A PROJECT SEMINAR

ON

“Content Based Image Retrieval Framework

for Speech Annotated Images”

PRESENTED BY:
Cherukuri Nishma - 111517106020
Konduru Joshita - 111517106057
Cherukuri Jyothirmayi -111517106019

Under the Esteemed Guidance of

Mrs. R. Aarthi , M.E,(Ph.D)
Assistant professor,
Electronics and communication
Engineering
 ABSTRACT

 To search for a relevant image from an archive is a challenging

research problem for computer vision research community.

Most of the search engines retrieve images on the basis of traditional

text-based approaches that rely on captions and metadata.

In the last two decades, extensive research is reported for content-
based image retrieval (CBIR), image classification, and analysis.

 In this project we proposed to develop a novel framework for

content based image retrieval for speech annotated digital images.
AUTHORS JOURNAL NAME AND YEAR METHODOLOGY

Illustrated about energy

computation and bandwidth along
C. Chelba m Soft indexing of speech content with are ensuing in an awesome
Jorge silva for search in spoken documents sum of records of diverse types
Alex acero and 2006 being created, replaced, and
accumulated. Speech hunt has less
interest possibly due to the huge
albums of spoken material have
formerly no longer been to be had.
 
J Mamou Metric inverted - an efficient Described a retrieving information
Benjamin Zander inverted indexing method for from speech data using phonetic
Yosi Mass metric spaces and 2008 search. The paradigm of query
expansion to speech retrieval and
consists of phonetically expanding
the query for improving search on
phonetic transcripts.
AUTHORS JOURNAL NAME AND YEAR METHODOLOGY

Discovered the NLG method for

Pradipta Biswas Designing inclusive interfaces another new application
Peter Robinson through user modelling and supporting challenged children to
Patrick Langdon simulation and 24th June 2011 be in part of discussion. They have
used a supple technique where
importance is given on suppleness
and usability of the device

Yuk Wah Wong Comparative experiments on Proposed a unique statistical

Razvan Bunescu learning information extractors dealing towards semantic parsing,
Ruifang Ge for proteins and their WASP for building a whole,
Rohit J Kate interactions and July 2004 meaning of a sentence in formally.
The primary novelty of WASP is its
use of most recent ideas in
statistical device conversion
methods.
AUTHORS JOURNAL NAME AND METHODOLOGY
YEAR

John Eakins Content based image The effectiveness of all current CBIR
Margaret graham retrieval- university of systems is inherently limited by the
northumbria at Newcastle fact that they can operate only at the
-2009 primitive feature level. None of them
can search effectively for, say, a photo
of a dog - though some semantic
queries can be handled by specifying
them in terms of primitives

we need effective and efficient

techniques that meet user
Manesh kokare A survey on current content requirements, to access large volumes
BN Chatterji based image retrieval of digital images and video data. The
P k biswas methods survey includes a of system design
and 26 march 2015 and applications of CBIR, image
feature representation and extraction,
Multidimensional indexing.
AUTHORS JOURNAL NAME AND YEAR METHODOLOGY

The growth in reputation of digital

Sakthidasan Content based image retrieval camera spots in the direction of
Sankaran process for speech annotated growing number of customers with
K. Kavitha digital images and 2014 huge album of digital images in
S. Haritha priya their computers which includes
gloss and retrieval.
Multidimensional scaling is used to
identify n-best users to deal with
recognition errors and is converted
into an image-like sample. 

Michael o . Content based image retrieval A content-based image retrieval

Odetayo approach for biometric security approach is proposed for biometric
Kashif Iqbal using color, texture and shape security. It is based on colour,
Anne James features controlled by fuzzy texture and shape features. Colour
heuristics and july 2012 histogram is used to extract the
colour features of an image. Gabor
filter is used to extract the texture
features.
 Project Introduction

The shared and stored multimedia data are growing, and to search or to
retrieve a relevant image from an archive is a challenging research
problem.

The fundamental need of any image retrieval model is to search and

arrange the images that are in a visual semantic relationship with the
query given by the user.

Most of the search engines on the Internet retrieve the images based on
text-based approaches that require captions as input.

The difference in human visual perception and manual

labeling/annotation is the main reason for generating the output that is
irrelevant
 Problem Statement

 In this project we proposed to develop a novel

framework for content based image retrieval for speech
annotated digital images.
 EXISTING SYSTEM
It is near to impossible to apply the concept of manual labeling to
existing large size image archives that contain millions of images.

The second approach for image retrieval and analysis is to apply an

automatic image annotation system that can label image on the basis of
image contents.

The approaches based on automatic image annotation are dependent on

how accurate a system is in detecting color, edges, texture, spatial layout,
and shape-related information.

Significant research is being performed in this area to enhance the

performance of automatic image annotation, but the difference in visual
perception can mislead the retrieval process.
 EXISTING SYSTEM

Content-based image retrieval (CBIR) is a framework that can

overcome the abovementioned problems as it is based on the visual
analysis of contents that are part of the query image.

