Journal Articles
Permanent URI for this collectionhttps://idr.nitk.ac.in/handle/123456789/19884
Browse
48 results
Search Results
Item Epileptic EEG detection using neural networks and post-classification(2008) Patnaik, L.M.; Manyam, O.K.Electroencephalogram (EEG) has established itself as an important means of identifying and analyzing epileptic seizure activity in humans. In most cases, identification of the epileptic EEG signal is done manually by skilled professionals, who are small in number. In this paper, we try to automate the detection process. We use wavelet transform for feature extraction and obtain statistical parameters from the decomposed wavelet co-efficients. A feed-forward backpropagating artificial neural network (ANN) is used for the classification. We use genetic algorithm for choosing the training set and also implement a post-classification stage using harmonic weights to increase the accuracy. Average specificity of 99.19%, sensitivity of 91.29% and selectivity of 91.14% are obtained. © 2008 Elsevier Ireland Ltd. All rights reserved.Item An efficient framework for segmentation and identification of tumours in brain MR images(Inderscience Enterprises Ltd., 2016) Parameshwari, D.S.; Aparna., P.In this research work, two efficient textural feature extraction (TFE) algorithms (TFEA-I and TFEA-II) are proposed for a class of brain magnetic resonance imaging (MRI) applications. TFEA-I employs higher order statistical cumulant, namely, Kurtosis in order to generate a feature set based on the probability density function (PDF) of generalised Gaussian model that represents thewavelet coefficient energies of the sub-bands of decomposed image. TFEA-II derives a feature set employing cooccurrence matrix model for second order statistical characterisation of wavelet coefficients. In conjunction with TFEA-I and TFEA-II, we propose segmentation framework to compute coarse and smooth segmented boundaries for the tumour. When compared with the conventional TFEA methods reported in the literature, the use of proposed TFEA-I and TFEA-II results in two important advantages; considerable reduction in the feature set size and elimination of the need for using specialised feature selection/reduction algorithms thereby making them highly attractive for a class of brain MR imaging application. © © 2016 Inderscience Enterprises Ltd.Item Condition monitoring of roller bearing by K-star classifier and K-nearest neighborhood classifier using sound signal(Tech Science Press sale@techscience.com, 2017) Sharma, R.K.; Sugumaran, V.; Kumar, H.; Amarnath, M.Most of the machineries in small or large scale industry have rotating element supported by bearings for rigid support and accurate movement. For proper functioning of machinery, condition monitoring of the bearing is very important. In present study sound signal is used to continuously monitor bearing health as sound signals of rotating machineries carry dynamic information of components. There are numerous studies in literature that are reporting superiority of vibration signal of bearing fault diagnosis. However, there are very few studies done using sound signal. The cost associated with condition monitoring using sound signal (Microphone) is less than the cost of transducer used to acquire vibration signal (Accelerometer). This paper employs sound signal for condition monitoring of roller bearing by K-star classifier and k-nearest neighborhood classifier. The statistical feature extraction is performed from acquired sound signals. Then two layer feature selection is done using J48 decision tree algorithm and random tree algorithm. These selected features were classified using K-star classifier and k-nearest neighborhood classifier and parametric optimization is performed to achieve the maximum classification accuracy. The classification results for both K-star classifier and k-nearest neighborhood classifier for condition monitoring of roller bearing using sound signals were compared. © Copyright 2017 Tech Science Press.Item A novel mathematical morphology based algorithm for shoreline extraction from satellite images(Taylor and Francis Ltd., 2017) Rishikeshan, C.A.; Ramesh, H.Shoreline extraction is fundamental and inevitable for several studies. Ascertaining the precise spatial location of the shoreline is crucial. Recently, the need for using remote sensing data to accomplish the complex task of automatic extraction of features, such as shoreline, has considerably increased. Automated feature extraction can drastically minimize the time and cost of data acquisition and database updating. Effective and fast approaches are essential to monitor coastline retreat and update shoreline maps. Here, we present a flexible mathematical morphology-driven approach for shoreline extraction algorithm from satellite imageries. The salient features of this work are the preservation of actual size and shape of the shorelines, run-time structuring element definition, semi-automation, faster processing, and single band adaptability. The proposed approach is tested with various sensor-driven images with low to high resolutions. Accuracy of the developed methodology has been assessed with manually prepared ground truths of the study area and compared with an existing shoreline classification approach. The proposed approach is found successful in shoreline extraction from the wide variety of satellite images based on the results drawn from visual and