Content Based Image Retrieval by K-Means Clustering Algorithm

2015

In recent years, there is an increasing research focus on Bag-of-Words based visual search. The state-ofthe-art in visual search systems are built based upon a visual vocabulary model with an inverted indexing structure. This search paradigm is achieved by systems that are inspired by text retrieval. Visual words are image descriptors selected from a discrete vocabulary. These words are mapped from highdimensional descriptors characterizing local regions of images. In this paper, we address problems of visual vocabulary constructing by comparing k-means with kmeans++ algorithms. As expected, k-means++ is clearly superior to the classical algorithm k-means both in terms of quality and running time.

International Journal of …, 2010

Published by: PIONEER RESEARCH & DEVELOPMENT GROUP (www.prdg.org) 1 Abstract— Content-based image retrieval (CBIR) is a new but widely adopted method for finding images from vast and annotated image databases. As the network and development of multimedia technologies are becoming more popular, users are not satisfied with the traditional information retrieval techniques. So nowadays the content based image retrieval (CBIR) are becoming a source of exact and fast retrieval. In recent years, a variety of techniques have been developed to improve the performance of CBIR. An image retrieval system that takes the input query image and retrieves the similar images according to the spatial coordinates and which uses the k means clustering algorithm for its segmentation. Most existing Content Based Image Retrieval based on the images of color, text documents, informative charts, and shape.

There are many methods to retrieve an image from an amassment of images in the database in order to meet users demand with image content kindred attribute, edge pattern homogeneous attribute, color homogeneous attribute, etc. An image retrieval system offers an efficient way to access or retrieve set of similar images by directly computing the image features from images by directly computing the image features from an image as reported by utilizing different kinds of techniques as well as algorithm. Content based image retrieval (CBIR) is most recently used technique for image retrieval from large image database. The reason behind content based image retrieval is to get perfect and fast result. There are many technique of CBIR utilized for image retrieval. A Block Truncation Coding technique is the famous method used for image retrieval. In the proposed system the advanced technique of BTC is used that is Ordered Dither Block Truncation Coding (ODBTC). ODBTC encoded data stream to construct the image features namely Color Co-occurrence and Bit Pattern features. After the extraction of this feature similarity distance is computed for retrieving a set of similar images. And to make the search more accurate K-means clustering method is used. The most similar images to the query image are selected and these features are being appended together and k-means clustering is applied. This method retrieves more similar images to the query image than the first search. The proposed scheme can be considered as very good in color image retrieval application. The process is implemented in a MATLAB 2014.

2013

This paper provides an overview of the technical achievements in the research area of k-means clustering in content-based image retrieval (CBIR). K-means clustering is a powerful technique in CBIR systems, in order to improve the performance of CBIR effectively. It is an open research area to the researcher to reduce the local maximum trapping and number of iteration to find target image. The paper covers the current state of art of the research in k-means clustering in CBIR and query point method for the system.

csjournals.com

Content-based image retrieval (CBIR) systems are capable to use query for visually related images by identifying similarity between a query image and those in the image database. The CBIR systems can be classified broadly into two classes as Low level feature based system and High level Semantic feature based system. Image contents are plays significant role for image retrieval. The most common contents are color, texture and shape. An efficient image retrieval system must be based on well-organized image feature extraction. K-means clustering is used to group similar and dissimilar objects in an image database into k disjoint clusters whereas neural network is used as a retrieval engine to measure the overall similarity between the query and the images. Relevance feedback is a query modification technique in the field of content-based image retrieval to improve the retrieval performance.

The field of image retrieval has been an active research area for several decades and has gained steady momentum in recent years as a result, large collection of digital images are growing day by day in government, hospitals, banking, etc. Interest in image retrieval has increased in large part due the rapid growth of the World Wide Web. With the proliferation of image data, the need to search and retrieve images efficiently and accurately from a large collection of image databases has drastically increased. Query based text retrieval from database is done through tools like SQL, MYSQL etc. In this process we consider constraint based query to retrieve text records from database. But image retrieval is not as easy as query processing. For this proposed system a broad literature survey of data mining technique have been made. Clustering technique has been identified as best suited for this system. In clustering technique there exist algorithms like BIRCH, CURE, DBSCAN, STING, K-Means, K-Medoids etc, out of which K-Means have been chosen as best suited algorithm for the proposed system. In Image retrieval Shape and Color of an object plays an important role. To address such a demand, Content-Based Image Retrieval (CBIR) through Clustering is proposed in this system for implementing CBIR system and identifying cluster of images for medical applications. With this process, the knowledge discovery through pattern recognition is carried out, through which the medical researchers can take predictive measures.

International Journal of Electrical, Electronics and Computers, 2017

Nowadays, Content-Based Image Retrieval has received a massive attention in the literature of image information retrieval, and accordingly a broad range of techniques have been proposed. However, these techniques are not free of defects in terms of recognition. In this paper, content based image retrieval has been proposed with a new method of building feature vector to represente an image for the clustertnig, which consiss of 140 elements taken from several feature types as following color historgram, color moments, Gabor filters, GLCM matrix, wavelet transformation, tamura feature, and moment invaraints. Aftering preparing the feature vector, clustering operation named K-Mean is exploited here to give the centroid of each image features. Finally Minkowski-Form Distance and Euclidean distance as a similarity measurement are applied for clustering groups of images having the same charactersitcs, shape and colors. The experiment is run on IMPLIcity database which has 1000 colored images. The evaluation of this proposed algorithm was by selecting random five images as query images, a fruitful result has been gotten as clustering set of images as illustared in the result section of this paper.

International Journal of Computer Applications, 2012

This paper presents novel techniques for image retrieval using the clustering features extracted from images based on Row Mean Clustering, Column Mean clustering, Row Mean DCT Clustering ,Column Mean DCT Clustering, Row Mean Wavelet Clustering and Column Mean Wavelet Clustering. The proposed techniques are compared with well known traditional technique such as Hierarchical Clustering. Hierarchical clustering starts by calculating the Euclidean distance measure for all patterns in data set, which is not required to calculate in proposed techniques. Hence number of clusters used for comparison of proposed techniques is less as compared to existing technique (Hierarchical Clustering). All the CBIR techniques are implemented on a database having 665 images spread across 31 classes. The results of proposed techniques have shown performance improvement (higher precision and Recall) as compared to existing technique at reduced computations.

The content based image retrieval (CBIR) is the well-liked and heart favorite area of research in the field of digital image processing. The key goal of content based image retrieval (CBIR) is to excerpt the visual content of an image directly, like color, texture, or shape. There are several applications of the CBIR technique such as forensic laboratories, crime detection, image searching etc. For the purpose of feature extraction of well-matched images from the database, a universal CBIR system utilizes texture, color and shape based techniques. In this presented work, we have offered an efficient approach for the content based image retrieval, where images are decomposed using the wavelet transform, it means that the image features are converted in the matrix form and a color feature data set is prepared. In this paper, we are proposing the algorithm with the help of that we can improve the image retrieval.