1 COLOR BASED VIDEO RETRIEVAL USING BLOCK …...2.1 Color based video retrieval using DCT The DCT block computes the unitary discrete cosine transform (DCT) of each channel in the

International Journal of Information Technology & Management Information System (IJITMIS), ISSN 0976 –

6405(Print), ISSN 0976 – 6413(Online), Volume 5, Issue 2, May - August (2014), pp. 01-14 © IAEME

1

COLOR BASED VIDEO RETRIEVAL USING BLOCK AND GLOBAL

METHODS

Smita Chavan

Information Technology Department, Government Engineering College,

Osmanpura, Aurangabad, India

ABSTRACT

Interacting with multimedia data and video in particular, requires more than

connecting with data banks and delivering data via networks to customer’s homes or offices.

We still have limited tools and applications to describe, organize, and manage video data.

The fundamental approach is to Index video data and make it a structured media manually

generating video. Color features are key-elements which are widely used in video retrieval,

video content-analysis, and object recognition. In the present study we are using the video

retrieval technique, based on color content, the standard way of extracting a color from an

image is to generate a color histogram. Therefore, the core research in content-based video

retrieval is developing technologies to automatically parse video, audio, and text to identify

meaningful composition structure and to extract and represent content attributes of any video

sources. A typical scheme of video-content analysis and indexing, as proposed by many

researchers, involves four primary processes: feature extraction, structure analysis,

abstraction, and indexing. Each process poses many challenging research problems. The main

processes in a proposed system includes search image, apply two methods for color feature

extraction, find matching videos as compared with image input given, calculation of distance

metrics using Euclidean distance measurement.

Keywords: Absolute Difference, Distance Metrics, Euclidean Distance, Feature Extraction,

Similarity Measurement.

1. INTRODUCTION

Content-Based video search is according to the user-supplied in the bottom

Characteristics, directly find out images containing specific content from the image library

the basic process: First of all, do appropriate pre-processing of images like size and image

INTERNATIONAL JOURNAL OF INFORMATION TECHNOLOGY &

MANAGEMENT INFORMATION SYSTEM (IJITMIS)

ISSN 0976 – 6405(Print)

ISSN 0976 – 6413(Online)

Volume 5, Issue 2, May - August (2014), pp. 01-14

© IAEME: http://www.iaeme.com/IJITMIS.asp

Journal Impact Factor (2014): 6.2217 (Calculated by GISI) www.jifactor.com

IJITMIS

© I A E M E



2

transformation and noise reduction is taking place, and then extract image characteristics

needed from the image according to the contents of images to keep in the database. When we

retrieve to identify the image, extract the corresponding features from a known image and

then retrieve the image database to identify the images which are similar with it, also we can

give some of the characteristics based on a query requirement, then retrieve out the required

images based on the given suitable values. In the whole retrieval process, feature extraction is

essential it is closely related to all aspects of the feature, such as color, shape, texture and

space. A color image is a combination of some basic colors. In MATLAB breaks each

individual pixel of a color image termed true color down into Red, Green and Blue values.

We are going to get as a result, for the entire image is 3 matrices, each one representing color

features. The three matrices are arranging in sequential order, next to each other creating a 3

dimensional m by n by 3 matrixes. An image which has a height of 5 pixels and width of 10

pixels the resulting in MATLAB would be a 5 by 10 by 3 matrixes for a true color image. For

improved retrieval performance, results from the different content retrieval methods may be

combined. Fusion of multimedia search results is query-dependent, and an area of ongoing

research. Besides merging results from different content retrieval methods, when retrieving

whole programs, it is necessary to combine shot-level results from within the program. There

has not been much work in this area, and approaches consist of assigning binary values to

each shot and aggregating these to the program level, averaging the scores of all shots for a

particular feature, using the maximum score of a shot in a program, or simply using the

central shot in a program to represent the whole video[1].Overall paper contains section I

intoduction,section II history, section III methods, section IV implementation, section V

experimental Result with BCVR and GCVR, section VI conclusion.