quantitative assessments. © 2017 Wuhan University. Published by Informa UK Limited, trading as Taylor & Francis Group.Item Automated multi-level malware detection system based on reconstructed semantic view of executables using machine learning techniques at VMM(Elsevier B.V., 2018) M.a, A.K.; Jaidhar, C.D.In order to fulfill the requirements like stringent timing restraints and demand on resources, Cyber–Physical System (CPS) must deploy on the virtualized environment such as cloud computing. To protect Virtual Machines (VMs) in which CPSs are functioning against malware-based attacks, malware detection and mitigation technique is emerging as a highly crucial concern. The traditional VM-based anti-malware software themselves a potential target for malware-based attack since they are easily subverted by sophisticated malware. Thus, a reliable and robust malware monitoring and detection systems are needed to detect and mitigate rapidly the malware based cyber-attacks in real time particularly for virtualized environment. The Virtual Machine Introspection (VMI) has emerged as a fine-grained out-of-VM security solution to detect malware by introspecting and reconstructing the volatile memory state of the live guest Operating System (OS) by functioning at the Virtual Machine Monitor (VMM) or hypervisor. However, the reconstructed semantic details by the VMI are available in a combination of benign and malicious states at the hypervisor. In order to distinguish between these two states, extensive manual analysis is required by the existing out-of-VM security solutions. To address the foremost issue, in this paper, we propose an advanced VMM-based guest-assisted Automated Multilevel Malware Detection System (AMMDS) that leverages both VMI and Memory Forensic Analysis (MFA) techniques to predict early symptoms of malware execution by detecting stealthy hidden processes on a live guest OS. More specifically, the AMMDS system detects and classifies the actual running malicious executables from the semantically reconstructed process view of the guest OS. The two sub-components of the AMMDS are: Online Malware Detector (OMD) and Offline Malware Classifier (OFMC). The OMD recognizes whether the running processes are benign or malicious using its Local Malware Signature Database (LMSD) and online malware scanner and the OFMC classify unknown malware by adopting machine learning techniques at the hypervisor. The AMMDS has been evaluated by executing large real-world malware and benign executables on to the live guest OSs. The evaluation results achieved 100% of accuracy and zero False Positive Rate (FPR) on the 10-fold cross-validation in classifying unknown malware with maximum performance overhead of 5.8%. © 2017 Elsevier B.V.Item A novel pre-processing procedure for enhanced feature extraction and characterization of electromyogram signals(Elsevier Ltd, 2018) Powar, O.S.; Chemmangat, K.; Figarado, S.In the analysis of electromyogram signals, the challenge lies in the suppression of noise associated with the measurement and signal conditioning. The main aim of this paper is to present a novel pre-processing step, namely Minimum Entropy Deconvolution Adjusted (MEDA), to enhance the signal for feature extraction resulting in better characterization of different upper limb motions. MEDA method is based on finding the set of filter coefficients that recover the output signal with maximum value of kurtosis while minimizing the low kurtosis noise components. The proposed method has been validated on surface electromyogram dataset collected from seven subjects performing eight classes of hand movements (wrist flexion, wrist radial deviation, hand close, tripod, wrist extension, wrist ulnar deviation, cylindrical and key grip) with only two pairs of electrodes recorded from flexor carpi radialis and extensor carpi radialis on the forearm. The performance of the MEDA has been compared across four classifiers namely J-48, k-nearest neighbours (KNN), Naives Bayes and Linear Discriminant Analysis (LDA) attaining the classification accuracy of 85.3 ± 4%, 85.67 ± 5%, 76 ± 6% and 91.2 ± 2% respectively. Practical results demonstrate the feasibility of the approach with mean percentage increase in classification accuracy of 20.5%, without significant increase in computational time across seven subjects demonstrating the significance of MEDA in classification. © 2018 Elsevier LtdItem An ANN supported mathematical morphology based algorithm for lakes extraction from satellite images(Taylor and Francis Ltd. michael.wagreich@univie.ac.at, 2018) Rishikeshan, C.A.; Ramesh, H.With advances in remote sensing (RS) technology and platforms, more and more high-quality and fine spatial resolution satellite images are available. Manual method of feature extraction from remote sensing imagery is a tedious and time-consuming process. Thus automated and replicable technique plays vital role in updating lake database to evaluate the spatial and temporal evolution of lakes and ponds especially for vastly growing urban areas. This research work presents an artificial neural network (ANN) computed threshold value-based mathematical morphology (MM)-driven approach for extraction of lakes from satellite imageries with better accuracy. Accuracy of developed methodology has been assessed with the ground truths of the study area revealing better performance with different data-sets compared to existing methods. On an average scale for all