| 灰度共生矩阵纹理特征对SAR海冰漂移监测的增强性能研究 |
| |
| 引用本文： | 李小娜,张 杰,戴永寿,张 晰.灰度共生矩阵纹理特征对SAR海冰漂移监测的增强性能研究[J].海洋科学,2018,42(4):9-17. |
| |
| 作者姓名： | 李小娜 张 杰 戴永寿 张 晰 |
| |
| 作者单位： | 中国石油大学(华东)信息与控制工程学院;国家海洋局第一海洋研究所 |
| |
| 基金项目： | 国家重点研发计划项目(No.2016YFA0600102); 中央级公益性科研院所基本科研业务费专项资金资助项目(2014G31) |
| |
| 摘 要： | 海冰漂移监测对气候变化分析、船只航行、海上石油平台等海上活动安全作业具有重要意义。当前主流的SAR海冰漂移监测方法多是基于SAR灰度图开展的,其受噪声、环境等因素的影响较大,导致其在海冰漂移探测时,特征失配率高,匹配正确率低。针对这一问题,本文尝试利用SAR海冰纹理特征来增强海冰漂移探测性能。首先对比分析了8种纹理特征对海冰漂移探测中特征匹配的增强性能,筛选出能够有效增强特征匹配性能的最优纹理特征;其次进一步分析了海冰类型、入射角和分辨率对基于纹理特征的海冰漂移探测性能增强的影响。实验结果表明,均值是最优的纹理特征,与SAR强度图相比,特征匹配正确率提高了约7%。
|
| 关 键 词： | 海冰漂移 纹理特征 SAR |
| 收稿时间： | 2017/8/3 0:00:00 |
| 修稿时间： | 2017/9/15 0:00:00 |
Research on the enhanced performance of texture feature for sea ice drift monitoring based on gray level co-occurrence matrices |
| |
| LI Xiao-n,ZHANG Jie,DAI Yong-shou and ZHANG Xi.Research on the enhanced performance of texture feature for sea ice drift monitoring based on gray level co-occurrence matrices[J].Marine Sciences,2018,42(4):9-17. |
| |
| Authors: | LI Xiao-n ZHANG Jie DAI Yong-shou and ZHANG Xi |
| |
| Abstract: | Sea ice drift monitoring is important for climate change analysis, vessel navigation, and offshore maritime safety operations. The current methods of synthetic aperture radar (SAR) sea ice drift tracking are based on SAR intensity image analysis. However, intensity images are easily influenced by noise and have a high probability of mismatching and low matching accuracy for sea ice drift detection. Considering the above problems, this paper attempts to use the SAR sea ice texture features to enhance the drift detection performance. First, the enhancement performance of eight texture features for pattern matching in sea ice drift detection is analyzed, and the optimal texture feature is selected. Then the effects of sea ice type, incident angle, and resolution on the enhanced performance of texture features are analyzed. The results show that the mean value is the optimal texture feature, and compared with the SAR intensity image analysis approach, the proposed method improves the pattern matching accuracy by about 7%. |
| |
| Keywords: | sea ice drift texture feature SAR |
| 本文献已被 CNKI 等数据库收录！ |
| 点击此处可从《海洋科学》浏览原始摘要信息 |
| 点击此处可从《海洋科学》下载免费的PDF全文 |
|