1.1 Color Features

Fig-1.1: Basic Content Based Video Retrieval Method

Image

Query

Feature

Extraction

Query

Image

Similarity

Measures

Retrieved

Images

Feature

DB

Feature

Extraction

Image

Collection



3

Color is one of the most widely used visual features in multimedia context and image

/ video retrieval, To support communication over the Internet, the data should compress well

and be suitable for heterogeneous environment with a variety of the user platforms and

viewing devices, large scatter of the user's machine power, and changing viewing conditions.

The CBVR systems are not aware usually of the difference in original, encoded, and

perceived colors, e.g., differences between the colorimetric and device color data. It is also an

important feature of perception, comparing with other image features such as texture and

shape etc, color features are very stable and robust[2].

It is not sensitive to rotation, translation and scale changes. Moreover, the color

feature calculation is relatively simple. A color histogram is the most used method to extract

color features.

2. HISTORY

Shape-based image/video retrieval is one of the hardest problems in general

image/video retrieval. This is mainly due to the difficulty of segmenting objects of interest in

the images. Consequently, shape retrieval is typically limited to well distinguish objects in

the image. In order to detect and determine the border of an object an image may need to be

preprocessed. The preprocessing algorithm or filter depends on the application. Different

object types such as skin lesions, brain tumors, persons, flowers, and airplanes may require

different algorithms. If the object of interest is known to be darker than the background, then

a simple intensity thresholding scheme may isolate the object. For more complex scenes,

noise removal and transformations invariant to scale and rotation may be needed. Once the

object is detected and located, its boundary can be found by edge detection and boundary

following algorithms. The detection and shape characterization of the objects becomes more

difficult for complex scenes where there are many objects with occlusions and shading. Once

the object border is determined the object shape can be characterized by measures such as

area, eccentricity (e.g., the ratio of major and minor axes), circularity (closeness to a circle of

equal area), shape signature (a sequence of boundary-to-center distance numbers), shape

moments, curvature (a measure of how fast the border contour is turning), fractal dimension

(degree of self similarity) and others. All of these are represented by numerical values and

can be used as keys in a multidimensional index structure to facilitate retrieval.

In dimensionality color histograms may include hundreds of colors. In order to

efficiently compute histogram distances, the dimensionality should be reduced. Transform

methods such as K-L and Discrete Fourier Transform (DFT), Discrete Cosine Transform

(DCT) or various wavelet transforms can be used to reduce the number of significant

dimensions. Another idea is to find significant colors by color region extraction and compare

images based on presence of significant colors only partitioning strategies such as the

Pyramid-Technique map n-dimensional spaces into a one-dimensional space and use a B+

tree to manage the transformed data. The resulting access structure shows significantly better

performance for large number of dimensions compared to methods such as R-tree variants.

Texture and shape retrieval differ from color in that they are defined not for individual

pixels but for windows or regions. Texture segmentation involves determining the regions of

image with homogeneous texture. Once the regions are determined, their bounding boxes

may be used in an access structure like an R-tree. Dimensionality and cross correlation

problems also apply to texture and can be solved by similar methods as in color. Fractal

codes capture the self similarity of texture regions and are used for texture based indexing

and retrieval. Shapes can be characterized by methods and represented by n-dimensional



4

numeric vectors which become points in n-dimensional space. Another approach is to

approximate the shapes by well defined, simpler shapes. For instance, triangulation or a

rectangular block cover can be used to represent an irregular shape. In this latter scheme, the

shape is approximated by collection of triangles or rectangles, whose dimensions and

locations are recorded. The advantage of this approach is that storage requirements are lower,

comparison is simpler and the original shape can be recovered with small error [3].

