10-06-2017, 07:08 AM
An Efficient K-Means Cluster Based Image Retrieval Algorithm using Learning: An Innovative Approach from Relevance Feedback
[attachment=548]
INTRODUCTION
content-based image retrieval has been an active
research area in recent years. The interest in this
research area has inspired from the need to search
and well manage large volumes of Multimedia
information [1, 6]. CBIR extracts low-level features
which is Inbuilt in the images to present the contents of
images. Each image has Visual features such as
classified into three main classes: color [2, 11], texture
[9, 11] and shape [2, 10] features. Color is an important
image feature such as used in Content-Based Image
Retrieval [2, 9, 13]. These features have potential to
identify objects [10] and retrieve similar images on the
basis of their contents.
Proposed Work
Each image has three features Color, Shape and Texture.
For fast and improve Image retrieval performance we are
using color feature extraction. Using color feature
extraction firstly we converted color image into grey
level, this is containing values from 0 to 255.
Result and Analysis
In this section, some experiments are conducted in
order to test the performance of histograms and here
we used James Z. Wang et al database. [6] To test
the proposed method. Firstly image converted to
grayscale image. Then the image were quantized
and the features described in section II were
calculated which is based on coherent and
incoherent clusters. Followed these feature set;
images were grouped in similar clusters using Kmeans
clustering method.
Conclusion
The algorithm is based on the color and texture features
of images. The grayscale values mean variance, contrast
level and various sizes of the intensity values are
considered as appropriate features for retrieval. We have
shown that k-means clustering is quite useful for
relevant image retrieval queries. One achievable way for
future work is to further improve the cluster picking
method by investigating heuristic functions.