A Modified Framework of a Clustering Algorithm for Image Processing Applications (original) (raw)

Two Way Clustering Based on Minimum Spanning Tree and Dbscan Algorithm for Image Mining

2017

Image mining is currently a growing yet active research focus in computer science. Image mining is connected with the development of information mining inside the field of image processing. Image mining handles with the concealed data extraction and additional examples that are not obviously characterized inside the pictures. Image mining incorporates systems like Image preparing, information handling, Robotics and machine learning. Semantic maps are used to visualize the image information which is stored in image databases. But to build the semantic maps we propose one graph optimization technique that is spanning tree techniques. After the development of semantic maps, data mining techniques are used to extract the image information. In this paper we propose a novel algorithm, Two Way Clustering based on Minimum Spanning Tree and DBSCAN (density-based spatial clustering of applications with noise) (TWCMSTDBSCAN) for Image Mining to segment the given image and to detect anomalous o...

Analysis of Different Clustering Algorithms on Image Databases

2011

When we apply Image Retrieval techniques to large image Databases .It provides restriction of search space to provide adequate response time. This restriction can be done minimized by using Clustering technique to partition the image dataset into subspaces of similar elements .In this article we will apply different clustering algorithms on large image database and then evaluate and analyse the performance of these algorithms to determine which algorithm is best for image retrieval.

Fast approximate minimum spanning tree based clustering algorithm

Neurocomputing, 2017

Minimum spanning tree (MST) based clustering algorithms have been employed successfully to detect clusters of heterogeneous nature. Given a dataset of n random points, most of the MST-based clustering algorithms first generate a complete graph G of the dataset and then construct MST from G. The first step of the algorithm is the major bottleneck which takes O(n 2) time. This paper proposes an algorithm namely MST-based Clustering on Partitionbased nearest neighbor graph for reducing the computational overhead. By using a centroid based nearest neighbor rule, the proposed algorithm first generates a sparse Local Neighborhood Graph (LNG) and then the approximate MST is constructed from LNG. We prove that both size and computational time to construct the graph (LNG) is O(n 3/2), which is a O(√ n) factor improvement over the traditional algorithms. The approximate MST is constructed from LNG in O(n 3/2 lg n) time, which is asymptotically faster than O(n 2). Experimental analysis on both synthetic and real datasets demonstrates that the computational time has been reduced significantly by maintaining the quality of clusters obtained from the approximate MST.

Graph-theoretic clustering for image grouping and retrieval

1999

Image retrieval algorithms are generally based on the assumption that visually similar images are located close to each other in the feature space. Since the feature vectors usually exist in a very high dimensional space, a parametric characterization of their distribution is impossible, so non-parametric approaches, like the k-nearest neighbor search, are used for retrieval.

A Novel Approaches on Clustering Algorithms And it's Applications

2012

Graph clustering algorithms are Random walk and minimum spanning tree algorithms. Random walk has been used to identify significant vertices in the graph that receive maximum flow while minimum spanning tree algorithm has been used to identify significant edges in the graph .We believe these two graph algorithms have useful applications in clustering, namely for identifying centroids and for identifying edges to merge or split clusters such that intra-cluster similarity is maximized while inter-cluster similarity is minimized. This paper investigates the graph algorithms, graph- based clustering algorithms, and their applications. graph algorithms and graph-based clustering algorithms, we propose novel variants of Star clustering algorithm that use different techniques for identifying centroids, and two novel graph-based clustering algorithms: MST-Sim and Ricochet. The variant graph algorithms and graph based clustering algorithms achieve higher performance in terms of effectiveness...

Minimization of Image Retrieval Time using Clustering

2014

Image Mining is a apecial technique used to study different aspects in detail of an image. In case of image mining there are certain algorithms with which different details of an image can be studied. Some of them are k-means, c-means, fuzzy approaches. Out of which we have adopted the technique of c-means clustering to achieve our target. In this algorithm we generally make different clusters on the basis of similarity and differences of images and form an image set. This approach depends on the membership function which clarifies the similarity and dissimilarity between the images. It depends on the mathematical value that specifies the membership function. Values of membership function is either near zero or one. Zero means similarity between images is very less and one means images are highly similar.

Partitioning Based Clustering Method for Image Retrieval

Clustering is a major technique used for grouping of numerical and image data in data mining and image processing applications. Clustering makes the job of image retrieval easy by finding the images as similar as given in the query image. The images are grouped together in some given number of clusters. Image data are grouped on the basis of some features such as color, texture, shape etc. contained in the images in the form of pixels. Content Based Image Retrieval (CBIR) is a collection of techniques for retrieving images from large database. The images are retrieved based on content. The term " content " demonstrate to color, texture and shapes. In this system, color and texture features are retrieved from images. The color features are obtained using Dynamic Color Distribution Entropy of Neighborhoods (D_CDEN). The texture features are obtained using Gray Level Co-occurrence Matrix (GLCM). The clustering technique is presented to solve the above problem. In this system, K-Means and Contribution based clustering techniques are used. The K-Means clustering algorithm optimizes only intra cluster similarity. Contribution based clustering enhance both intra and inter cluster similarity. The experimental results shows the comparison between average dispersion measures for both K-Means and Contribution based clustering.

Graph based k-means clustering

Signal Processing, 2012

An original approach to cluster multi-component data sets is proposed that includes an estimation of the number of clusters. Using Prim's algorithm to construct a minimal spanning tree (MST) we show that, under the assumption that the vertices are approximately distributed according to a spatial homogeneous Poisson process, the number of clusters can be accurately estimated by thresholding the sequence of edge lengths added to the MST by Prim's alorithm. This sequence, called the Prim trajectory, contains sufficient information to determine both the number of clusters and the approximate locations of the cluster centroids. The estimated number of clusters and cluster centroids are used to initialize the generalized Lloyd algorithm, also known as k-means, which circumvents its well known initialization problems. We evaluate the false positive rate of our cluster detection algorithm, using Poisson approximations in Euclidean spaces. Applications of this method in the multi/hyper-spectral imagery domain to a satellite view of Paris and to an image of Mars are also presented.

A fast hybrid clustering technique based on local nearest neighbor using minimum spanning tree

Expert Systems with Applications, 2019

With rapid explosion of information, clustering emerged as an active research area for knowledge discovery. Most of the existing clustering algorithms become ineffective when inappropriate parameters are provided or applied on a dataset which consists of clusters of diverse shapes, sizes, and varying densities. To overcome these issues, many graph based hybrid clustering algorithms have been proposed but these algorithms first generate a complete graph of the dataset which takes O(N 2) time where N is the number of data points which limits their application on large datasets. This paper proposes an algorithm namely a fast hybrid clustering technique based on local nearest neighbor using minimum spanning tree to reduce the computational overhead. In the first step, the algorithm partitions the dataset into large number of sub-clusters based on dispersion of data points to capture the geometry of clusters. After partitioning the dataset, a minimum spanning tree based on the centroids of each of the sub-clusters is constructed to identify the adjacent pairs. A novel merge method is proposed to find the genuine clusters by repeatedly merging the adjacent sub-clusters. The cohesion and intra-similarity are introduced to compute the level of dispersion of data points with respect to the centers of an adjacent pair and average edge weight of a sub-clusters respectively. The algorithm takes O(N 3/2) time which is a √ N factor improvement over the popular hybrid clustering algorithms. Experimental analyses on both synthetic as well as gene expression datasets demonstrate that the proposed technique shows significant improvement over competing clustering algorithms in terms of execution time and improved cluster quality. Moreover, the proposed algorithm does not require any user defined parameters and it can estimate the number of clusters more accurately.