2.1 Color based video retrieval using DCT

The DCT block computes the unitary discrete cosine transform (DCT) of each

channel in the M-by-N input matrix, u. y = dct(u) % Equivalent MATLAB code When the

input is a sample-based row vector, the DCT block computes the discrete cosine transform

across the vector dimension of the input. For all other N-D input arrays, the block computes

the DCT across the first dimension. The size of the first dimension (frame size), must be a

power of two. To work with other frame sizes, use the Pad block to pad or truncate the frame

size to a power-of-two length. When the input to the DCT block is an M-by-N matrix, the

block treats each input column as an independent channel containing M consecutive samples.

The block outputs on-by-N matrix whose lth column contains the length-M DCT of the

corresponding input column [4].

Fig 2.1: Comparison of DCT with row and column

3. METHODS

Content-Based video search is according to the user-supplied in the bottom

Characteristics, directly find out images containing specific content from the image library

the basic process, First of all, do appropriate pre-processing of images like size and image

transformation and noise reduction is taking place, and then extract image characteristics

needed from the image according to the contents of videos to keep in the database. When we

retrieve to identify the video, extract the corresponding features from a known image and

then retrieve the video database to identify the images which are similar with it, also we can

give some of the characteristics based on a query requirement, then retrieve out the required

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

DCT ROW COLOUMN

Series1



5

videos based on the given suitable values. In the whole retrieval process, feature extraction is

essential it is closely related to all aspects of the feature, such as color, shape, texture and

space.

Feature extraction based on the color characteristics in the proposed method is

classified into two types.

3.1 Global color based method

Because an RGB color space does not meet the visual requirements of people, for

video retrieval, the video normally converts from an RGB space to other color spaces. The

HSV space is used as it is a more common color space.

The global color is a simple way of extracting image features. High effective

calculation and matching are its main advantage. The feature is invariable to rotation and

translation.

The drawback is that the global color only calculates the frequency of colors. The

spatial distribution of color information is lost. Two completely different images can get the

same global color difference graph, which will cause retrieval errors.

3.2 Block color based method

For the block color, an image is separated into n×n blocks. Each block will have less

meaning if the block is too large, while computation of retrieval process will be increased if

the block is too small. A two-dimensional space divided into 3 × 3 will be more effective. For

each block, we carry out a calculation of the color space converting and color quantization.

The normalized color features for each block can be calculated. Usually in order to highlight

the specific weight of different blocks, a weight coefficient is distributed to each block, and

the weight of middle block is often larger [5].

3.3 Similarity of Frame sequence

Video access is one of the important design issues in the development of multimedia

information system, video-on-demand and digital library. Video can be accessed by attributes

of traditional database techniques, by semantic descriptions of traditional information

retrieval technique, by visual features and by browsing. Access by attributes of database

technique and access by semantic descriptions of information retrieval technique are

insufficient for video access owing to the numerous interpretations of visual data. Besides,

the automatic extraction of semantic information from general video programs is outside the

capability of the current technologies of machine vision.

3.3.1 Symmetric similarity measures

A similarity measure is symmetric if D (U, V) = D (V, U). The straightforward

measure of similarity is the one to- one optimal mapping. In the one to one mapping, we try

to map as many pairs of frames as possible under the constraint that each frame ui in U

corresponds to only one frame vj in V. Obviously, the maximal number of mapping pairs is

equal to the number of frames of shorter shot (shot with less number of frames). The

mapping with minimum sum of frame distance is selected as the optimal mapping. The

formal definitions are given as follows. The effect of video segmentation also affects the

performance of sequence mapping. Video segmentation with lower threshold produces shots

with much variation. It is possible that the shot U is very similar to the V except that some

frames are very dissimilar. Using OM, these dissimilar frame pairs produce large distance.

Therefore, in the next definition, the mapping is constrained by a threshold. Two frames with



6

distance larger than are not allowed to match. In addition, each unmatched frame is

associated with a penalty value v in the computation of shot distance. Otherwise, the result of

the distance-constrained optimal mapping with minimum cost would be no matching at all.

No matching produces the distance of zero.

