首页> 外文会议>International conference on mechanical and electronic engineering >Comparative Study of Water-Body Information Extraction Methods Based on Electronic Sensing Image
【24h】

Comparative Study of Water-Body Information Extraction Methods Based on Electronic Sensing Image

机译:基于电子感测图像的水体信息提取方法对比研究

获取原文

摘要

In order to extract accurately wetland information of Tianjin Binhai New Area from electronic sensing images, the paper summarized and analyzed several water extraction methods, including Gray-level Threshold Method, Spectrum-photometric Method, Band Ratio Method and Modified Normalized Difference Water Index (MNDWI). Water information of the TM data was extracted using ENVI software by these methods, respectively. Finally, the above-mentioned results were analyzed and compared. The results show that Gray-level Threshold Method is the easiest and the least accurate, it is not appropriate for the extraction work of small water bodies. Spectrum-photometric Method is more accurate and gets clear water boundary. MNDWI is the most accurate, suitable for all kinds of water, but the boundary definition is not as good as the formers. Therefore, no one is accurate completely, but if we combine the method reasonably according to the characteristics of data, we can achieve better results.
机译:为了从电子感测图像中提取天津滨海新区天津滨海新区的准确湿地信息,本文总结和分析了几种水提取方法,包括灰度阈值法,频谱 - 光度法,带比法和改进的归一化差异水指数(MNDWI )。这些方法分别使用envi软件提取TM数据的水信息。最后,分析了上述结果并进行了比较。结果表明,灰度级阈值方法是最简单,最不准确的,这是不合适的小水体的提取工作。光谱 - 光度法更准确,透明的水边界。 MNDWI是最准确的,适合各种水,但边界定义并不像成型器一样好。因此,没有人完全准确,但如果我们根据数据的特征合理地结合该方法,我们可以实现更好的结果。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号