image enhancement in spatial domain

Upload: dev268

Post on 19-Oct-2015

43 views

Category:

Documents


0 download

DESCRIPTION

this ppt consist of the image enhancement in spatial domain.

TRANSCRIPT

  • 5/28/2018 image enhancement in spatial domain

    1/68

    Image Enhancement

    (Spatial Domain Methods)

  • 5/28/2018 image enhancement in spatial domain

    2/68

    What Is Image Enhancement?

    Image enhancement is the process of making

    images more useful

    The reasons for doing this include:

    Highlighting interesting detail in images

    Removing noise from images

    Making images more visually appealing

  • 5/28/2018 image enhancement in spatial domain

    3/68

    Image Enhancement

    Enhance otherwise hidden information Filter important image features

    Discard unimportant image features Emphasize, sharpen or smoothen image

    features

  • 5/28/2018 image enhancement in spatial domain

    4/68

    Classification of Image enhancement

    Spatial Domain

    Process intensity of pixels

    Two types- intensity transformation and spatialfiltering

    Transform Domain

    Transform image, process it and then find inversetransform to get image in spatial domain

  • 5/28/2018 image enhancement in spatial domain

    5/68

    Basic of Spatial Domain Filtering

    Origin y

    x Image f (x, y)

    (x, y)

    g (x, y) = T[ f (x, y)]

    f (x, y)is the

    input image

    g (x, y)is

    the processed image

    and Tis operator definedover some neighbourhood

    of (x, y)

  • 5/28/2018 image enhancement in spatial domain

    6/68

    Point Processing

    Point processing operations take the form

    s = T ( r )

    srefers to the processed image pixel value and rrefers to the original image pixel value

    Tis referred to as agrey level transformation function

    or apoint processing operationf(x,y) g(x,y)

  • 5/28/2018 image enhancement in spatial domain

    7/68

    Spatial Domain

    The operator T can be defined over

    The set of pixels (x,y) of the image

    The set of neighborhoods, N(x,y) ofeach pixel

  • 5/28/2018 image enhancement in spatial domain

    8/68

    Point operation

    Mask operation

    Global operation

    Classification of spatial domain

  • 5/28/2018 image enhancement in spatial domain

    9/68

    Brightness modification

    Contrast manipulation

    Histogram manipulation

    Point operation

  • 5/28/2018 image enhancement in spatial domain

    10/68

    Operation on the set of image-pixels

    6 8 2 0

    12 200 20 10

    3 4 1 0

    6 100 10 5

    Spatial Domain

    (Operator: Div. by 2)

  • 5/28/2018 image enhancement in spatial domain

    11/68

    Operation on the set of neighborhood

    6 8 2 0

    12 200 20 10

    226

    Spatial Domain

    6 8

    12 200

    (Operator: sum)

  • 5/28/2018 image enhancement in spatial domain

    12/68

    Global Operation

    6 8 2 0

    12 200 20 10

    Spatial Domain

    5 5 1 0

    2 20 3 4

    11 13 3 0

    14 220 23 14

    (Operator: sum)

  • 5/28/2018 image enhancement in spatial domain

    13/68

    Gray Level/Intensity Transformations

    Brightness modification

    Image negatives Piecewise-Linear transformationFunctions

    Log transformations Power Law transformations

    Transformations

  • 5/28/2018 image enhancement in spatial domain

    14/68

    Intensity Level Transformations

    Linear

    Negative/Identity

    Logarithmic

    Log/Inverse log

    Power law

    nth

    power/nth

    root

  • 5/28/2018 image enhancement in spatial domain

    15/68

    Suited for enhancing white or grey detail embedded indark region and black area predominates

    Image Negative

    Input gray level

    O

    utputgraylev

    el g(x,y)= 255- f(x,y)

  • 5/28/2018 image enhancement in spatial domain

    16/68

    Logarithmic Transformations

    The log transformation maps a narrow range of low

    input grey level values into a wider range of output

    values

    The inverse log transformation performs theopposite transformation

    g(x,y) = c * log (1+ f(x,y))

  • 5/28/2018 image enhancement in spatial domain

    17/68

    Log Transformations

    InvLog Log

    Input grey level values has large range of values

    T f i

  • 5/28/2018 image enhancement in spatial domain

    18/68

    Log Transformations

    Logarithm of FT reveals more details

    Range, 0 to 106becomes 0 to 6.2

    P L T f ti

  • 5/28/2018 image enhancement in spatial domain

    19/68

    Power Law Transformations

    T(f) = c*f

    f

  • 5/28/2018 image enhancement in spatial domain

    20/68

    > 1

    Compresses dark values Expands bright values

    < 1 Expands dark values Compresses bright values

    Transformations

  • 5/28/2018 image enhancement in spatial domain

    21/68

    Power Law Transformations

    s = c * r

    Map a narrow range of dark input values into awider range of output values or vice versa

    Varying gives a whole family of curves

  • 5/28/2018 image enhancement in spatial domain

    22/68

    Application of gamma correction

    A cathode ray tube (CRT) converts a video signal

    to light in a nonlinear way.

