← 返回刚才阅读的位置

知识卡片

普通读书笔记卡 · 8916.a · 生成模型:GPT-6

希尔伯特变换移动相位

理解等级:高级

在书中约定下,Hilbert 变换的频域乘子为 −j·sgnω,正频率相移−π/2,负频率相移+π/2,幅度保持不变。正交相位关系要求这种频率相关的相移,而不是简单把所有频率乘−1;二维推广还须指定判定正负半平面的方向。

这张卡片看不懂?

已有反馈 · 0 条

    还没有人提交反馈。

    在整书笔记中的位置

    原书出处 · 1 处

    • · 查看原书依据

      For notational simplicity, we'll describe quadrature pair filters in the time domain, but they extend naturally to two or more dimensions. Consider an even symmetric zero-mean signal, ℓ(t), and with Fourier transform ℒ(ω), with ℒ(0) = 0 because the signal is zero-mean. In the Fourier domain, two functions, ℒ(ω) and ℒh(ω), are said to be Hilbert transform pairs if:

    参考来源

    - 位置:《Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)》第22章《滤波器组》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)/_epub-src/OPS/xhtml/c022.xhtml) - 结论依据:摘录中关于希尔伯特变换移动相位的定义、比较或条件支撑卡片;原书表述不严谨的范围已在覆盖记录逐项限定,不把案例结果外推为普遍结论。 - 原始内容:For notational simplicity, we'll describe quadrature pair filters in the time domain, but they extend naturally to two or more dimensions. Consider an even symmetric zero-mean signal, ℓ(t), and with Fourier transform ℒ(ω), with ℒ(0) = 0 because the signal is zero-mean. In the Fourier domain, two functions, ℒ(ω) and ℒh(ω), are said to be Hilbert transform pairs if: