scispace - formally typeset
Search or ask a question
Author

K. Rajakumar

Other affiliations: College of Engineering, Guindy
Bio: K. Rajakumar is an academic researcher from VIT University. The author has contributed to research in topics: Visual Word & Image texture. The author has an hindex of 5, co-authored 12 publications receiving 63 citations. Previous affiliations of K. Rajakumar include College of Engineering, Guindy.

Papers
More filters
Proceedings ArticleDOI
30 Mar 2019
TL;DR: This paper has proposed a novel method in which microscopic blood images are taken as an input image and the features are extracted and were used for classification using Convolutional Neural Network (CNN).
Abstract: Detection of White Blood Cell (WBC) cancer diseases like Acute Myeloid Leukemia (AML), Acute Lymphoblastic Leukemia (ALL), and Myeloma is a complex task in medical field because they are sudden in onset. Our proposed method consists of designing and developing an automated system which will assist the medical professionals in correctly diagnosing all the types and sub-types of this disease. In this paper, we have proposed a novel method in which we have taken microscopic blood images as an input image. A dataset of 100 images in which 62 training and 38 testing images is taken. After that we have converted the image to proper format (YCbCr) for segmentation. For segmenting, we have used the combination of Gaussian Distribution, Otsu Adaptive Thresholding and for clustering we have used K-Means method. Using Gray Level Co-occurrence Matrix (GLCM), the features are extracted and were used for classification using Convolutional Neural Network (CNN). The overall accuracy of the system obtained after processing is 97.3%.

20 citations

Proceedings ArticleDOI
29 Jul 2010
TL;DR: The experimental results show that the proposed algorithm is easy to identify main objects and reduce the influence of background in the image, and thus improve the performance of image retrieval.
Abstract: In content based image retrieval system, the reliability of retrieval results depends much on the image features used for measuring image similarity. In this paper, a new medical image retrieval method using energy efficient wavelet transform is proposed. Wavelet transformation can also be easily extended to 2-D (image) or 3-D (volume) data by successively applying 1-D transformation on different dimensions. The proposed algorithm was tested using energy efficient wavelet transform and performance analysis was done with Gabor, Bi-orthogonal, and Haar. The retrieval image is the relevance between a query image and any database image, the relevance similarity is ranked according to the closest similar measures computed by the Euclidean distance. The experimental results show that the proposed algorithm is easy to identify main objects and reduce the influence of background in the image, and thus improve the performance of image retrieval.

12 citations

Journal ArticleDOI
TL;DR: The experimental results show that, using the proposed MDCT approach, it is easy to identify main objects and reduce the influence of background in the image, and thus improve the performance of medical image retrieval.

11 citations

Journal ArticleDOI
TL;DR: The experimental results show that the proposed algorithm is easily identifies target object and reduces the influence of background in the image and thus improves the performance of MRI image retrieval.
Abstract: In content based image retrieval (CBIR) system, the images are represented based upon its feature such as color, texture, shape, and spatial relationship etc. In this paper, we propose a MRI Image Retrieval using wavelet transform with mahalanobis distance measurement. Wavelet transformation can also be easily extended to 2-D (image) or 3-D (volume) data by successively applying 1-D transformation on different dimensions. The proposed algorithm has tested using wavelet transform and performance analysis have done with HH and H* elimination methods. The retrieval image is the relevance between a query image and any database image, the relevance similarity is ranked according to the closest similar measures computed by the mahalanobis distance measurement. An adaptive similarity synthesis approach based on a linear combination of individual feature level similarities are analyzed and presented in this paper. The feature weights are calculated by considering both the precision and recall rate of the top retrieved relevant images as predicted by our enhanced technique. Hence, to produce effective results the weights are dynamically updated for robust searching process. The experimental results show that the proposed algorithm is easily identifies target object and reduces the influence of background in the image and thus improves the performance of MRI image retrieval.

11 citations

01 Jan 2013
TL;DR: It is shown that the texture features are extracted by using curvelet transform and statistical similarity measure is done by using mahalanobis distance measurement, showing that the proposed method significantly gives better retrieval results.
Abstract: Content-based image retrieval (CBIR) is the most commonly used method for searching large-scale medical image databases. Images are generally retrieved on the basis of either low level features, such as colour,texture and shape.Most texture based image retrieval systems are still incapable of providing better retrieval result through high retrieval accuracy and less computational complexity. To tackle this problem, we propose a texture based medical image retrieval using curvelet transform with mahalanobis distance measurement. We show that the texture features are extracted by using curvelet transform and statistical similarity measure is done by using mahalanobis distance . The proposed method gives a better retrieval rate . Experimental results on a database of 200 medical images show that the proposed method significantly gives better retrieval results.

6 citations


Cited by
More filters
Journal ArticleDOI
TL;DR: An inclusive taxonomy for architectural, algorithmic and technologic aspects of fog computing is provided and base architectures for application, software, security, computing resource management and networking are presented and evaluated using a proposed maturity model.
Abstract: Fog computing is an emerging technology to address computing and networking bottlenecks in large scale deployment of IoT applications. It is a promising complementary computing paradigm to cloud computing where computational, networking, storage and acceleration elements are deployed at the edge and network layers in a multi-tier, distributed and possibly cooperative manner. These elements may be virtualized computing functions placed at edge devices or network elements on demand, realizing the “computing everywhere” concept. To put the current research in perspective, this paper provides an inclusive taxonomy for architectural, algorithmic and technologic aspects of fog computing. The computing paradigms and their architectural distinctions, including cloud, edge, mobile edge and fog computing are subsequently reviewed. Practical deployment of fog computing includes a number of different aspects such as system design, application design, software implementation, security, computing resource management and networking. A comprehensive survey of all these aspects from the architectural point of view is covered. Current reference architectures and major application-specific architectures describing their salient features and distinctions in the context of fog computing are explored. Base architectures for application, software, security, computing resource management and networking are presented and are evaluated using a proposed maturity model.

96 citations

Journal Article
TL;DR: The simulation experiment with IRIS data set shows that the proposed algorithm converges faster and the value k found is close to the actual value, which proves the validity of the algorithm.
Abstract: Aiming at the problemsof too much iterative times in selecting initial centroids stochastically for K-Means algorithm,a method is proposed to optimize the initial centroids through cutting the set into k segmentations and select one point in each segmentation as initial centroids for iterative computing. A new valid function called clustering-index is defined as the sum of clustering-density and clustering-significance and can be used to search the optimization of k in the internal of [1,n(1/2) ]. The simulation experiment with IRIS data set shows that the proposed algorithm converges faster and the value k found is close to the actual value,which proves the validity of the algorithm.

88 citations

Journal ArticleDOI
TL;DR: The task offloading delay model is derived based on three different velocity models, i.e., a constant velocity model, vehicle-following model, and traveling-time statistical model and a pricing-based one-to-one matching algorithm is proposed and validated based on simulation results.
Abstract: Vehicular edge computing has emerged as a promising technology to accommodate the tremendous demand for data storage and computational resources in vehicular networks. By processing the massive workload tasks in the proximity of vehicles, the quality of service can be guaranteed. However, how to determine the task offloading strategy under various constraints of resource and delay is still an open issue. In this paper, we study the task offloading problem from a matching perspective and aim to optimize the total network delay. The task offloading delay model is derived based on three different velocity models, i.e., a constant velocity model, vehicle-following model, and traveling-time statistical model. Next, we propose a pricing-based one-to-one matching algorithm and pricing-based one-to-many matching algorithms for the task offloading. The proposed algorithm is validated based on three different simulation scenarios, i.e., straight road, the urban road with the traffic light, and crooked road, which are extracted from the realistic road topologies in Beijing and Guangdong, China. The simulation results confirm that significant delay decreasing can be achieved by the proposed algorithm.

54 citations

Proceedings Article
01 Nov 2017
TL;DR: In this article, the authors proposed two distributed information fusion (DIFFUSION) schemes, in which the information shared among sensors consists of: 1) contacts, generated by the local detection stage and 2) tracks, generated from the local tracking stage.
Abstract: Surveillance in antisubmarine warfare has traditionally been carried out by means of submarines or frigates with towed arrays. These techniques are manpower intensive. Alternative approaches have recently been suggested using distributed stationary and mobile sensors, such as autonomous underwater vehicles (AUVs). In contrast with the use of standard assets, these small, low-power, and mobile devices have limited processing and wireless communication capabilities. However, when deployed in a spatially separated network, these sensors can form an intelligent network achieving high performance with significant features of scalability, robustness, and reliability. The distributed information FUSION (DIFFUSION) strategy, in which the local information is shared among sensors, is one of the key aspects of this intelligent network. In this paper, we propose two DIFFUSION schemes, in which the information shared among sensors consists of: 1) contacts, generated by the local detection stage and 2) tracks, generated by the local tracking stage. In the first DIFFUSION scheme, contacts are combined at each nodes using the optimal Bayesian tracking based on the random finite set formulation. In the second DIFFUSION scheme, tracks are combined using the track-to-track association/fusion procedure, then a sequential decision based on the association events is exploited. A full validation of the DIFFUSION schemes is conducted by the NATO Science and Technology Organization—Center for maritime research and experimentation during the sea trials Exercise Proud Manta 2012–2013 using real data. Performance metrics of DIFFUSION and of local tracking/detection strategies are also evaluated in terms of time-on-target (ToT) and false alarm rate (FAR). We demonstrate the benefit of using DIFFUSION against the local noncooperative strategies. In particular DIFFUSION improves the level of TOT (FAR) with respect to the local tracking/detection strategies. In particular, the TOT is increased over 90%–95% while the FAR is reduced of two order of magnitude. The problem of communication failures, data not available from the collaborative AUV during certain periods of time, is also investigated. The robustness of DIFFUSION with respect to these communication failures is demonstrated, and the related performance results are reported here. In particular, with 75% of communication failures the ToT is over 90%–95% with a relatively small increase of the FAR with respect to the case of perfect communication.

52 citations

Proceedings ArticleDOI
01 Nov 2016
TL;DR: A review of some major work in area of image denoising is presented, which provides ample information about the human soft tissue, which helps in the diagnosis of human diseases.
Abstract: The advent of computer aided technologies image processing techniques have become increasingly important in a wide variety of medical applications. Intervention between the protection of useful diagnostic information and noise suppression must be treasured in medical images. Image denoising is an applicable issue found in diverse image processing and computer vision problems. There are various existing methods to denoise images. The important property of a good image denoising model is that it should completely remove noise as far as possible as well as preserve edges. This paper presents a review of some major work in area of image denoising. The objective in all discipline is to extract information about the scene being imaged. The rapid progress in computerized medical image reconstruction and the associated developments in analysis methods and computer-aided diagnosis has boosted medical imaging into one of the most important sub-fields in scientific imaging Ultrasound, MRI, CT-Scan, PET Scan are the medical techniques mainly used by the radiologist for visualization of internal structure of the human body without any surgery. These provide ample information about the human soft tissue, which helps in the diagnosis of human diseases.

35 citations