An Image Clustering and Feedback-based Retrieval Framework

2010 ◽  
Vol 1 (1) ◽  
pp. 55-74 ◽  
Author(s):  
Chengcui Zhang ◽  
Liping Zhou ◽  
Wen Wan ◽  
Jeffrey Birch ◽  
Wei-Bang Chen
Keyword(s):  
Image Retrieval ◽  
Clustering Algorithm ◽  
Image Database ◽  
Image Clustering ◽  
Single Object ◽  
Image Region ◽  
Retrieval Accuracy ◽  
Retrieval Systems ◽  
Object Based ◽  
Region Matching

Most existing object-based image retrieval systems are based on single object matching, with its main limitation being that one individual image region (object) can hardly represent the user’s retrieval target, especially when more than one object of interest is involved in the retrieval. Integrated Region Matching (IRM) has been used to improve the retrieval accuracy by evaluating the overall similarity between images and incorporating the properties of all the regions in the images. However, IRM does not take the user’s preferred regions into account and has undesirable time complexity. In this article, we present a Feedback-based Image Clustering and Retrieval Framework (FIRM) using a novel image clustering algorithm and integrating it with Integrated Region Matching (IRM) and Relevance Feedback (RF). The performance of the system is evaluated on a large image database, demonstrating the effectiveness of our framework in catching users’ retrieval interests in object-based image retrieval.

An Image Clustering and Feedback-Based Retrieval Framework

2012 ◽  
pp. 62-80
Author(s):  
Chengcui Zhang ◽  
Liping Zhou ◽  
Wen Wan ◽  
Jeffrey Birch ◽  
Wei-Bang Chen
Keyword(s):  
Image Retrieval ◽  
Relevance Feedback ◽  
Time Complexity ◽  
Clustering Algorithm ◽  
Image Database ◽  
Image Clustering ◽  
Image Region ◽  
Retrieval Systems ◽  
Object Based ◽  
Region Matching

Most existing object-based image retrieval systems are based on single object matching, with its main limitation being that one individual image region (object) can hardly represent the user’s retrieval target, especially when more than one object of interest is involved in the retrieval. Integrated Region Matching (IRM) has been used to improve the retrieval accuracy by evaluating the overall similarity between images and incorporating the properties of all the regions in the images. However, IRM does not take the user’s preferred regions into account and has undesirable time complexity. In this article, we present a Feedback-based Image Clustering and Retrieval Framework (FIRM) using a novel image clustering algorithm and integrating it with Integrated Region Matching (IRM) and Relevance Feedback (RF). The performance of the system is evaluated on a large image database, demonstrating the effectiveness of our framework in catching users’ retrieval interests in object-based image retrieval.

Interface for visualization of image database in adaptive image retrieval systems (AIRS)

2005 ◽  
Author(s):  
Anca Doloc-Mihu ◽  
Vijay V. Raghavan ◽  
Surendra Karnatapu ◽  
Chee-Hung H. Chu
Keyword(s):  
Image Retrieval ◽  
Image Database ◽  
Retrieval Systems

WAVELET-BASED MULTIRESOLUTION HISTOGRAM FOR FAST IMAGE RETRIEVAL

2004 ◽  
Vol 02 (01) ◽  
pp. 59-73 ◽  
Author(s):  
PAWAN JAIN ◽  
S. N. MERCHANT
Keyword(s):  
Data Structure ◽  
Image Retrieval ◽  
Retrieval System ◽  
Performance Enhancement ◽  
Image Database ◽  
Retrieval Systems ◽  
Candidate Image ◽  
Image Retrieval System ◽  
Resolution Level ◽  
Structure Algorithm

Most of the content-based image retrieval systems require a distance computation of feature vectors for each candidate image in the image database. This exhaustive search is highly time-consuming and inefficient. This limits the usefulness of such system. Thus there is a growing need for a fast image retrieval system. Multiresolution data-structure algorithm provides a good solution to the above problem. In this paper we propose a wavelet-based multiresolution data-structure algorithm. Wavelet-based multiresolution data-structure further reduce the number of computation by around 50%. In the proposed approach we reuse the information obtained at lower resolution levels to calculate the distance at a higher resolution level. Apart from this, the proposed structure saves memory overheads by about 50% over multiresolution data-structure algorithm. The proposed algorithm can be easily combined with other algorithms for performance enhancement.4 In this paper we use the proposed technique to match luminance histogram for image retrieval. Fuzzy histograms enhances performance by considering the similarity between neighboring bins. We have extended the proposed approach to fuzzy histograms for better performance.

