← 返回起始笔记位置

知识卡片

普通读书笔记卡 · 8915.a · 生成模型:GPT-6

复伽博核分离两种相位

理解等级:中级

复 Gabor 核 ψ=gσ(x,y)e^(j(u₀x+v₀y)),其实部为高斯加权余弦,虚部为高斯加权正弦。二者共享中心频率与包络,可共同描述局部结构的幅度和相位。频率向量指向波纹变化最快的方向,通常与条纹延伸方向垂直。

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

      Gabor functions can be extended to two dimensions by using 2D Gaussians and waves. We can also use complex waves (chapter 16) as they will give us a more general form of the Gabor function. We can multiply a complex Fourier basis function, exp (j (u0x + v0y)), by a spatially localized 2D Gaussian window, to obtain a Gabor function, ψ(x, y):

    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:摘录中关于复伽博核分离两种相位的定义、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:Gabor functions can be extended to two dimensions by using 2D Gaussians and waves. We can also use complex waves (chapter 16) as they will give us a more general form of the Gabor function. We can multiply a complex Fourier basis function, exp (j (u0x + v0y)), by a spatially localized 2D Gaussian window, to obtain a Gabor function, ψ(x, y):