    The light intensityIis proportional to a power ()

    of the source voltage V, (I=V)

    For a computer CRT, is about 2.2

    To view image on monitors -correction is required

    Application of gamma correction

  • 5/28/2018 image enhancement in spatial domain

    23/68

    Application of gamma-correction

  • 5/28/2018 image enhancement in spatial domain

    24/68

    Power Law Example

    Magnetic Resonance

    (MR) image of a

    fractured human

    spine

  • 5/28/2018 image enhancement in spatial domain

    25/68

    Power Law Example ( = 0.6)

    = 0.6

    0

    0.1

    0.2

    0.3

    0.4

    0.5

    0.6

    0.7

    0.8

    0.9

    1

    0 0.2 0.4 0.6 0.8 1

    Old Intensities

    TransformedInte

    nsities

    l ( )

  • 5/28/2018 image enhancement in spatial domain

    26/68

    Power Law Example ( = 0.4)

    = 0.4

    0

    0.1

    0.2

    0.3

    0.4

    0.5

    0.6

    0.7

    0.8

    0.9

    1

    0 0.2 0.4 0.6 0.8 1

    Original Intensities

    TransformedIntensities

  • 5/28/2018 image enhancement in spatial domain

    27/68

    Power Law Example ( = 0.3)

    = 0.3

    0

    0.1

    0.2

    0.3

    0.4

    0.5

    0.60.7

    0.8

    0.9

    1

    0 0.2 0.4 0.6 0.8 1

    Original Intensities

    TransformedInten

    sities

    P L E l ( )

  • 5/28/2018 image enhancement in spatial domain

    28/68

    Power Law Example (cont)

    s = r 0.6

    s=

    r0.4

    Power Law Example

  • 5/28/2018 image enhancement in spatial domain

    29/68

    Power Law Example

    (Image with washed out appearance)

    An aerial view

    of a runway

  • 5/28/2018 image enhancement in spatial domain

    30/68

    Image after gamma correction (> 1)

    = 5.0

    0

    0.1

    0.2

    0.3

    0.4

    0.5

    0.60.7

    0.8

    0.9

    1

    0 0.2 0.4 0.6 0.8 1

    Original Intensities

    TransformedInten

    sities

  • 5/28/2018 image enhancement in spatial domain

    31/68

    Different

    curves

    highlight

    different

    detail

    s = r 3.0

    s=

    r4.0

    Brightness/contrast modification

  • 5/28/2018 image enhancement in spatial domain

    32/68

    Brightness/contrast modification

    g(m,n) = f(m,n) + k (increase brightness)

    g(m,n) = f(m,n)k (decrease brightness)

    Piecewise Linear Transformations

  • 5/28/2018 image enhancement in spatial domain

    33/68

    Thresholding Function

    g(x,y) = L-1, f(x,y) > t

    = 0, f(x,y) < t

    t = threshold level

    Piecewise Linear Transformations

    Input gray level

    Outputg

    raylevel

    Thresholding

  • 5/28/2018 image enhancement in spatial domain

    34/68

    Thresholding

    Thresholding transformations are particularly useful

    for segmentation in which we want to isolate an

    object of interest from a background

    s =1.0

    0.0 r threshold

    Contrast stretching

  • 5/28/2018 image enhancement in spatial domain

    35/68

    Contrast stretching

    Gray/Intensity Level Slicing

  • 5/28/2018 image enhancement in spatial domain

    36/68

    Highlight a specific range of gray values

    Two approaches:

    Display high value for range of interest, lowvalue else (discard background)

    Display high value for range of interest,original value else (preserve background)

    y y g

    Gray Level Slicing, example

  • 5/28/2018 image enhancement in spatial domain

    37/68

    y g, p

  • 5/28/2018 image enhancement in spatial domain

    38/68

    Bit Plane Slicing

    Isolate particular bits ofintensity value

    Shows contribution of

    each bit

    Higher-order bits usually

    contain most of the

    significant visual

    informationLower-order bits

    contain subtle details

    Intensity= (b7b6b5b4b3b2b1b0)

  • 5/28/2018 image enhancement in spatial domain

    39/68

    y 6 5 3 0

    BP 7

    BP 5

    BP 0

    Bit Plane Slicing (example)

  • 5/28/2018 image enhancement in spatial domain

    40/68

    Bit Plane Slicing (example)

    Intensity= (b7

    b6

    b5

    b4

    b3

    b2

    b1

    b0

    )

  • 5/28/2018 image enhancement in spatial domain

    41/68

    Bit Plane Slicing (plane 1)

    i l Sli i ( l 2)

  • 5/28/2018 image enhancement in spatial domain

    42/68

    Bit Plane Slicing (plane 2)

    Bi Pl Sli i ( l 3)

  • 5/28/2018 image enhancement in spatial domain

    43/68

    Bit Plane Slicing (plane 3)

