Study Of Skeletons

Using Various
Preprocessing Methods

By :
B. Siva Kesava
 Types of Digital Images
1. Binary Image
Black & White colors (0&1) are exists in the image.

2. Grey Scale Image

The pixel intensities ranges from some minimum to
maximum.

3. Color Image
Various colors are exists in the image
 METHODOLOGY

Various
Pre- Skeleton
Input processing Preprocessed Conversion Binary Skeletanization Of the
Grey Level Grey Level Image Binary
Image Steps
Image Image
 PREPROCESSING
Preprocessing is very useful in variety of
situations since it helps to suppress information
that is not relevant to the specific image
processing or analysis task.
Improvement of the image data that suppresses
undesired distortions or enhances some image
features important for further processing,
although geometric transformations of image are
also classified as preprocessing methods.
The preprocessed grey level image is converted
into binary by global average.
Various methods for preprocessing of a grey
level image are explained with a 5×5 grey level
image shown in figure.
10 40 60 24 34

30 60 70 65 56

40 60 70 65 56

20 25 35 40 90

50 60 40 60 40

Each preprocess method uses 3×3 mask. So the

resultant image is of 3×3 size.
Preprocessing by Local Maximum
Select the 3×3 mask from the original 5×5
image.
Calculate the maximum grey level value with
in the selected 3×3 mask.
Replace the central pixel (CP) with the
maximum value calculated in step2.

Repeat the steps 1-3 for all possible 3×3

masks in the original 5×5 image.
10 40 60 24 34
30 60 70 65 56
40 60 70 65 56 70 70 70
20 25 35 40 90 70 70 90
50 60 40 60 40 70 70 90

Original 5x5 Image Preprocessed output

Preprocessing by Local Minimum
Select the 3×3 mask from the original 5×5
image.
Calculate the minimum grey level value with in
the selected 3×3 mask.
Replace the central pixel (CP) with the
minimum value calculated in step2.

Repeat the steps 1-3 for all possible 3×3 masks

in the original 5×5 image.
10 40 60 24 34
30 60 70 65 56
40 60 70 65 56 10 24 24
20 25 35 40 90 20 25 35
50 60 40 60 40 20 25 35

Original 5x5 Image Preprocessed output

 Conversion from grey to binary
images
By using the below procedure we can get
the binary image for the corresponding grey
level images.
Calculate the global average of the grey
image values.
Compare the pixel value and global value,
If pixel value less than global value set the
pixel value to zero, other wise set to one.
 10 40 60 24 34 0 0 1 0 0

30 60 70 65 56 0 1 1 1 1

40 60 70 65 56 0 1 1 1 1

20 25 35 40 90 0 0 0 0 1

50 60 40 60 40 1 1 0 1 0

Original image binary Image

 skeletonization
The process of skeletonization involves
the following formulae.
Results:

Fig 1(a) Fig1(b) Fig 1(c)

Fig 2(a) Fig2(b) Fig 2(c)

Fig 3(a) Fig3(b) Fig 3(c)

1(a):Original Image 1(b): Binary Image 1(c): Skeleton

2(a): Local Maximum 2(b): Binary Image 2(c): Skeleton
3(a): Local Minimum 3(b): Binary Image 3(c): Skeleton
 Fig 4(a) Fig4(b) Fig 4(c)

Fig 5(a) Fig 5(b) Fig 5(c)

Fig 6(a) Fig 6(b) Fig 6(c)

Fig 7(a) Fig 7(b) Fig 7(c)

4(a): Local (Max-Min)/2 4(b): Binary Image 4(c): Skeleton
5(a): Local Mean 5(b): Binary Image 5(c): Skeleton
6(a): Local Median 6(b): Binary Image 6(c): Skeleton
7(a): Local Mode 7(b): Binary Image 7(c): Skeleton
 CONCLUSIONS
one can not say which preprocessing
method is superior; this depends upon the
type of image, the way the grey levels are
spread, and type of application.
It is better to choose one of the
preprocessing methods by applying and
comparing each instead depending upon a
constant method.