3.3.2 Asymmetric similarity measures

A similarity measure is asymmetric if D (U, V) ¹ D (V, U). In general, asymmetric

similarity measure is used to map between the query sequence of frames and video sequence

of frames. The simplest proposed asymmetric similarity measure is the Optimal Subsequence

Mapping (OSM). The algorithm of OSM is similar to that of OM except that the query video

sequence must be the shorter sequence. Therefore, it is not necessary to compare the length

between the video shot U and query video shot Q [6].

3.4 Find matching videos Here absolute difference (AD) is used for similarity measurement of feature vectors of

the videos in content based video retrieval. In this method, the feature vectors of a query

video are compared with feature vectors of all the videos stored in the database by taking

absolute difference between them. The differences will be ordered in increasing fashion so as

to show most relevant searches on top, as the videos with which the absolute difference will

be lower will be most relevant ones[4].

3.5 Calculation of Distance metrics In the image database and video database proposed method compares with the images

given in query and compare with video database features as with each frames from the video

.for the each frames method calculates Euclidean distance takes the mean of the distances

calculated and then it is stored into the feature vector.

4. IMPLEMENTATION

For the purpose of implementation MATLAB 2012 is used. MATLAB is a "Matrix

Laboratory", and as such it provides many convenient ways for creating vectors, matrices,

and multi-dimensional arrays. In the MATLAB vernacular, a vector refers to a one

dimensional (1×N or N×1) matrix, commonly referred to as an array in other programming

languages. A matrix generally refers to a 2-dimensional array, i.e. an m x n array where m

and n are greater than or equal to 1. Arrays with more than two dimensions are referred to as

multidimensional arrays. Operating system windows 7 is used for the proposed method.

The Software Requirement: Matlab (2012a), The Hardware Requirement: The RAM:

2 GB with Operating System Windows 7/XP.

The first step for video retrieval based on the partitioning of a video sequence into

shots. A shot is defined as an image sequence. Key-frames are images extracted from original

video data. Key-frames have been frequently used to supplement the text of a video log,

though they were selected manually in the past .Key-frames, if extracted properly, are a very

effective visual abstract of video contents and are very useful for fast video browsing. key-

frames can also be used in representing video in retrieval video index may be constructed

based on visual features of key-frames, and queries may be directed at key-frames using

query by retrieval algorithms.

Next step is to extract features. The features are typically extracted with the above

mentioned methods i.e. For color feature extractions methods used is global and block color



7

based system. Low-level features such as object motion, color, shape, texture, loudness,

power spectrum, bandwidth, and pitch are extracted directly from video in the database.

High-level features are also called semantic features. Features such as timbre, rhythm,

instruments.

Next step is to find matching videos with both methods using Euclidean distance

formula. Each video is divided into five frames. For each frame Euclidean distance is

calculated take mean of that Euclidean distance, arrange it into ascending order, we will get

most closed matching videos to the given image.

4.1 Color description In general, color is one of the most dominant and distinguishable low-level visual

features in describing video image. Many CBIR systems employ color to retrieve images In

theory, it will lead to minimum error by extracting color feature for retrieval using real color

image directly, but the problem is that the computation cost and storage required will expand

rapidly In fact, for a given color video image, the number of actual colors only occupies a

small proportion of the total number of colors in the whole color space. In the HSV color

space, each component occupies a large range of values. If we use direct values of H, S and V

components to represent the color feature, it requires lot of computation. So it is better to

quantify the HSV color space into non-equal intervals. At the same time, because the power

of human eye to distinguish colors is limited, we do not need to calculate all segments.

Unequal interval quantization according the human color perception has been applied on H, S

and V components [7].

4.1.1. Color set notation

Transformation

The color set representation is defined as follows: given the color triple (r; g; b) in the

3-D RGB color space and the transform T between RGB and a generic color space denoted

XY Z, for each (r; g; b) let the triple (x; y; z) represent the transformed color such that (x; y;

z) = T(r; g; b):

Quantization

Let QM be a vector quantizer function that maps a triple (x; y; z) to one of M bins.

