基于Otsu与海域统计特性的SAR图像海陆分割算法
发布时间：2018-11-18 20:58
【摘要】：合成孔径雷达图像海陆分割在海面目标检测、海岸线提取等海洋应用方面具有重要的意义。针对合成孔径雷达图像的特点,本文了提出粗阈值与精确阈值相结合的海陆分割算法,并提供了一套完整的海陆分割方案。首先利用改进的Otsu法计算粗阈值,获得大致海域范围;然后根据海域统计特性近似高斯分布的特点,在分析了高斯分布统计特性的基础上,结合粗阈值分割结果,提出了计算出精确的分割阈值的方法,进而获得准确的海陆分割图像;最后通过去除杂散点进一步完善分割结果。实验结果表明,本文提出的方法具有较高的处理精度和较快的处理速度。
[Abstract]:Synthetic Aperture Radar (SAR) image segmentation plays an important role in sea surface target detection, coastline extraction and other marine applications. According to the characteristics of synthetic Aperture Radar (SAR) images, a sea and land segmentation algorithm combining rough threshold and accurate threshold is proposed, and a complete sea and land segmentation scheme is provided. First, the rough threshold is calculated by using the improved Otsu method, and the approximate sea area is obtained. Then, according to the characteristics of the statistical characteristics of the sea area approximating Gao Si's distribution, on the basis of analyzing the statistical characteristics of Gao Si's distribution, combined with the result of rough threshold segmentation, a method to calculate the accurate segmentation threshold is put forward. Then the accurate land and sea segmentation images are obtained. Finally, the segmentation results are further improved by removing stray points. The experimental results show that the proposed method has higher processing accuracy and faster processing speed.
【作者单位】： 南京电子技术研究所;
【分类号】：TN957.52
本文编号：2341205
[Abstract]:Synthetic Aperture Radar (SAR) image segmentation plays an important role in sea surface target detection, coastline extraction and other marine applications. According to the characteristics of synthetic Aperture Radar (SAR) images, a sea and land segmentation algorithm combining rough threshold and accurate threshold is proposed, and a complete sea and land segmentation scheme is provided. First, the rough threshold is calculated by using the improved Otsu method, and the approximate sea area is obtained. Then, according to the characteristics of the statistical characteristics of the sea area approximating Gao Si's distribution, on the basis of analyzing the statistical characteristics of Gao Si's distribution, combined with the result of rough threshold segmentation, a method to calculate the accurate segmentation threshold is put forward. Then the accurate land and sea segmentation images are obtained. Finally, the segmentation results are further improved by removing stray points. The experimental results show that the proposed method has higher processing accuracy and faster processing speed.
【作者单位】： 南京电子技术研究所;
【分类号】：TN957.52
【相似文献】
相关期刊论文 前1条
1 陶海东;龚军;;有线电视故障维修一例[J];中国有线电视;2006年12期
相关硕士学位论文 前1条
1 李云龙;港口区域SAR图像目标检测技术研究[D];国防科学技术大学;2011年
，本文编号：2341205
本文链接：https://www.wllwen.com/kejilunwen/wltx/2341205.html
