02 Filter

Download as pdf or txt
Download as pdf or txt
You are on page 1of 73

Image Processing

COS 426, Spring 2014


Tom Funkhouser
Image Processing
Goal: read an image, process it, write the result

input.jpg output.jpg

imgpro input.jpg output.jpg –histogram_equalization


Image Processing Operations
• Luminance • Linear filtering
 Brightness  Blur & sharpen
 Contrast.  Edge detect
 Gamma  Convolution
 Histogram equalization
• Non-linear filtering
• Color  Median
 Black & white  Bilateral filter
 Saturation
 White balance
• Dithering
 Quantization
 Ordered dither
 Floyd-Steinberg
Image Processing Operations
• Luminance • Linear filtering
 Brightness  Blur & sharpen
 Contrast.  Edge detect
 Gamma  Convolution
 Histogram equalization
• Non-linear filtering
• Color  Median
 Black & white  Bilateral filter
 Saturation
 White balance
• Dithering
 Quantization
 Ordered dither
 Floyd-Steinberg
What is Luminance?
Measures perceived “gray-level” of pixel
L = 0.30*red + 0.59*green + 0.11*blue

0.5
0.0
0.7

0.9
Luminance
Measures perceived “gray-level” of pixel

Luminance
x
Values of luminance for positions
on one horizontal scanline

Figure 19.9 FvDFH


Luminance
Measures perceived “gray-level” of pixel

Luminance
x
Samples of luminance for pixels
on one horizontal scanline

Figure 19.9 FvDFH


Adjusting Brightness
• What must be done to the RGB values
to make this image brighter?

Luminance
x
Adjusting Brightness
• Method 1: Convert to HSV, scale V, convert back
• Method 2: Scale R, G, and B directly
o Multiply each of red, green, and blue by a factor
o Must clamp to [0..1] … always

Luminance
x
Original Brighter
Adjusting Contrast
• Compute one mean luminance L* for whole image
Scale deviation from L* for each pixel component

L*

Original More Contrast


Adjusting Gamma
Apply non-linear function to account for difference
between brightness and perceived brightness
of display

Perceived brightness
Iout = Iin

Amount of light

 depends on camera and monitor


Histogram Equalization
Change distribution of luminance values to
cover full range [0-1]
# Pixels

# Pixels

Luminance Luminance
Image Processing Operations
• Luminance • Linear filtering
 Brightness  Blur & sharpen
 Contrast.  Edge detect
 Gamma  Convolution
 Histogram equalization
• Non-linear filtering
• Color  Median
 Black & white  Bilateral filter
 Saturation
 White balance
• Dithering
 Quantization
 Ordered dither
 Floyd-Steinberg
Color processing
• Color models
 RGB
 CMY
 HSV
 XYZ
 La*b*
 Etc.

HSV Color Model


Black & White
Convert from color to gray-levels

Original Black & White


(actually gray levels)
Black & White
Convert from color to gray-levels

Original Black & White


(actually gray levels)

Method 1: Convert to HSV, set S=0, convert back to RGB


Method 2: Set RGB of every pixel to (L,L,L)
Adjusting Saturation
Increase/decrease color saturation of every pixel
Adjusting Saturation
Increase/decrease color saturation of every pixel

Method 1: Convert to HSV, scale S, convert back


Method 2: Set each pixel to factor * (R-L, G-L, B-L)
White Balance
Adjust colors so that a given RGB value is
mapped to a neutral color
White Balance
Conceptually:
Provide an RGB value W that should be mapped to white
Perform transformation of color space

W
White Balance
Von Kries method: adjust colors in LMS color space
 LMS primaries represent the responses of
the three different types of cones in our eyes
White Balance
For each pixel RGB:
1) Convert to XYZ color space