Then m, where m is the index of color (x; y; z) assigned by the quantizer function and is

given by m = QM(x; y; z):

Binary Color Space

Let BM be an M dimensional binary space that corresponds to the index produced by

QM such that each index value m corresponds to one axis in BM.

Definition 1: Color Set, A color set is a binary vector in BM. A color set corresponds to a

selection of colors from the quantized color space. For example, let T transform RGB to HSV

and let QM where M = 8 vector quantize the HSV color space to 2 hues, 2 saturations and 2

values. QM assigns a unique index m to each quantized HSV color. Then B8 is the 8-

dimensionalbinary space whereby each axis in B8 corresponds one of the quantized HSV

colors. Then a color set c contains a selection from the 8 colors. If the color set corresponds

to a unit length binary vector then one color is selected. If a color set has more than one non-

zero value then several colors are selected. For example, the color set c = (10010100)



8

corresponds to the selection of three colors, m = 7; m = 4 and m = 2, from the quantized HSV

color space.

4.2 Database Small Training Databases: The videos are categorized in number of frames as each video is

divided into five frames assorted categories containing 8 similar videos. So, in total there are

8 videos in data set for experimentation. All the videos are normalized in each category. The

videos are chosen such that those of same categories differ very minutely in color content so

as to find accurate result.

Test Database: In proposed method test database contains 8 videos .search image is compared

with the video database for color feature extraction. Comparative difference graph is plotted

by Euclidean distance.

Created video database is used for the proposed method.

Large Training Database: the images are considered as large training database. There are 15

search images in the image database. All the images are compared with video database for

color feature extraction. Following image database is used in the proposed method.

Fig 4.2: Sample Image Database

4.3 Video retrieval algorithm is as follows Step 1: First divide the video into no of frames.

Step 2: Uniformly divide each video in the database and the target image

Step 3: Extract Feature with color based methods for Efficient Video retrieval

Step 4: construct a combined feature vector for color and differential graph



9

Step 5: Find the distances between feature vector of query Video and the feature vectors of

target video using Euclidean distance.

Step 6: Sort the Euclidean distances by comparing the query image and database video in

ascending order

Step 7: Retrieve most similar Video with minimum Distance from Video data.

5. EXPERIMENTAL RESULT WITH BCVR AND GCVR

5.1 Comparisons In the proposed method results are shown with block color based video retrieval and

global color based video r retrieval including calculation of Euclidean distances.

Table 5.1: Search image results

Search image 12 Search image 3 Search image 14

GCVR BCVR GCVR BCVR GCVR BCVR

2.15444e+09 4.13454e+08 5.11385e+09 8.69056e+08 3.30671e+09 6.47014e+08

2.78191e+09 7.09396e+08 6.7226e+09 9.24809e+08 3.96584e+09 7.34128e+08

3.85552e+09 7.26439e+08 6.83174e+09 1.07362e+09 4.64586e+09 8.46765e+08

4.06727e+09 7.39199e+08 7.70484e+09 1.10147e+09 5.36469e+09 8.951e+08

4.86135e+09 7.41091e+09 8.94948e+09 1.11417e+09 6.66659e+09 9.7927e+08

Fig-5.1: Search image comparison

0.00E+00

1.00E+09

2.00E+09

3.00E+09

4.00E+09

5.00E+09

6.00E+09

7.00E+09

8.00E+09

9.00E+09

1.00E+10

GCVR BCVR GCVR BCVR GCVR BCVR

Search image 12 Search image 3 Search image 14

Series1

Series2

Series3

Series4

Series5

Euclidean distance

measurements



10

Fig- 5.2: Difference graph

5.2 Accuracy with GCVR and BCVR

Table 5.2: Search image Accuracy percentage

Search Image

Method

GCVR

Accuracy %

BCVR

Accuracy %

Image 1 75 79

Image 2 65 69

Image 3 85 90

Image 4 90 95

Image 5 50 55

Image 6 90 91

Image 7 45 41

Image 8 89 92

Image 9 65 76

Image 10 32 43

Image 11 67 53

Image 12 89 92

Image 13 91 96

Image 14 92 96

Image 15 54 60



11

Fig 5.2: Search image accuracy by GCVR and BCVR

6. CONCLUSION

Despite the considerable progress of academic research in video retrieval, there has

