会员体验
专利管家(专利管理)
工作空间(专利管理)
风险监控(情报监控)
数据分析(专利分析)
侵权分析(诉讼无效)
联系我们
交流群
官方交流:
QQ群: 891211   
微信请扫码    >>>
现在联系顾问~
热词
    • 2. 发明申请
    • MOVING IMAGE CODING APPARATUS, MOVING IMAGE DECODING APPARATUS, CONTROL METHOD THEREFOR, COMPUTER PROGRAM, AND COMPUTER-READABLE STORAGE MEDIUM
    • 移动图像编码装置,移动图像解码装置,其控制方法,计算机程序和计算机可读存储介质
    • WO2005088977A1
    • 2005-09-22
    • PCT/JP2005/004851
    • 2005-03-11
    • CANON KABUSHIKI KAISHAKAJIWARA, HiroshiMAEDA, MitsuruSUZUKI, MasakiKISHI, Hiroki
    • KAJIWARA, HiroshiMAEDA, MitsuruSUZUKI, MasakiKISHI, Hiroki
    • H04N7/30
    • H04N19/64H04N19/169H04N19/17H04N19/184H04N19/46H04N19/61H04N19/63
    • This invention provides a technique of preventing errors from being gradually accumulated on the decoding side, while maintaining high scalability, even when a technique for transformation to subbands as a plurality of frequency components, like discrete wavelet transformation, is used for moving image coding. The image data of one input frame is decomposed into a plurality of subbands having different frequency components by a discrete wavelet transformation unit (101). A lower bitplane coding unit (109) codes, for each bitplane, predetermined lower bits of each coefficient data of a subband LL and the coefficient data of subbands other than the subband LL. The data of the upper bits of the subband LL is stored in a frame memory (105). A motion vector detection unit (107) detects a predicted value and motion vector on the basis of the decoded data of the subband LL in a preceding frame. A subtracter (103) obtains the difference between the detected predicted value and the current frame. A coefficient coding unit (104) codes the obtained difference. A motion vector coding unit (108) codes the motion vector. A multiplexing unit (111) multiplexes the code data obtained by the coding units (104, 108, 109).
    • 本发明提供了一种防止错误逐渐积累在解码侧的技术,同时保持高的可扩展性,即使将用于变换为多个频率分量的子带(如离散小波变换)的变换技术用于运动图像编码也是如此。 通过离散小波变换单元(101)将一个输入帧的图像数据分解成具有不同频率分量的多个子带。 低位平面编码单元(109)针对每个位平面编码子带LL的每个系数数据的预定低位和子带LL以外的子带的系数数据。 子带LL的高位的数据被存储在帧存储器(105)中。 运动矢量检测单元(107)根据前一帧中的子带LL的解码数据来检测预测值和运动矢量。 减法器(103)获得检测到的预测值与当前帧之间的差。 系数编码单元(104)对获得的差值进行编码。 运动矢量编码单元(108)对运动矢量进行编码。 复用单元(111)对由编码单元(104,108,109)获得的编码数据进行复用。