Spatial Filtering

CS474/674 - Prof. Bebis

Sections 3.4, 3.5, 3.6, 3.7, 3.8
Spatial Filtering Methods

output image
 Spatial Filtering (cont’d)
• Need to define:
(1) a neighborhood (or mask)
(2) an operation

output image
 Spatial Filtering – Neighborhood
• Defines which pixels will be processed.
• Usually, it has a square shape K x K.
– e.g., 3x3 or 5x5
 Spatial filtering - Operation
• Defines how to manipulate the pixel values within
the specified neighborhood.

Example: z’5 = 5z1 -3z2+z3-z4-2z5-3z6+z8-z9-9z7

output image
 Linear vs Non-Linear filters
• A filter is called linear/non-linear when the operator being used is
linear/non-linear.

z’5 = 5z1 -3z2+z3-z4-2z5-3z6+z8-z9-9z7 linear

Examples:
z’5 = max(z1,z2,z3,z4,z5,z6,z7,z8,z9) non-linear

output image
 Linear Operators

• Two common linear operators are:

– Correlation
– Convolution

• The output is a linear combination of the inputs.

 Correlation (linear operator)
• The output of correlation is a weighted sum of input pixels.

Need to define the

weights in the
KxK
output image neighborhood (i.e.,
mask weights)
 Correlation (cont’d)
w(i,j)
g(i,j)
Filtered
Image

f(i,j)

Filtered image is
generated by moving
the center of the mask
at every location.

K /2 K /2
g (i, j )  w(i, j )  f (i, j )   
s  K /2 t  K /2
w( s, t ) f (i  s, j  t )
Handling Locations Close to Boundaries

Usually, we pad with zeroes

0 0 0 ……………………….0

0 0 0 ……………………….0
 Correlation (cont’d)

Often used in applications where

we need to measure the similarity
between images or parts of images
(e.g., template matching).

Simple template matching

does not work in most
practical cases.
 Convolution (linear operator)

• Similar to correlation except that the mask is first flipped both

horizontally and vertically.

K /2 K /2
g (i, j )  w(i, j )  f (i, j )   
s  K /2 t  K /2
w( s, t ) f (i  s, j  t )

• Note: if w(i, j) is symmetric (i.e., w(i, j)=w(-i,-j)), then

convolution is equivalent to correlation!
 Example

Correlation:

Convolution:
 Filter Categories

• We will focus on two types of filters:

– Smoothing (low-pass) filters
– Sharpening (high-pass) filters
 Smoothing Filters (low-pass)

• Useful for reducing noise and eliminating small details.

– The elements of the mask must be positive.
– Mask elements sum to 1 (assuming normalized weights).
Smoothing filters – Example

input image smoothed image

 Sharpening Filters (high-pass)

• Useful for highlighting fine details.

– The elements of the mask contain both positive and
negative weights.
– Mask elements sum to 0.
 Sharpening Filters - Example
• Useful for highlighting fine details.
– e.g., emphasize edges
 Sharpening Filters - Example

input image sharpened image

(for better visualization, the original

image has been added to the
sharpened image)
 Common Smoothing Filters

• Averaging

• Gaussian

• Median filtering (non-linear)

Smoothing Filters: Averaging
Smoothing Filters: Averaging (cont’d)
• Mask size determines degree of smoothing (i.e., loss of detail).

original 3x3 5x5 7x7

15x15 25x25
Smoothing Filters: Averaging (cont’d)
Example: extract largest, brightest objects

15 x 15 averaging image thresholding

 Smoothing filters: Gaussian
• The mask weights are samples of a 2D Gaussian function:
 Smoothing filters: Gaussian (cont’d)
• Mask size depends on σ, e.g.:
 Smoothing filters: Gaussian (cont’d)
• Since mask size depends on σ,
it controls the amount of smoothing!
σ=3
σ = 1.4
Smoothing filters: Gaussian (cont’d)
Averaging vs Gaussian Smoothing
Averaging