To provide a query image as an input is the main requirement of CBIR

and it matches the visual contents of query image with the images that
are placed in the archive, and closeness in the visual similarity in terms
of image feature vector provides a base to find images with similar
contents.

In CBIR, low-level visual features (e.g., color, shape, texture, and
spatial layout) are computed from the query and matching of these
features is performed to sort the output 
 EXISTING SYSTEM

 According to the literature, Query-By-Image Content (QBIC) and

Simplicity are the examples of image retrieval models that are based
on the extraction of low-level visual semantic.

 After the successful implementation of the abovementioned models,

CBIR and feature extraction approaches are applied in various
applications such as medical image analysis, remote sensing, crime
detection, video analysis, military surveillance, and textile industry.
 Disadvantages Of
EXISTING SYSTEM
1.) Not accurate.
2.)Highly complex.
3.) Image retrieval Efficiency is very less.
4.) Time consuming method.
5.)Requires huge hardware.
6.) High processing time.
7.) Existing Approaches cannot detect and retrieve the speech
annotated images.
8.)Consumes huge power .
9.) High Operational and maintenance cost.
10.)Computationally Very Complex
 Proposed Work:
In this project we proposed to develop a novel framework for
content based image retrieval for speech annotated digital images.

The schematic block overview of the proposed Optical image

classification algorithm is shown .

The proposed system first creates a database of several significant

features derived from the different standard high resolution optical
satellite images .

While Creating the data base each data base image is preprocessed
first and then processed with the robust Discrete Wavelet
Transform.
 Proposed System

Fig: Schematic Block diagram of the proposed system.

 Proposed System
Several significant object frequencies and features are extracted
from the DWT processed speech annotated images and all these
features are stored in the data base.

 Now after deploying the system according the operational flow

described in fig.

First we consider the speech annotated image under test and

subject it to preprocessing where its quality will be improved.

Next after enhancing the quality of the speech annotated image,

the preprocessed speech annotated image is subjected to the DWT
analysis to obtain the sub band features at fine to coarser levels.
 Proposed System

 A sub band analysis is carried out to extract the features and to

construct the feature vector.

The constructed feature vector of the test speech annotated image

is subjected to the comparison with the database features and based
on the matching status the corresponding data or images will be
retrieved from the database.

This system finds its applications in almost all diverse fields due its
ability in pattern detection and recognition applications.
 Advantages of the Proposed System

1.) Highly accurate.

2.)Very simple and flexible framework.
3.) Image retrieval Efficiency is very high.
4.) Implements a high speed process.
5.)Requires very less hardware.
6.) Reduced processing time.
7.) Can detect and retrieve the speech annotated images with high

precision.
8.)Consumes less power .
9.) Low Operational and maintenance cost.
10.)Computationally Very Efficient.
 Tools & Techniques

 Matlab Software

 Image Processing Toolbox

output
References

D. A. James, “A system for unrestricted topic retrieval from radio news
broadcasts,” in Proc. IEEE Int. Conf. Acoustics, Speech, and Signal
Processing, pp.279–282,1994.
B. Logan, J.-M. Van Thong, and P. J. Moreno, “Approaches to reduce
the effects of OOV queries on indexed spoken audio,” IEEE Trans.
Multimedia, vol. 7, no. 5, pp. 899–906, Oct.2005.
K.Sakthidasan @ Sankaran, S. Bhuvaneshwari and Dr.V.Nagarajan “A
new edge preserved technique using iterative median filter” in IEEE
International Conference on Communication and Signal
Processing(ICCSP 2014), pp:1750 – 1754, 2014.
J. G. Wilpon and L. R. Rabiner, “A modified K-mea ns clustering
algorithm for use in isolated work recognition,” IEEE Trans. Acoust.,
Speech, Signal Process., vol. ASSP-33, no. 3, pp. 587–594,1985.
1. K.Sakthidasan @ Sankaran, G . Ammu and Dr.V.Nagarajan “Non local image
restoration using iterative method” in IEEE International Conference on
Communication and Signal Processing(ICCSP 2014), pp:1740-1744,2014.

2. C.-H.Wu and Y.-J. Chen, “Multi-keyword spotting of telephone speech using a fuzzy
search algorithm and keyword-driven two-level CBSM,” Speech Commun., vol. 33,
pp. 197– 212, 2001.

3. B. Chen, H.-M.Wang, and L.-S. Lee, “Disc riminating capabilities of syllable-based

features and approaches of utilizing them for voice retrieval of speech information in
Mandarin Chinese,” IEEE Trans. Speech Audio Process., vol. 10, no. 5, pp. 303–314,
Jul. 2002.

K. Sakthidasan alias Sankaran & V. Nagarajan, “Noise Removal Through the Exploration
of Subjective and Apparent Denoised Patches Using Discrete Wavelet Transform”, IETE
Journal of Research, ISSN