data-sets used, the proposed algorithm is able to extract lakes with 99.47% accuracy and 0.9397 correlation coefficient (MCC). The existing classification method exhibited an accuracy of 98.75% and correlation coefficient of 0.89049. Similarly, the existing threshold-driven method has 99.31% accuracy and 0.90374 correlation coefficient. Maintenance of actual size and shape of the lakes, run-time control over structuring elements, semi-automation, faster processing, and single band adaptability are features of this work. © 2017 Indian Society for Hydraulics.Item Dynamic video anomaly detection and localization using sparse denoising autoencoders(Springer New York LLC barbara.b.bertram@gsk.com, 2018) Narasimhan, M.G.; Kamath S?, S.The emergence of novel techniques for automatic anomaly detection in surveillance videos has significantly reduced the burden of manual processing of large, continuous video streams. However, existing anomaly detection systems suffer from a high false-positive rate and also, are not real-time, which makes them practically redundant. Furthermore, their predefined feature selection techniques limit their application to specific cases. To overcome these shortcomings, a dynamic anomaly detection and localization system is proposed, which uses deep learning to automatically learn relevant features. In this technique, each video is represented as a group of cubic patches for identifying local and global anomalies. A unique sparse denoising autoencoder architecture is used, that significantly reduced the computation time and the number of false positives in frame-level anomaly detection by more than 2.5%. Experimental analysis on two benchmark data sets - UMN dataset and UCSD Pedestrian dataset, show that our algorithm outperforms the state-of-the-art models in terms of false positive rate, while also showing a significant reduction in computation time. © 2017, Springer Science+Business Media, LLC.Item A novel feature extraction technique for pulmonary sound analysis based on EMD(Elsevier Ireland Ltd, 2018) Mondal, A.; Banerjee, P.; Tang, H.Background and objective: The stethoscope based auscultation technique is a primary diagnostic tool for chest sound analysis. However, the performance of this method is limited due to its dependency on physicians experience, knowledge and also clarity of the signal. To overcome this problem we need an automated computer-aided diagnostic system that will be competent in noisy environment. In this paper, a novel feature extraction technique is introduced for discriminating various pulmonary dysfunctions in an automated way based on pattern recognition algorithms. Method: In this work, the disease correlated relevant characteristics of lung sounds signals are identified in terms of statistical distribution parameters: mean, variance, skewness, and kurtosis. These features are extracted from selective morphological components of the mapped signal in the empirical mode decomposition domain. The feature set is fed to the classifier model to differentiate their corresponding classes. Results: The significance of features developed are validated by conducting several experiments using supervised and unsupervised classifiers. Furthermore, the discriminating power of the proposed features is compared with three types of baseline features. The experimental result is evaluated by statistical analysis and also validated with physicians inference. Conclusions: It is found that the proposed features extraction technique is superior to the baseline methods in terms of classification accuracy, sensitivity and specificity. The developed method gives better results compared to baseline methods in any circumstance. The proposed method gives a higher accuracy of 94.16, sensitivity of 100 and specificity of 93.75 for an artificial neural network classifier. © 2018 Elsevier B.V.Item Automatic seizure detection by modified line length and Mahalanobis distance function(Elsevier Ltd, 2018) Pathak, A.; Ramesh, A.; Mitra, A.; Majumdar, K.Automatic seizure detection with high accuracy and in linear time has profound implications on therapeutic intervention mechanisms. In this work taking into account 12 popular seizure detection algorithms we have shown that line length is one feature that is extractable in linear time from EEG signals and capable of automatic seizure onset detection with highest accuracy among linear time extractable features. Also line length is less prone to give false positives. The detection accuracy has been ascertained by ROC curve analysis on Freiburg Seizure Prediction Project data containing intracranial EEG recordings of 87 seizures from 21 patients with sufficient interictal signals. Next, we have modified the classical line length feature extraction algorithm to improve its accuracy without any additional computational burden. Finally, we have applied both classical line length (LL) and modified line length (MLL) on all focal channels and detected seizures on multidimensional focal channel signals by Mahalanobis distance function (MDF). Both detected 73 out of 87 seizures. Area under the ROC curve (AUC), detection delay and false positive for LL and MLL are 0.951, 11.903 s, 0.201/h and 0.954, 11.698 s, 0.198/h respectively. Since LL has already been incorporated into an FDA approved commercially available closed loop intervention system, even this minute improvement may have significant healthcare implications. © 2018 Elsevier Ltd