    Bi Pl Sli i ( l 4)

  • 5/28/2018 image enhancement in spatial domain

    44/68

    Bit Plane Slicing (plane 4)

    Bit Pl Sli i ( l 5)

  • 5/28/2018 image enhancement in spatial domain

    45/68

    Bit Plane Slicing (plane 5)

    Bit Pl Sli i ( l 6)

  • 5/28/2018 image enhancement in spatial domain

    46/68

    Bit Plane Slicing (plane 6)

    Bit Pl Sli i ( l 7)

  • 5/28/2018 image enhancement in spatial domain

    47/68

    Bit Plane Slicing (plane 7)

    Bit Pl Sli i ( l 8)

  • 5/28/2018 image enhancement in spatial domain

    48/68

    Bit Plane Slicing (plane 8)

    Bit Plane Slicing (cont )

  • 5/28/2018 image enhancement in spatial domain

    49/68

    Bit Plane Slicing (cont)

    Reconstructed image

    using only bit planes 8

    and 7

    Reconstructed image

    using only bit planes 8, 7

    and 6

    Reconstructed image

    using only bit planes 7, 6

    and 5

    Histogram

  • 5/28/2018 image enhancement in spatial domain

    50/68

    gray level

    Number

    ofPixels

    0 1 2 3

    1 3 0 1

    4

    1 2

    5

    Plot of number of occurrences of grey levels against

    grey level values

    Histogram of image

  • 5/28/2018 image enhancement in spatial domain

    51/68

    Histogram Examples

  • 5/28/2018 image enhancement in spatial domain

    52/68

    g p

    Histogram Examples (cont)

  • 5/28/2018 image enhancement in spatial domain

    53/68

    g p ( )

    Histogram Examples (cont)

  • 5/28/2018 image enhancement in spatial domain

    54/68Histogram Examples (cont)

  • 5/28/2018 image enhancement in spatial domain

    55/68Histogram Examples (cont)

  • 5/28/2018 image enhancement in spatial domain

    56/68

    High contrast

    image has the mostevenly spaced histogram

    Histogram Equalization

  • 5/28/2018 image enhancement in spatial domain

    57/68

    Equal number of pixels for every gray-value Histogram is constant Preprocessing technique to enhance contrast in

    natural images

    Find gray level transformation function T to transformimage f such that the histogram of T(f) is equalized

    Spreading out the frequencies in an image (orequalising the image) improves dark or washed out

    images

    Equalisation Transformation Function

  • 5/28/2018 image enhancement in spatial domain

    58/68Histogram Equalisation

  • 5/28/2018 image enhancement in spatial domain

    59/68

    rk:input intensity

    sk:processed intensity k: the intensity range

    nj:the frequency of intensityj

    n: the sum of all frequencies

    )( kk rTs

    k

    j

    jj rp1

    )(

    k

    j

    j

    n

    n

    1

    Histogram Equalisation

  • 5/28/2018 image enhancement in spatial domain

    60/68

    Spread out gray levels to evenly distributein the range

    Find cumulative frequency distribution Normalize by dividing by total number ofpixels

    Multiply by maximum gray value

    Map gray levels

    Equalisation Transformation Functions

  • 5/28/2018 image enhancement in spatial domain

    61/68

    Histogram Equalization

  • 5/28/2018 image enhancement in spatial domain

    62/68How does it work ?

  • 5/28/2018 image enhancement in spatial domain

    63/68

    Mean value (or average gray level)

  • 5/28/2018 image enhancement in spatial domain

    64/68

    m = irip(ri)

    =1*0.3+2*0.1+3*0.2+4*0.1+5*0.2+6*0.1=2.6

    Mean value represents overall brightness

    P(r)

    0.3

    0.2

    0.1

    0.01 2 3 4 5 6 r

    Variance

  • 5/28/2018 image enhancement in spatial domain

    65/68

    gives a measure of the distribution of

    histogram values around the mean

    0.3

    0.2

    0.1

    0.0

    0.3

    0.2

    0.10.0

    V1 =3.34 V2=0.24

    v = 2= i(ri-m)2p(ri) M=2.6, v1=(1-2.6)2x0.3+

    Standard Deviation

    A l th l l i h i

  • 5/28/2018 image enhancement in spatial domain

    66/68

    A value on the gray level axis, showing average

    distance of all pixels to the mean

    0.3

    0.2

    0.1

    0.0

    0.3

    0.2

    0.1

    0.0

    D1 > D2

    = sqrt(v)

    Histograms

    V i d St d d D i ti f th

  • 5/28/2018 image enhancement in spatial domain

    67/68

    Variance and Standard Deviation of the

    histogram represent average contrast of the

    image

    The higher the Variance (=the higher the

    Standard Deviation), the higher the imagescontrast

    Histograms

    Hi t ith d St d d d i ti

  • 5/28/2018 image enhancement in spatial domain

    68/68

    Histograms with mean and Standard deviation

    M=0.73 D=0.32 M=0.71 D=0.27