计算机工程与应用 ›› 2008, Vol. 44 ›› Issue (8): 104-106.

• 学术探讨 • 上一篇    下一篇

基于小波半规则网格压缩均方误差的近似方法

胡 云1,2,顾耀林1   

  1. 1.江南大学 信息工程学院,江苏 无锡 214122
    2.无锡广播电视大学 信息工程系,江苏 无锡 214021
  • 收稿日期:2007-07-16 修回日期:2007-09-24 出版日期:2008-03-11 发布日期:2008-03-11
  • 通讯作者: 胡 云

Mean square error approximation for wavelet-based semiregular mesh compression

HU Yun1,2,GU Yao-lin1   

  1. 1.Department of Information Technology,Southern Yangtze University,Wuxi,Jiangsu 214122,China
    2.Department of Information Technology,Wuxi Radio & Television University,Wuxi,Jiangsu 214021,China
  • Received:2007-07-16 Revised:2007-09-24 Online:2008-03-11 Published:2008-03-11
  • Contact: HU Yun

摘要: 为了实现数据量大的图像在网络上的快速传输,就必须对其进行高压缩。现在一般采用比较先进的小波压缩方法,但用压缩后的数据重构得到的图像与原图像相比有明显的差别。为了降低这种失真,开始部分提出了一种有效的基于位分配模型的性能优化方法;为了实现这种快速和低复杂度的分配方法,提出了和半规则网格几何编码相关的重建均方误差的近似方法。并用实验结果说明,在基于模型分配的过程中将上述近似方法得到的结果作为失真可以改善小波编码器的性能,提高编码增益,从而达到在不降低图像压缩率下提高图像质量的目的。

Abstract: In order to achieve a large quantity of data images on the network of rapid transmission,it must carry out its high compression.Now generally uses more advanced wavelet compression method,However,the use of compressed data to reconstruct the image with the original images than the obvious differences.To reduce the distortion of this,the beginning of this article proposes an effective allocation model based on the performance optimization method;To achieve such a rapid and low complexity of the distribution,and it proposes a semiregular mesh geometry coding related to the reconstruction of the mean square error of approximation.The experimental results show that the model-based allocation process will be similar to the above method results as a small distortion can be improved wave encoder performance,improve coding gain,and thereby achieve without lowering the rate of image compression to improve the image quality.