𝑋 0.4124 0.3576 0.1805 𝑅


𝑌 = 0.2126 0.7152 0.0722 𝐺
𝑍 0.0193 0.1192 0.9502 𝐵

2) Convert to LMS color space


𝐿 0.40024 0.7076 −0.08081 𝑋
𝑀 = −0.2263 1.16532 0.0457 𝑌
𝑆 0 0 0.91822 𝑍

3) Divide by LWMWSW
4) Convert back to RGB
Image Processing Operations
• Luminance • Linear filtering
 Brightness  Blur & sharpen
 Contrast.  Edge detect
 Gamma  Convolution
 Histogram equalization
• Non-linear filtering
• Color  Median
 Black & white  Bilateral filter
 Saturation
 White balance
• Dithering
 Quantization
 Ordered dither
 Floyd-Steinberg
Blur
What is the basic operation for each pixel
when blurring an image?
Basic Operation: Convolution
Output value is weighted sum of values in
neighborhood of input image
 Pattern of weights is the “filter” or “kernel”

Box Filter

Input Output
Basic Operation: Convolution
Output value is weighted sum of values in
neighborhood of input image
 Pattern of weights is the “filter” or “kernel”

Triangle Filter

Input Output
Basic Operation: Convolution
Output value is weighted sum of values in
neighborhood of input image
 Pattern of weights is the “filter” or “kernel”

Gaussian Filter

Input Output
Convolution with a Gaussian Filter

pixel
position
unrelated uncertain related uncertain unrelated
pixels pixels pixels pixels pixels
Convolution with a Gaussian Filter
Output value is weighted sum of values in
neighborhood of input image

0.4
0.2 0.2 Filter

Input Output
Convolution with a Gaussian Filter
Output value is weighted sum of values in
neighborhood of input image

0.4
0.2 0.2 Filter

Input Output
Convolution with a Gaussian Filter
Output value is weighted sum of values in
neighborhood of input image

0.4
0.2 0.2 Filter

Input Output
Convolution with a Gaussian Filter
Output value is weighted sum of values in
neighborhood of input image

0.4
0.2 0.2 Filter

Input Output
Convolution with a Gaussian Filter
Output value is weighted sum of values in
neighborhood of input image

0.4
0.2 0.2 Filter

Input Output
Convolution with a Gaussian Filter
What if filter extends beyond boundary?

0.4
0.2 0.2 Filter

Input Output
Convolution with a Gaussian Filter
What if filter extends beyond boundary?

0.8
0.4 Modified Filter

Input Output
Convolution with a Gaussian Filter
Output contains samples from smoothed input

Filter

Input Output
Figure 2.4 Wolberg
Linear Filtering
2D Convolution
o Each output pixel is a linear combination of input pixels
in 2D neighborhood with weights prescribed by a filter

Filter

Input Image Output Image


Linear Filtering
2D Convolution
o Each output pixel is a linear combination of input pixels
in 2D neighborhood with weights prescribed by a filter

Filter

Input Image Output Image


Linear Filtering
2D Convolution
o Each output pixel is a linear combination of input pixels
in 2D neighborhood with weights prescribed by a filter

Filter

Input Image Output Image


Linear Filtering
2D Convolution
o Each output pixel is a linear combination of input pixels
in 2D neighborhood with weights prescribed by a filter

Filter

Input Image Output Image


Linear Filtering
2D Convolution
o Each output pixel is a linear combination of input pixels
in 2D neighborhood with weights prescribed by a filter

Filter

Input Image Output Image


Gaussian Blur

per-pixel multiplication

*
input output
Gaussian Blur
Output value is weighted sum of values in
neighborhood of input image

normalized
Gaussian function
1

0
input
Gaussian blur
Linear Filtering
• Many interesting linear filters
 Blur
 Edge detect
 Sharpen
 Emboss
 etc.

Filter =
?
Blur
Convolve with a 2D Gaussian filter

Original

Blur Filter =
Edge Detection
Convolve with a 2D Laplacian filter that finds
differences between neighbor pixels

Original Detect edges

Filter =
Sharpen
Sum detected edges with original image

Original Sharpened

Filter =
Emboss
Convolve with a filter that highlights
gradients in particular directions

