Scribd Logo
Upload

Welcome to Scribd!

  • 72 views

    2-D Fourier Transform Properties

    Uploaded by

    asif1253
    The document discusses the 2-D Fourier transform, which extends the 1-D Fourier transform to two-dimensional signals and images. The 2-D Fourier transform decomposes a 2-D signal into sinusoidal basis functions with frequencies in both the x and y directions. It has properties similar to the 1-D Fourier transform, such as linearity and rotational invariance. The discrete 2-D Fourier transform (DFT) is commonly used for digital images and has computational advantages over the continuous transform.

    Copyright:

    Attribution Non-Commercial (BY-NC)
    Download as PDF, TXT or read online from Scribd

    2-D Fourier Transform Properties

    Uploaded by

    asif1253
    72 views3 pages
    The document discusses the 2-D Fourier transform, which extends the 1-D Fourier transform to two-dimensional signals and images. The 2-D Fourier transform decomposes a 2-D signal into sinusoidal basis functions with frequencies in both the x and y directions. It has properties similar to the 1-D Fourier transform, such as linearity and rotational invariance. The discrete 2-D Fourier transform (DFT) is commonly used for digital images and has computational advantages over the continuous transform.

    Original Description:

    2-D Fourier Transform Properties

    Copyright

    © Attribution Non-Commercial (BY-NC)

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    The document discusses the 2-D Fourier transform, which extends the 1-D Fourier transform to two-dimensional signals and images. The 2-D Fourier transform decomposes a 2-D signal into sinusoidal basis functions with frequencies in both the x and y directions. It has properties similar to the 1-D Fourier transform, such as linearity and rotational invariance. The discrete 2-D Fourier transform (DFT) is commonly used for digital images and has computational advantages over the continuous transform.

    Copyright:

    Attribution Non-Commercial (BY-NC)
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    72 views3 pages

    2-D Fourier Transform Properties

    Uploaded by

    asif1253
    The document discusses the 2-D Fourier transform, which extends the 1-D Fourier transform to two-dimensional signals and images. The 2-D Fourier transform decomposes a 2-D signal into sinusoidal basis functions with frequencies in both the x and y directions. It has properties similar to the 1-D Fourier transform, such as linearity and rotational invariance. The discrete 2-D Fourier transform (DFT) is commonly used for digital images and has computational advantages over the continuous transform.

    Copyright:

    Attribution Non-Commercial (BY-NC)
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 3

    The 2-D Fourier Transform

    The 2-D Fourier Transform Introduction

    Two-Dimensional Continous Fourier Transform


    Basis functions are product of

    The 2-D Fourier Transform


    CS 450: Introduction to Digital Signal and Image Processing

    generalized sinusoids with frequency u in the x direction generalized sinusoids with frequency v in the y direction b(u , v ) = ei 2ux ei 2vy = ei 2(ux +vy )
    1 1 1 0.8 0.8 0.8

    Bryan Morse BYU Computer Science

    0.6

    0.6

    0.6

    0.4

    0.4

    0.4

    0.2

    0.2

    0.2

    0 0 0.2 0.4 0.6 0.8 1

    0 0 0.2 0.4 0.6 0.8 1

    0 0 0.2 0.4 0.6 0.8 1

    u = 4, v = 0

    u = 0, v = 5 (real parts)

    u = 4, v = 5

    The 2-D Fourier Transform Continuous Transform

    The 2-D Fourier Transform Examples

    Two-Dimensional Continous Fourier Transform


    The transform now becomes:

    Example

    F (u , v ) =

    f (x , y ) ei 2(ux +vy ) dx dy

    Similar process for the inverse:


    f (x , y ) =

    F (u , v ) ei 2(ux +vy ) du dv Image Fourier Transform (magnitude)

    The 2-D Fourier Transform Examples

    The 2-D Fourier Transform Examples

    One-Dimensional Fourier Transform: 1-D Square Pulse

    Two-Dimensional Fourier Transform: 2-D Square Pulse

    The 2-D Fourier Transform Examples

    The 2-D Fourier Transform Discrete Transform

    Two-Dimensional Fourier Transform: Another Example

    Two-Dimensional Discrete Fourier Transform


    For an M N image, these basis functions are fu ,v [x , y ] = ei 2ux /M ei 2vy /N = ei 2(ux /M +vy /N ) The Fourier Transform and Inverse: F [u , v ] = 1 MN
    M 1 N 1

    f [x , y ] ei 2(ux /M +vy /N )
    x = 0 y =0

    M 1 N 1

    f [x , y ] =
    u =0 v =0

    F [u , v ] ei 2(ux /M +vy /N )

    What is the computational complexity?


    The 2-D Fourier Transform Discrete Transform The 2-D Fourier Transform Discrete Transform

    Two-Dimensional DFT: Storage


    +u (0,0) -u

    Centering the Frequency Domain


    Sometimes useful to see the frequency domain with the origin centered:
    +u (0,0) +v -u -v

    +v
    -v

    -u

    (0,0)

    +u

    +v

    DFT storage

    Centered

    -v

    Can center either Shift storage after transform / before inverse transform Mathematically preprocess to cause the shift to happen: multiply by 1x ,y in time/spatial domain before and after

    The 2-D Fourier Transform Properties

    The 2-D Fourier Transform Properties

    Properties

    Rotation
    Rotating a 2-D function rotates its Fourier Transform.

    All other properties from one-dimensional signals apply: Linearity Shift Scaling Rayleighs Theorem

    If f2 = rot f1 F1 = F (f1 ) F2 = F (f2 ) then F2 = rot F1 i.e., the Fourier Transform is rotationally invariant.

    The 2-D Fourier Transform Properties

    The 2-D Fourier Transform Properties

    Example: Rotation

    How to Think About the 2-D Fourier Transform


    The point (u , v ) in the frequency domain corresponds to the basis function with frequency u in x and frequency v in y . OR The point (u , v ) in the frequency domain corresponds to the basis function with frequency |(u , v )| in the direction (u , v ). This follows from rotational invariance.
    Be careful: this is not the same as the magnitude and phase of the complex value F (u , v ).

    Image

    Fourier Transform (magnitude)

    The 2-D Fourier Transform Properties

    The 2-D Fourier Transform Properties

    Rotation: Example

    Transforms of Separable Functions


    A separable function has the following property: f (x , y ) = f (x ) g (y ) The Fourier Transform of a separable function is also separable: F (u ) G(v )

    The 2-D Fourier Transform Properties

    The 2-D Fourier Transform Application

    Linear Separability of The Fourier Transform


    The 2-D Fourier Transform is itself linearly separable: the Fourier Transform of a 2-D image is the transform of the rows followed by the transforms of the resulting columns (or vice versa). F [u , v ] = 1 MN 1 MN 1 MN
    M 1 N 1

    Spatial Frequencies
    Intuition: Gradual transitions require only low-frequency sinusoids Rapid transitions require high-frequency sinusoids So, Places with low spatial frequency content: smooth regions Places with high spatial frequency content: edges, texture And Low-pass ltering = smoothing High-frequency boosting = sharpening

    f [x , y ] ei 2(ux /M +vy /N )
    x =0 y = 0 M 1 N 1

    f [x , y ] ei 2ux /M ei 2vy /N )
    x =0 y = 0 N 1 M 1

    f [x , y ] ei 2ux /M ei 2vy /N )
    y =0 x =0

    What is the computational complexity of calculating it this way?

    You might also like