FUZZY ART-BASED IMAGE CLUSTERING METHOD FOR CONTENT-BASED IMAGE RETRIEVAL

2007 ◽  
Vol 06 (02) ◽  
pp. 213-233 ◽  
Author(s):  
SANG-SUNG PARK ◽  
KWANG-KYU SEO ◽  
DONG-SIK JANG
Keyword(s):  
Image Retrieval ◽  
Content Based Image Retrieval ◽  
Image Clustering ◽  
Clustering Method ◽  
Image Region ◽  
Shape Information ◽  
Fuzzy Art ◽  
The Neural Network ◽  
Joint Histogram ◽  
Noise Input

In this paper, an image clustering method that is essential for content-based image retrieval in large image databases efficiently is proposed by color, texture, and shape contents. The dominant triple HSV (Hue, Saturation, and Value), which are extracted from quantized HSV joint histogram in the image region, are used for representing color information in the image. Entropy and maximum entry from co-occurrence matrices are used for texture information and edge angle histogram is used for representing shape information. Due to its algorithmic simplicity and the several merits that facilitate the implementation of the neural network, Fuzzy ART has been exploited for image clustering. Original Fuzzy ART suffers unnecessary increase of the number of output neurons when the noise input is presented. Therefore, the improved Fuzzy ART algorithm is proposed to resolve the problem by differently updating the committed node and uncommitted node, and checking the vigilance test again. To show the validity of the proposed algorithm, experimental results on image clustering performance and comparison with original Fuzzy ART are presented in terms of recall rates.

Application of Optimized Partitioning Around Medoid Algorithm in Image Retrieval

2021 ◽  
Vol 12 (1) ◽  
pp. 77-94
Author(s):  
Yanxia Jin ◽  
Xin Zhang ◽  
Yao Jia
Keyword(s):  
Image Retrieval ◽  
Clustering Algorithm ◽  
Particle Swarm ◽  
Image Data ◽  
Experimental Comparison ◽  
Cluster Center ◽  
Particle Swarm Algorithm ◽  
Query Image ◽  
Retrieval Accuracy ◽  
Swarm Algorithm

In image retrieval, the major challenge is that the number of images in the gallery is large and irregular, which results in low retrieval accuracy. This paper analyzes the disadvantages of the PAM (partitioning around medoid) clustering algorithm in image data classification and the excessive consumption of time in the computation process of searching clustering representative objects using the PAM clustering algorithm. Fireworks particle swarm algorithm is utilized in the optimization process. PF-PAM algorithm, which is an improved PAM algorithm, is proposed and applied in image retrieval. First, extract the feature vector of the image in the gallery for the first clustering. Next, according to the clustering results, the most optimal cluster center is searched through the firework particle swarm algorithm to obtain the final clustering result. Finally, according to the incoming query image, determine the related image category and get similar images. The experimental comparison with other approaches shows that this method can effectively improve retrieval accuracy.

scholarly journals Color image retrieval using taken images

2017 ◽  
Vol 7 (1) ◽  
Author(s):  
S. Selvam
Keyword(s):  
Image Retrieval ◽  
Color Image ◽  
Visual Features ◽  
Image Region ◽  
Retrieval Accuracy ◽  
Semantic Classes ◽  
Image Retrieval System ◽  
Speed Up ◽  
Almost All ◽  
The Given

