• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Journal of information and computational science >Adaptive Band Combination Dimensional Reduction Based on Point of Accumulation for Hyperspectral Image
【24h】

Adaptive Band Combination Dimensional Reduction Based on Point of Accumulation for Hyperspectral Image

机译：基于累积点的高光谱图像自适应波段组合降维

获取原文
获取原文并翻译 | 示例
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Lossy compression of hyper-spectral image based on vector quantization could get a great compression ratio, but its time complexity is very high. To solve the high complexity of vector quantization, and to reduce the compressed time, this paper puts forward a method of Adaptive Band Combination (ABC) dimensional reduction based on point of accumulation for hyper-spectral images. This method gave adequate consideration in hyper-spectral image's high spectral correlation, set a reasonable threshold then calculated the correlation coefficient between bands. If the correlation coefficient was greater than the threshold, then merged the band into the same bands group. We calculated the average value of each bands group to obtain a matrix constituted of mean value. This matrix was the data of image after reducing dimensions. Lastly, the matrix was quantified which instead of the original hyper-spectral images. This method greatly reduces the compressed time and improves the efficiency of compression. It is a good method of dimension reduction which is used for vector quantization for lossy compression of hyper-spectral image.
机译：基于矢量量化的高光谱图像有损压缩可以获得很高的压缩率，但时间复杂度很高。为了解决矢量量化的高复杂度并减少压缩时间的问题，提出了一种基于累积点的高光谱图像自适应带组合（ABC）降维方法。该方法充分考虑了高光谱图像的高光谱相关性，设定了合理的阈值，然后计算了波段之间的相关系数。如果相关系数大于阈值，则将频段合并到相同频段组中。我们计算每个频带组的平均值，以获得由平均值组成的矩阵。该矩阵是缩小尺寸后的图像数据。最后，对矩阵进行量化，而不是原始的高光谱图像。该方法大大减少了压缩时间，提高了压缩效率。这是一种很好的降维方法，用于矢量量化，以对高光谱图像进行有损压缩。

著录项

  • 来源
    《Journal of information and computational science》 |2015年第11期|4479-4486|共8页
  • 作者

    Xuejun Zhao; Kaimin Yu; Xiaojuan Wang; Xu Qiao;

  • 作者单位

    School of Mechanical Electrical and Information Engineering, China University of Mining and Technology (Beijing), Beijing 100083, China;

    School of Mechanical Electrical and Information Engineering, China University of Mining and Technology (Beijing), Beijing 100083, China;

    School of Mechanical Electrical and Information Engineering, China University of Mining and Technology (Beijing), Beijing 100083, China;

    School of Mechanical Electrical and Information Engineering, China University of Mining and Technology (Beijing), Beijing 100083, China;

  • 收录信息
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类
  • 关键词

    Compression; Vector Quantization; Adaptive Band Combination; Correlation;

    机译：压缩;向量量化;自适应频段组合相关性;

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