been relatively little impact of content based video retrieval research on commercial

applications with some niche exceptions such as video segmentation. Choosing features that

reflect real human interest remains an open issue. One promising approach is to use Meta

learning to automatically select or combine appropriate features. Another possibility is to

develop an interactive user interface based on visually interpreting the data using a selected

measure to assist the selection process. We would be interested to see how these techniques

can be incorporated in an interactive video retrieval system. The usefulness of color

correlation statistics in semantic retrieval of video and images. It shows HSV Color is to be

efficient feature over the other more traditional color-based features. MPEG-7 color structure

descriptor is similar to HSV color correlogram and is very efficient in retrieval of

images/videos. This proposed method covers the following tasks: image search that means

search is based on the images. extraction of features of static key frames using color features

specifically HSV color features using conversion of RGB to HSV, video may be TV video,

natural images, collection of frames, video search including interface, similarity measure,

video retrieval and distance metrices.we made program in two parts database creation,

videos, samples, one program is for video file reading then save into database. Search images

option will search images which are closer to the video. Advanced filtering and retrieval

methods for all types of multimedia data are highly needed. Current image and video retrieval

systems are the results of combining research from various fields. Hence color-based retrieval

has been the backbone for content-based image and video retrieval compared with the other

methods like shape or texture.

6.1 Future Scope Many issues are still open and deserve further research, especially in the following

areas. Most current video indexing approaches depend heavily on prior domain knowledge.

This limits their extensibility to new domains. The elimination of the dependence on domain

0

20

40

60

80

100

120

Ima

ge

1

Ima

ge

2

Ima

ge

3

Ima

ge

4

Ima

ge

5

Ima

ge

6

Ima

ge

7

Ima

ge

8

Ima

ge

9

Ima

ge

10

Ima

ge

11

Ima

ge

12

Ima

ge

13

Ima

ge

14

Ima

ge

15

GCVR Accuracy %

BCVR Accuracy %



12

knowledge is a future research problem. Fast video search using hierarchical indices are all

interesting research questions. Video indexing and retrieval in the cloud computing

environment, where the individual videos to be searched and the dataset of videos are both

changing dynamically, will form a new and flourishing research direction in video retrieval in

the very near future. Video affective semantics describe human psycho-logical feelings such

as romance, pleasure, violence, sadness, and anger.

6.2 Applications

Specially used in supercomputers and mainframe computers, because they have high

computation time. In professional and educational activities that involve generating or using

large volumes of video and multimedia data are prime candidates for taking advantage of

video-content analysis techniques. Automated authoring of web content media organizations

and TV broadcasting companies have shown considerable interest in presenting their

information on the web. Automated authoring of web content searching and browsing large

video archives major news agencies and TV broadcasters own large archives of video that

have been accumulated over many years. Intelligent video segmentation and sampling

techniques can reduce the visual contents of the video program to a small number of static

images. Easy access to educational material the availability of large multimedia libraries that

we can efficiently search has a strong impact on education. Indexing and archiving

multimedia presentation.

Consumer domain applications, video content analysis research are geared toward

large video archives. However, the widest audience for video content analysis is consumers.

We all have video content pouring through broadcast TV and cable. Also, as consumers, we

own unlabeled home video and recorded tapes. To capture the consumer’s perspective, the

management of video information in the home entertainment area will require sophisticated

yet feasible techniques for analyzing, filtering, and browsing video by content [8].

7. REFERENCES

[1] R.Venkata Ramana Chary, Dr.D.Rajya Lakshmi and Dr. K.V.N Sunitha, Feature

Extraction Methods for Color Image Similarity Advanced computing, An

International Journal ( ACIJ ), Vol.3, No.2, March 2012, 147-157.

