Abstract:
Content-based image retrieval (CBIR) systems
are a special type of Information Retrieval (IR) system
where the elements in the repository are pictures. IR
works with finding digital resources in large databases.
CBIR is developed to retrieve the desired target image
from the large collection of images based on the
contents of the given query image. The Contents of
image can be extracted from any images which are
specified color, shape, texture or any other
characteristics of images. The system derived two types
of different color features - color moment and color
auto-correlogram. The moments of image can be used
to indicate color distribution of an image which can be
described as a probability distribution of colors. To
calculate the spatial correlation of pairs of color
different with distance, color correlogram is introduced
in the system. Gabor wavelets are used to express
texture of natural image. The characteristics of Gabor
wavelets are similar to those of human visual features.
The system firstly created the features vector with these
three types of features and used to get higher retrieval
results of system. To gives the better result in retrieval
of system and classify the query image, Support Vector
Machine classifier is used with the combination of
these visual features.
