robustness of detection. Deep learning has also been applied to SAR ship detection [19][20][21][22][23][24][25].…”

“…In terms of one-stage detectors, Wang et al [20] apply the end-to-end RetinaNet to SAR ship detection, construct a multi-resolution and complex background dataset and achieve high detection accuracy. In order to reduce computational time with relatively competitive detection accuracy, Chang et al [21] develop a new architecture with less number of layers called YOLOv2-reduced. With respect to two-stage detectors, Hu et al [22] use Faster-RCNN to detect SAR ships under the multi-resolution condition.…”