知识卡片
分析与合成需要左逆
按本章约定r=Pᵀℓ,其中P是N×M,合成ℓ=Qr要求QPᵀ=Iₙ。方阵且可逆时Q=(Pᵀ)⁻¹;满行秩的过完备P可用Q=(PPᵀ)⁻¹P。复数表示需用共轭转置,不能把普通转置直接用于傅里叶等复系数变换。
这张卡片看不懂?
这张卡片已通过人工审查,不再接受失败反馈。
已有反馈 · 0 条
还没有人提交反馈。
在整书笔记中的位置
- · 图像金字塔
原书出处 · 1 处
- · 查看原书依据
Let's first look at some general properties of linear image transforms. For an input image ℓ with N pixels, a linear transform is: (23.1) where r is a vector of dimensionality M, and P is a matrix of size N × M. The columns of P = [P0, P1, …, PM–1] are the projection vectors. The vector r contains the transform coefficients: . The vector
参考来源
- 位置:《Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)》第23章《图像金字塔》(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/c023.xhtml)
- 结论依据:摘录中关于分析与合成需要左逆的定义、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:Let's first look at some general properties of linear image transforms. For an input image ℓ with N pixels, a linear transform is: (23.1) where r is a vector of dimensionality M, and P is a matrix of size N × M. The columns of P = [P0, P1, …, PM–1] are the projection vectors. The vector r contains the transform coefficients: . The vector