[2] B V Patel and B B Meshram, Content Based Video Retrieval Systems, International

Journal of UbiComp (IJU), Vol.3, No.2, April 2012, 13-30.

[3] Y. Alp Aslandogan and Clement T. Yu, Techniques and Systems for Image and Video

Retrieval, IEEE Transactions on Knowledge and Data Engineering, VOL. 11, NO. 1,

January/February 1999, 56-62.

[4] Dr. Sudeep D. Thepade, Ajay A. Narvekar, Ameya V. Nawale, Color Content Based

Video Retrieval using Discrete Cosine Transform Applied on Rows and Columns of

Video Frames with RGB Color Space, International Journal of Engineering and

Innovative Technology Volume 2, Issue 11, May 2013, 133-135.

[5] Jun Yue, Zhenbo Li, Lu Liu, Zetian Fu, Content-based image retrieval using color and

texture fused features, Mathematical and Computer Modelling journal homepage

www.elsevier.com/locate/mcm 2011, 1121-1127.

[6] Man-Kwan Shan and Suh-Yin Lee, Content-based Video Retrieval based on

Similarity of Frame Sequence.



13

[7] Ch.Kavitha, Dr.B.Prabhakara Rao, Dr.A.Govardhan, Image Retrieval Based on Color

and Texture Features of the Image Sub-blocks, International Journal of Computer

Applications (0975 – 8887) Volume 15– No.7, February 2011, 33-37.

[8] Bouke Huurnink, Cees G. M. Snoek, Maarten de Rijke and Arnold W. M. Smeulders,

Content-Based Analysis Improves Audiovisual Archive Retrieval IEEE Transactions

on Multimedia, VOL. 14, NO. 4, AUGUST 2012, 1166-1178.

[9] Nevenka Dimitrova, Hong-Jiang Zhang, Behzad Shahraray, Ibrahim Sezan, Thomas

Huang, Avideh Zakhor,Applications of Video-Content Analysis and Retrieval, IEEE

2002, 42-55.

[10] Che-Yen Wen, Liang-Fan Chang, Hung-Hsin Li, Content based video retrieval with

motion vectors and the RGB color model, Received: October 31, 2007 / Accepted:

November 12, 2007.

[11] Y. Alp Aslandogan and Clement T. Yu, Techniques and Systems for Image and Video

Retrieval, IEEE Transactions on Knowledge and Data Engineering, VOL. 11, NO. 1,

January/February 1999, 56-62.

[12] Sagar Soman, Mitali Ghorpade, Vrushali Sonone, Satish Chavan Content Based

Image Retrieval Using Advanced Color and Texture Features, International

Conference in Computational Intelligence (ICCIA) 2011 Proceedings published in

International Journal of Computer Applications® (IJCA) 2011, 10-14.

[13] International Journal of Emerging Technology and Advanced Engineering Website:

www.ijetae.com (ISSN 2250-2459 (Online), An ISO 9001:2008 Certified Journal,

Volume 3, Special Issue 1, January 2013) International Conference on Information

Systems and Computing (ICISC-2013), INDIA.

[14] Marco Bertini, Alberto Del Bimbo, Andrea Ferracani, Lea Landucci, Daniele

Pezzatini Interactive multi-user video retrieval systems, (Springer Science+Business

Media, LLC) 2011, 111-137.

[15] Meenakshi A. Thalor and S. T. Patil, Comparison of Color Feature Extraction

Methods in Content based Video Retrieval, International Conference in Recent

Trends in Information Technology and Computer Science (ICRTITCS - 2012)

Proceedings published in International Journal of Computer Applications® (IJCA),

0975 – 8887, 10-14.

[16] Ranjit.M.Shende1, Dr P.N.Chatur, Dominant Color and Texture Approached for

Content Based Video Images Retrieval, IOSR Journal of Computer Engineering

(IOSR-JCE) e-ISSN: 2278-0661, p- ISSN: 2278-8727Volume 9, Issue 5, Mar. - Apr.

2013, 69-74.

