Transcript 按一下以新增標題

Wavelet-Based Multiresolution
Matching for
Content-Based Image Retrieval
Presented by
Tienwei Tsai
Department of Computer Science and Engineering
Tatung University
2005/12/10
Outline
1. Introduction
2. Related Works
3. Proposed Image Retrieval System
4. Experimental Results
5. Conclusions
1. Introduction
• Two approaches for image retrieval:
– query-by-text (QBT): annotation-based image
retrieval (ABIR)
– query-by-example (QBE): content-based
image retrieval (CBIR)
• Standard CBIR techniques can find the
images exactly matching the user query
only.
• In QBE, the retrieval of images basically has
been done via the similarity between the query
image and all candidates on the image database.
– Euclidean distance
• Transform type feature extraction techniques
– Wavelet, Walsh, Fourier, 2-D moment, DCT, and
Karhunen-Loeve.
• In our approach, the wavelet transform is used
to extract low-level texture features.
2. Related Works
• Content-based image retrieval is a technology to search
for similar images to a query based only on the image
pixel representation.
– However, the query based on pixel information is quite timeconsuming
– Therefore, how to choose a suitable color space and reduce the
data to be computed is a critical problem in image retrieval.
• Some of the systems employ color histograms.
– The histogram measures are only dependent on summations of
identical pixel values and do not incorporate orientation and
position.
• Therefore, we propose an image retrieval scheme to
retrieve images from their transform domain, which tries
to reduce data and still retains their local information.
• In this paper, we focus on the QbE approach.
The user gives an example image similar to the
one he/she is looking for.
• Finally, the images in the database with the
smallest distance to the query image will be
given, ranking according to their similarity.
– We can define the QbE problem as follows. Given a
query image Q and a database of images X1, X2,…,
Xn, find the image Xi closest to Q. The closeness is to
be computed using a distance measuring function
D(Q, Xn).
3. The Proposed Image
Retrieval System
Figure 1. The proposed system architecture.
Feature Extraction
• Features are functions of the
measurements performed on a class of
objects (or patterns) that enable that class
to be distinguished from other classes in
the same general category.
• Color Space Transformation
RGB (Red, Green, and Blue) ->
YUV (Luminance and Chroma channels)
YUV color space
• YUV is based on the CIE Y primary, and also
chrominance.
– The Y primary was specifically designed to follow the
luminous efficiency function of human eyes.
– Chrominance is the difference between a color and a
reference white at the same luminance.
• The following equations are used to convert from
RGB to YUV spaces:
– Y(x, y) = 0.299 R(x, y) + 0.587 G(x, y) + 0.114 B(x, y),
– U(x, y) = 0.492 (B(x, y) - Y(x, y)), and
– V(x, y) = 0.877 (R(x, y) - Y(x, y)).
Wavelet Transform
• Mallat' s pyramid algorithm
Figure 2.
Figure 3.
Distance Measurement
• In our experimental system, we define a
measure called the sum of squared differences
(SSD) to indicate the degree of distance (or
dissimilarity).
• The distance between Q and Xn under the LL(k)
subband can be defined as

DLL( k ) (Q, X n )   LL (m, n)  LL (m, n)
m
n
(k )
q
(k )
xn

2
• Based on the observation that subimages LL(1),
LL(2), …, LL(k) correspond to different resolution
levels of the original image, the retrieving
accuracy may be improved to consider these
subimages at the same time.
• Therefore, the distance between Q and Xn can
be modified as the weighted combination of
these subimages:
D(Q, X n )   wk DLL( k ) (Q, X n )
m
where wk is the weight of the distance under the
kth resoluiton level.
4. Experimental Results
• 1000 images downloaded from the WBIIS
database are used to demonstrate the
effectiveness of our system.
• The user can query by an external image or an
image from the database.
• To evaluate the retrieval efficiency of the
proposed method, we use the performance
measure, the precision rate, as follows:
Rr
precision rate 
Tr
where Rr is the number of relevant retrieved items, and
Tr is the number of all retrieved items.
Figure 4. Retrieved results via the comparison
of the original RGB images.
Figure 5. Retrieved results based on the Y-component images.
Figure 6. Retrieved results based on the
Y-component LL(1) subimages.
Figure 7. Retrieved results based on the
Y-component LL(2) subimages.
Figure 8. Retrieved results based on the
Y-component LL(3) subimages.
Figure 9. Retrieved results based on the
Y-component LL(4) subimages.
Figure 10. Retrieved results based on the
Y-component LL(5) subimages.
Figure 11. Retrieved results based on the
Y-component LL(6) subimages.
Figure 12. Retrieved results based on the combination of
the Y-component LL(2) and LL(3) subimages.
Figure 13. Retrieved results based on the combination of
the Y-component LL(3), LL(4), and LL(5) subimages.
5. Conclusions
• In this paper, we propose a CBIR method
based on the color space transformation
and the wavelet transform.
• We find that through the combination of
the features with different resolution level,
we can not only obtain the better precision
rate but also the good reduction rate.
Future Works
• Since only preliminary experiment has been
made to test our approach, a lot of works should
be done to improve this system:
– Since several features may be used simultaneously, it
is necessary to develop a scheme that can integrate
the similarity scores resulting from the matching
processes.
– A long-term aim is combining the semantic
annotations and low-level features to improve the
retrieval performance. That is, the retrieved images
should be somehow related to the objects contained
in the scenes.
Thank You !!!