Now-a-days in many applications content based image retrieval from large resources has become an area of wide interest. In this paper we present a color-based image retrieval system that uses color and texture as visual features to describe the content of an image region. To speed up retrieval and similarity computation, the database images are segmented and the extracted regions are clustered according to their feature vectors. This process is performed offline before query processing, therefore to answer a query our system need not search the entire database images; instead just a number of candidate images are required to be searched for image similarity. Our proposed system has the advantage of increasing the retrieval accuracy and decreasing the retrieval time. The experimental evaluation of the system is based on a 1,000 real taken color image database. From the experimental results, it is evident that our system performs significantly better and faster compared with other existing systems. In our analysis, we provide a comparison between retrieval results based on relevancy for the given ten classes. The results demonstrate that each type of feature is effective for a particular type of images according to its semantic contents, and using a combination of them gives better retrieval results for almost all semantic classes.

scholarly journals Image Retrieval Based on WBCH and Clustering Algorithm

2013 ◽  
Vol 5 (3) ◽  
pp. 604-613
Author(s):  
Asmita Bhaskar Shirsath ◽  
M. J. Chouhan ◽  
N. J Uke
Keyword(s):  
Image Retrieval ◽  
Large Scale ◽  
Clustering Algorithm ◽  
Similarity Measures ◽  
Image Database ◽  
Content Based Image Retrieval ◽  
Shape Information ◽  
Image Descriptors ◽  
Image Retrieval System ◽  
Retrieval Result

Research on content-based image retrieval has gained tremendous momentum during the last decade. Color, texture and shape information have been the primitive image descriptors in content based image retrieval systems. In order to get faster  retrieval result from large-scale image database ,we proposed image retrieval system in which image database is first pre-processed by Wavelet Based Color Histogram (WBCH) and K-means algorithm and then using Hierarchical clustering algorithm we index the previous result and then by using similarity measures we retrieve the images from pre-processed database. Experiments show that this proposed method offers substantial increase in retrieval speed but needs to be improved on retrieval results.

Image Retrieval via CURE Clustering and SIFT Algorithms

2013 ◽  
Vol 303-306 ◽  
pp. 1573-1576 ◽  
Author(s):  
Xian Tan
Keyword(s):  
Image Retrieval ◽  
Clustering Algorithm ◽  
Text Clustering ◽  
Image Clustering ◽  
Original Text ◽  
Retrieval Performance ◽  
Fusion Model ◽  
Semantic Clustering ◽  
Model Image ◽  
High Level

This paper, by using the short CURE clustering algorithm and image SIFT identification method, the establishment of a kind of image semantic clustering fusion model (image text clustering fusion model, referred to as ITCFM). The model is based on component method, the original image components, original text member, image clustering member, text clustering components, clustering fusion member five parts. In ITCM model for image semantic clustering characteristics on the basis of the description and extraction. The experimental results show that ITCM model can effectively to image to describe the high-level semantic, the image retrieval effect is good, and have stable retrieval performance.

A Hierarchial Classification Technique for Semantics-Based Image Retrieval

2011 ◽  
pp. 311-333 ◽  
Author(s):  
Mohammed Lamine Kherfi ◽  
Djemel Ziou
Keyword(s):  
Image Retrieval ◽  
Retrieval Accuracy ◽  
The Third ◽  
Retrieval Systems ◽  
Probabilistic Classification ◽  
Semantic Concepts ◽  
Similar Images ◽  
Hierarchical Index ◽  
Upper Level ◽  
Abstraction Levels

We present a new approach for improving image retrieval accuracy by integrating semantic concepts. First, images are represented according to different abstraction levels. At the lowest level, they are represented with visual features. At the upper level, they are represented with a set of very specific keywords. At the subsequent levels, they are represented with more general keywords. Second, visual content together with keywords are used to create a hierarchical index. A probabilistic classification approach is proposed, which allows to group similar images into the same class. Finally this index is exploited to define three retrieval mechanisms: the first is text-based, the second is content-based, and the third is a combination of both. Experiments show that our combination allows to nicely narrow the semantic gap encountered by most current image retrieval systems. Furthermore, we show that the proposed method helps reducing retrieval time and improving retrieval accuracy.

Sign in / Sign up

Export Citation Format

Share Document