[17] Chuan Wu*, Yuwen He, Li Zhao, Yuzhuo Zhong Motion Feature Extraction Scheme

for Content-based Video Retrieval, Proceedings of SPIE Vol. 4676 2002, 296-305.

[18] John R. Smith and Shih-Fu Chang, Tools and Techniques for Color Image Retrieval,

IS&T/SPIE PROCEEDINGS VOL. 2670 1-12.

[19] Tamizharasan.C1, Dr.S.Chandrakala, A Survey on Multimodal Content Based Video

Retrieval, International Conference on Information Systems and Computing

(ICISC-2013), INDIA, 69-76.

[20] Marco Bertini, Alberto Del Bimbo Andrea Ferracani, Lea Landucci·, Daniele

Pezzatini, Interactive multi-user video retrieval systems Multimedia Tools Appl,

(2013 62:111–137), (Springer DOI 10.1007/s11042-011-0888-9), 111-137.

[21] Hamdy K. Elminir, Mohamed Abu ElSoud, Multi feature content based video

retrieval using high level semantic concept, IJCSI International Journal of Computer

Science Issues, Vol. 9, Issue 4, No 2, July 2012, 254-260.



14

[22] T.N.Shanmugam and Priya Rajendran, An Enhanced Content-Based Video Retrieval

System Based on Query Clip, International Journal of Research and Reviews in

Applied Sciences ISSN: 2076-734X, EISSN: 2076-7366 Volume 1, Issue 3,

December 2009, 236-253.

[23] Smita Chavan and Shubhangi Sapkal, Color Content based Video Retrieval,

International Journal of Computer Applications (0975 – 8887) Volume 84 – No 11,

December 2013, 15-18.

[24] Vilas Naik, Vishwanath Chikaraddi and Prasanna Patil, “Query Clip Genre

Recognition using Tree Pruning Technique for Video Retrieval”, International Journal

of Computer Engineering & Technology (IJCET), Volume 4, Issue 4, 2013,

pp. 257 - 266, ISSN Print: 0976 – 6367, ISSN Online: 0976 – 6375.

[25] Payel Saha and Sudhir Sawarkar, “A Content Based Multimedia Retrieval System”,

International Journal of Computer Engineering & Technology (IJCET), Volume 5,

Issue 2, 2014, pp. 19 - 29, ISSN Print: 0976 – 6367, ISSN Online: 0976 – 6375.

[26] K.Ganapathi Babu, A.Komali, V.Satish Kumar and A.S.K.Ratnam, “An Overview of

Content Based Image Retrieval Software Systems”, International Journal of Computer

Engineering & Technology (IJCET), Volume 3, Issue 2, 2012, pp. 424 - 432,

ISSN Print: 0976 – 6367, ISSN Online: 0976 – 6375.

[27] Tarun Dhar Diwan and Upasana Sinha, “Performance Analysis is Basis on Color

Based Image Retrieval Technique”, International Journal of Computer Engineering &

Technology (IJCET), Volume 4, Issue 1, 2013, pp. 131 - 140, ISSN Print:

0976 – 6367, ISSN Online: 0976 – 6375.

[28] Abhishek Choubey, Omprakash Firke and Bahgwan Swaroop Sharma, “Rotation and

Illumination Invariant Image Retrieval using Texture Features”, International journal

of Electronics and Communication Engineering & Technology (IJECET), Volume 3,

Issue 2, 2012, pp. 48 - 55, ISSN Print: 0976- 6464, ISSN Online: 0976 –6472.

[29] Vilas Naik and Sagar Savalagi, “Textual Query Based Sports Video Retrieval by

Embedded Text Recognition”, International Journal of Computer Engineering &

Technology (IJCET), Volume 4, Issue 4, 2013, pp. 556 - 565, ISSN Print:

0976 – 6367, ISSN Online: 0976 – 6375.

Documents

1 COLOR BASED VIDEO RETRIEVAL USING BLOCK …...2.1 Color based video retrieval using DCT The DCT block computes the unitary discrete cosine transform (DCT) of each channel in the