首页> 外国专利> METHOD AND SYSTEM FOR CODING IMAGES USING HUMAN VISUAL SENSITIVITY CHARACTERISTIC

METHOD AND SYSTEM FOR CODING IMAGES USING HUMAN VISUAL SENSITIVITY CHARACTERISTIC

机译:利用人类视觉敏感性特征编码图像的方法和系统

摘要

PURPOSE: A method and a system for coding images using human visual sensitivity characteristic are provided to compress a motion picture in consideration of luminance and color components to accomplish high-efficiency compression close to human visual sensitivity characteristic. CONSTITUTION: On the basis of color difference characteristic of each pixel value stored in the first color look-up table, a macro block constructing a video signal to be coded is classified into a predetermined texture classes using texture masking. The macro block is divided into four sub-blocks to obtain the average color of each sub-block, and the average colors of the sub-blocks are classified to decide the minimum value of classification values of the sub-blocks as a color class of the macro block. The classified classes are combined to assort final classes of the macro block. The quantity of generation of bits is predicted using bit models classified into quantizing intervals and dispersion values with respect to the classes of the macro block(44). A target bit by frames suitable for variable length coding is calculated(48). The actual quantity of generation of bits is controlled using the predicted quantity of generation of bits and the calculated target bit.
机译:目的:提供一种使用人类视觉敏感度特征对图像进行编码的方法和系统,以考虑亮度和颜色分量来压缩运动图像,以实现接近人类视觉敏感度特征的高效压缩。构成:基于存储在第一颜色查找表中的每个像素值的色差特性,使用纹理遮罩将构成要编码的视频信号的宏块分类为预定的纹理类别。将宏块划分为四个子块以获得每个子块的平均颜色,并且对子块的平均颜色进行分类以将子块的分类值的最小值确定为颜色类别。宏块。组合分类的类别以分类宏块的最终类别。相对于宏块的类别,使用分类为量化间隔和分散值的比特模型来预测比特的产生量(44)。计算出适合可变长度编码的目标逐帧(48)。使用预测的比特生成量和计算的目标比特来控制比特的实际生成量。

著录项

  • 公开/公告号KR100316764B1

    专利类型

  • 公开/公告日2002-02-28

    原文格式PDF

  • 申请/专利权人 SAMSUNG ELECTRONICS CO. LTD.;

    申请/专利号KR19960001097

  • 发明设计人 RYU SEONG GEOL;

    申请日1996-01-19

  • 分类号H04N7/24;

  • 国家 KR

  • 入库时间 2022-08-22 00:30:04

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