首页> 外国专利> APPROACH TO KARHUNEN-LOÈVE TRANSFORM WITH LOW ARITHMETIC COST WITH IMAGE COMPRESSION APPLICATIONS

APPROACH TO KARHUNEN-LOÈVE TRANSFORM WITH LOW ARITHMETIC COST WITH IMAGE COMPRESSION APPLICATIONS

机译:具有图像压缩应用的低算术成本的卡尔鲁宁-洛维变换方法

摘要

The karhunen-loève transform (klt) is often used in data decorrelation and dimensionality reduction. Due to its characteristic of energy compression in few components, there are important applications of klt in the context of image compression. However, its use in real applications is impossible due to the high computational cost required to calculate each input signal. In this context, the present invention proposes a low computational version for klt by applying the signal function to the klt-induced matrix elements. we call the proposed transform signaled klt (sklt). Emphasis is placed on length 8 transforms because of the use of this length of transforms in image coding, as in the jpeg standard. sklt was numerically evaluated by coding and total distance measurements for the exact klt.
机译:karhunen-loève变换(klt)通常用于数据去相关和降维。由于其能量压缩特性很少,因此klt在图像压缩方面有重要的应用。但是,由于计算每个输入信号所需的高计算成本,因此无法在实际应用中使用它。在这种情况下,本发明通过将信号函数应用于klt引起的矩阵元素而提出了klt的低计算版本。我们将提出的变换信号称为klt(sklt)。像jpeg标准一样,由于在图像编码中使用了此长度的变换,因此重点放在了长度8变换上。通过编码和总距离测量对sklt进行准确的klt数值评估。

著录项

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

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