Gaussian
 Smoothing Filters: Median Filtering
(non-linear)
• Very effective for removing “salt and pepper” noise (i.e.,
random occurrences of black and white pixels).

median
averaging filtering
 Smoothing Filters: Median Filtering (cont’d)
• Replace each pixel by the median in a neighborhood
around the pixel.
• The size of the neighborhood controls the amount of
smoothing.
 Common Sharpening Filters

• Unsharp masking
• High Boost filter
• Gradient (1st derivative)
• Laplacian (2nd derivative)
 Sharpening Filters: Unsharp Masking
• Obtain a sharp image by subtracting a lowpass filtered
(i.e., smoothed) image from the original image:

- =
(with some contrast
enhancement)
 Sharpening Filters: High Boost
• Image sharpening emphasizes edges but details are lost.
• Idea: amplify input image, then subtract a lowpass image.

Note: textbook’s
formulation is
slightly different but
in essence equivalent.

(A-1) + =
 Sharpening Filters: High Boost (cont’d)
• If A=1, the result is unsharp masking.
• If A>1, part of the original image is added back to the high
pass filtered image.

High boost

One way to
implement high boost
filtering is using these
masks:
 Sharpening Filters: High Boost (cont’d)

A=1.4 A=1.9
 Sharpening Filters: Derivatives
• The derivative of an image results in a sharpened image.

• Image derivatives can be computed using the gradient:

 Gradient

• The gradient is a vector which has magnitude and direction:

(approximation)
f f
or | || |
x y
 Gradient (cont’d)

• Gradient magnitude: provides information about edge

strength.
• Gradient direction: perpendicular to the direction of the
edge.
 Gradient Computation

• Approximate partial derivatives using finite differences:

Δx

y
Gradient Computation (cont’d)

f(x3,y3)-f(x3,y2)
sensitive to horizontal edges
y3-y2

y2=y3+Dy, y3=y, x3=x, Dy=1

sensitive to vertical edges

 Example: visualize partial derivatives
• The gradient can be visualized by mapping the values to [0, 255]

f
x

f
y
 Implement Gradient Using Masks

• We can implement and using masks:

(x+1/2,y)
good approximation
at (x+1/2,y) (x,y+1/2)
*
*
good approximation
at (x,y+1/2)
 Implement Gradient Using Masks (cont’d)

• A different approximation of the gradient:

good approximation
(x+1/2,y+1/2)

• We can implement and using the following masks:

 Implement Gradient Using Masks (cont’d)
• Other approximations

Prewitt

Sobel
 Example: Gradient Magnitude Image
• The gradient magnitude can be visualized by mapping the
values to [0, 255]
f Gradient Magnitude

x

f
y
(isotropic, i.e.,
edges in all directions)
 Second Derivative
f ( x)  f ( x  1)  f ( x)

f ( x)  f ( x  1)  f ( x)  f ( x  1)  f ( x  1)  2 f ( x)
 Second Derivative (cont’d)
• Often, points that lie on an edge
can be detected by:

(1) Local maxima or

minima of the first derivative.

(2) Zero-crossings 1st derivative

of the second derivative (i.e., points
where the second derivative changes
2nd derivative
sign).
 Second Derivative (cont’d)

Example:
 Laplacian
The Laplacian is defined as:

(dot product)

Approximate
2nd derivatives:
 Laplacian (cont’d)
Laplacian Mask

input image output image

5 5 5 Edges can
5 -10 -5 -5
be found
5 -10
by detecting
10
the zero-
-10
crossings
 Example: Laplacian vs Gradient
Laplacian Sobel

• Laplacian localizes edges better (zero-crossings).

• Higher order derivatives are typically more sensitive to noise.
• Laplacian is less computational expensive (i.e., one mask).
• Laplacian can provide edge magnitude information
but no information about edge direction.
Example
Example
Example
 Quiz #3
• When: Monday (Oct 12th) at 2pm
• What: Spatial Transformations
• Duration
– 10 minutes for answering questions
– 5 minutes for uploading your answers on Canvas
• Suggested Apps
– Genius Scan
[Link]
– Canvas Student
[Link]