CIML MINI PROJECT
{ECE3020}
TOPIC- Image
Segmentation
Rohan S
20201ECE0197
5ECE3
Table of Contents
Description
Data set information
Attribute information
Steps to finish Project
Accuracy without PCA
Accuracy with PCA
Accuracy with SVM (PCA ON)
Accuracy with SWM(PCA OFF)
Conclusion
DESCRIPTION
Image segmentation involves converting an image into
a collection of regions of pixels that are represented
by a labeled image.
The goal of segmentation is to simplify and change the
representation of an image into something that is more
meaningful and easier to analyze.
Image segmentation is used in traffic control system,
video surveillance , Face recognition, Fingerprint
recognition etc.
DATA SET INFORMATION
The instances were drawn randomly from a database of
7 outdoor images.
The images were hand segmented to create a
classification for every pixel.
Each instance is a 3x3 region.
The 7 outdoor images are - Brick Face, Sky, Foliage,
Cement, Window, Path, Grass .
The values of REGIONCENTROIDROW,INTENSITYMEAN
were Given to Analize the Data
DATA Attribute Information
1. region-centroid-col: the column of the center pixel of the
region.
2. region-centroid-row: the row of the center pixel of the
region.
3. region-pixel-count: the number of pixels in a region = 9.
4. short-line-density-5: the results of a line extraction
algorithm that counts how many lines of length 5 (any
orientation) with low contrast, less than or equal to 5, go
through the region.
5. short-line-density-2: same as short-line-density-5 but counts
lines of high contrast, greater than 5.
6. saturation-mean
7. Hue-mean
8. intensity-mean: the average over the region of (R + G + B)/3
9. Raw red -mean: the average over the region of the R value.
10. Raw blue -mean: the average over the region of the B value.
11. Raw green -mean: the average over the region of the G value.
12. Ex red-mean: measure the excess red: (2R - (G + B))
13. Ex blue -mean: measure the excess blue: (2B - (G + R))
14. Ex green -mean: measure the excess green: (2G - (R + B))
Steps to complete project
Download the
data given to you Importing Data
from the link set to Matlab .
(Data set folder).
Obtaining scatter
Analyzing
plot of original
the Data.
Data set.
Applying SVM(PC
Applying PCA and
A ON) and
training the data
training the data
set.
set.
Applying
Checking
SVM(PCA OFF)
STEPS and training the
the accuracy of
the process.
data set.
Concluding
the Project.
Importing data set to Matlab
ORIGINAL DATA
SET SCATTER PLOT
Training
Data set
PCA is
Applied
Now the PCA is on (Enabled) the accuracy was
observed to be 71.9%
PCA can be abbreviated as Principal
Component Analysis.
What Is PCA
? PCA can be mainly used for Dimensionality
Reduction
The key aim of PCA is to reduce the number of
variables of a data set while preserving as
much information as possible.
SVM is applied
(PCA IS ON) To
Data set
When SVM was
applied to data set
and Trained with
PCA on accuracy
was observed to be
75.7%
Linear SVM
SVM stands for Support Vector Machine.
which is used for Classification as well as Regression problems.
The goal of the SVM algorithm is to create the best line that can
segregate n-dimensional space into classes so that we can easily
put the new data point in the correct category in the future.
Linear SVM is used for linearly separable data, which means if a
dataset can be classified into two classes by using a single straight
line.
CONCLUSION
The accuracy of the Data
is 71.9% when trained
data with PCA ON.
The accuracy of the Data
is 75.7% (SVM ) (PCA
ON)
Ø The accuracy of the Data
is 90% (SVM ) (PCA OFF)
THANK
YOU