Original Embossed

Filter =
Side Note: Separable Filters
Some filters are separable (e.g., Gaussian)
 First, apply 1-D convolution across every row
 Then, apply 1-D convolution across every column
 Big impact on performance
Image Processing Operations
• Luminance • Linear filtering
 Brightness  Blur & sharpen
 Contrast.  Edge detect
 Gamma  Convolution
 Histogram equalization
• Non-linear filtering
• Color  Median
 Black & white  Bilateral filter
 Saturation
 White balance
• Dithering
 Quantization
 Ordered dither
 Floyd-Steinberg
Non-Linear Filtering
Each output pixel is a non-linear function of
input pixels in neighborhood (filter depends on input)

Original Paint Stained Glass


Median Filter
Each output pixel is
median of input pixels
in neighborhood
Bilateral Filter
Gaussian blur uses same filter for all pixels
Blurs across edges as much as other areas

Original Gaussian Blur


Bilateral Filter
Gaussian blur uses same filter for all pixels
Prefer a filter that preserves edges (adapts to content)

Original Bilateral Filter


Bilateral Filter
Combine Gaussian filtering in both
spatial domain and color domain

 G || p  q || G | I  I q | I q
1
Bilateral [ I ]p  p
Wp qS
s r

Spatial Color
Proximity Proximity
Weight Weight
Bilateral Filtering
Combine Gaussian filtering in both
spatial domain and color domain
r = 
r = 0.1 r = 0.25 (Gaussian blur)

input
s = 2

s = 6

s = 18
Image Processing Operations
• Luminance • Linear filtering
 Brightness  Blur & sharpen
 Contrast.  Edge detect
 Gamma  Convolution
 Histogram equalization
• Non-linear filtering
• Color  Median
 Black & white  Bilateral filter
 Saturation
 White balance
• Dithering
 Quantization
 Ordered dither
 Floyd-Steinberg
Quantization
Reduce intensity resolution
o Frame buffers have limited number of bits per pixel
o Physical devices have limited dynamic range
Uniform Quantization

P(x, y) = round( I(x, y) )


where round() chooses nearest
value that can be represented.
I(x,y)
P(x,y)

I(x,y)
P(x,y)
(2 bits per pixel)
Uniform Quantization
Images with decreasing bits per pixel:

8 bits 4 bits 2 bits 1 bit

Notice contouring.
Reducing Effects of Quantization

• Intensity resolution / spatial resolution tradeoff

• Dithering
o Random dither
o Ordered dither
o Error diffusion dither

• Halftoning
o Classical halftoning
Dithering
Distribute errors among pixels
o Exploit spatial integration in our eye
o Display greater range of perceptible intensities

Original Uniform Floyd-Steinberg


(8 bits) Quantization Dither
(1 bit) (1 bit)
Random Dither
Randomize quantization errors
o Errors appear as noise

P(x,y)
P(x,y)

I(x,y) I(x,y)

P(x, y) = round(I(x, y) + noise(x,y))


Random Dither

Original Uniform Random


(8 bits) Quantization Dither
(1 bit) (1 bit)
Ordered Dither
Pseudo-random quantization errors
o Matrix stores pattern of threshholds

i = x mod n
j = y mod n
e = I(x,y) - trunc(I(x,y))
threshold = (D(i,j)+1)/(n2+1)
if (e > threshold)
0 1/5 2/5 3/5 4/5 1
P(x,y) = ceil(I(x, y))
else
P(x,y) = floor(I(x,y)) thresholds
Ordered Dither
Bayer’s ordered dither matrices
Ordered Dither

Original Random Ordered


(8 bits) Dither Dither
(1 bit) (1 bit)
Error Diffusion Dither
Spread quantization error over neighbor pixels
o Error dispersed to pixels right and below
o Floyd-Steinberg weights:

3/16 + 5/16 + 1/16 + 7/16  1.0


Figure 14.42 from H&B
Error Diffusion Dither

Original Random Ordered Floyd-Steinberg


(8 bits) Dither Dither Dither
(1 bit) (1 bit) (1 bit)
Summary
• Color transformations
• Different color spaces useful for different operations

• Filtering
• Compute new values for image pixels based on
function of old values in neighborhood

• Dithering
• Reduce visual artifacts due to quantization
• Distribute errors among pixels
Exploit spatial integration in our eye

You might also like