Multiple Classifier Based on Fuzzy C-Means for a Flower Image Retrieval

1
Multiple Classifier Based on Fuzzy C-Means for a
Flower Image Retrieval
Keita Fukuda, Tetsuya Takiguchi, Yasuo
Ariki Graduate School of Engineering, Kobe
University, Japan
2
Table of contents

• Introduction
• Purpose of Multiple Classifier based on Fuzzy
  C-means
• Overview of our flower recognition system
• Proposed Method
• Each Classifier
• Fuzzy C-means
• Experiments
• Summary and Future Work

3
Introduction
What is this flower ?
Retrieval system requires keywords. But it is
difficult to get keywords from images.
We are focusing on flower image retrieval system
In our proposed method
We take a flower picture and send it to a system.
We receive flower image information there and
then immediately.
4
Conventional techniques

• Conventional method
• Using the same features for classification.
• ?But flowers have various shape.
• We propose multiple classifier which selects
  important features for each flower type and
  weights the importance on each classifier using
  Fuzzy c-means.

It is required to select important features
according to flower type.
5
Overview of our system
Flower region extraction
Send image
Color and shape features extraction
Receive information
Database
Similarity by multiple classifier
contents based flower image retrieval
6
Flower region extraction
Color and Shape features are computed on them.
A large color regions locating at near center are
extracted as flower region
7
Feature extraction
Distribution histogram
Color feature
100 segments
10
10
Shape feature
d
l
G
d
Gravity to contour
8
Recognition with multiple classifier (1)
Which types is a query image associated with ?
Similarity for each classifier is calculated
Membership of a query image in each type is
obtained as weight for each similarity
We define 3 classifiers for 3 flower types
Linearly coupled similarity matching of 3
classifiers
Membership of query image (Weight)
Multiple classifier
Similarity for classifier
Query image
Similarity
0.03
0.93
0.04
Image
Features, Information, Similarity
Database

9
Each flower type
We define 3 classifiers for 3 flower types
A Near circle
B Clear one petal
C Many petals
Type Classifier For Similarity
A FA Near circle flowers ViA
B FB Clear one petal flowers ViB
C FC many petals flowers ViC
The similarity in each classifier is computed
using Weighted Histogram Intersection. The value
of weight represents the difference of each
classifier
10
Each Classifier
Characteristics
Peak (5) the number of petal
Important similarity
Query image
Gaussian Weight
Histogram Intersection
image
11
Recognition with multiple classifier (2)
Membership of query image (Weight)
Multiple classifier
Similarity for classifier
Query image
Similarity
0.03
0.93
0.04
Image
Features, Information, Similarity
Database

Weight for each similarity is membership of a
query image in type A, B and C ? It is difficult
that all flowers are classified into one of 3
types clearly.
12
Fuzzy C-means
It is based on minimization of the following
objective function
Fuzzy partitioning is carried out through an
iterative optimization of the objective function,
with the update of membership uij and the cluster
centers cj
Membership property is
Data elements can belong to more than one
cluster. associated with each element is a set of
membership.
13
Fuzzy C-means For flower retrieval system

1. Database images are clustered using fuzzy
   c-means.
2. Membership of a query image is computed.

Membership of a query image is obtained as weight
for each similarity
membership
B Clear one petal
C Many petals
A Near Circle
Data elements
Input data shape features compactness, entropy,
average Output data membership of the image in
each type.
14
Recognition with multiple classifier (3)
Multiple classifier
Similarity for classifier
Membership (Weight)
Query image
Similarity
0.03
0.93
0.04
Image
Features, Information, Similarity
Database

Linearly coupled similarity matching of 3
classifiers is calculated. This example, the
similarity between image i and a query image
15
Result information
Result information
Input image
Result information are shown to users up to fifth
rank based on the similarity M(i)
16
Experimental condition

• Flower images of 120 species with each 4 samples.
• (i.e. 480 images in total).
• Four Cross validation (evaluate cumulative
  recognition)
• One sample is used as a query image (120).
• The others are used as the database images
  (1203).

17
Conventional method
y
Shape features
Color features

• Compactness
• The number of petal (peak)
• Moment
• The ratio of the shortest width over the longest

• Largest segment
• X coordinate
• Y coordinate
• Its distributed value
• 2nd Largest segment
• X coordinate
• Y coordinate
• Its distributed value

x
peak
18
Experimental result
1st 3rd 5th 10th
Conventional method Conventional method 33.8 59.6 69.4 80.8
Multiple classifier No fuzzy 39.8 67.1 78.1 89.4
Multiple classifier fuzzy 42.7 69.6 81.3 92.5
Proposed method
Conventional method
query
19
Summary
Multiple Classifier Based on Fuzzy C-Means for a
Flower Image Retrieval
New concept multiple classifier which select
important features for each flower type
In future work research for more than three
classifiers
20
Thank you