检索结果-内蒙古大学图书馆

International Conference on Intelligent Computing, Instrumentation and Control Technologies (ICICICT)

作者： G Priyanka R Gandhirai K P Soman Center for Computational Engineering and Networking (CEN) Amrita School of Engineering Amrita Vishwa Vidyapeetham Coimbatore Tamil Nadu India SIERS Research Lab Amrita School of Engineering Amrita Vishwa Vidyapeetham Coimbatore Tamil Nadu India

Target classification is an important application in modern radar systems. Recognition of targets in radar data is a key task for surveillance and monitoring the areas such as military. Automatic classification systems is becoming one of the challenges in modern era. The feature extractor is often hand-designed with domain knowledge in machine learning and can have a significant impact on classification accuracy. Convolutional neural network (CNN) have recently achieved best results through the automatic learning of feature hierarchies from massive training data in many applications such as computer vision and speech recognition tasks. We build a four layer convolutional neural network for the target classification task. The Dataset used for radar classification is MSTAR (Moving and Stationary Target Acquisition and Recognition) which consists of 8 class targets. The data collected from different angles like 15°, 17°, 30° and 45° from each class is considered. This paper shows that there is increase in the performance of target classification by building the convolutional neural network (CNN). We achieved 95% accuracy in the target classification without any trivial preprocessing and feature extraction mechanism. The Support vector machine (SVM) in the machine learning algorithm was able to achieve 86.9% classification performance with bag of words features. Furthermore our CNN result with any preprocessing steps shows that it can be used in real time applications like civilian and military purposes.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Image Denoising Using Total Variation Wavelet Galerkin Method

引用

Procedia Computer Science 2018年 143卷 481-492页

作者： Divya Pankaj K.A Narayanankutty D. Govind Center for Computational Engineering and Networking (CEN) Amrita School of Engineering Coimbatore Amrita Vishwa Vidyapeetham India

Wavelet Galerkin method (WGM) is an efficient numerical method for solving boundary value problems. In this paper we propose WGM in image processing and formulated the algorithm as a numerical solution of the total variation (TV) denoising method. The properties of the wavelets such as, regularity, vanishing moments, smoothness and the properties of Galerkin method which uses an algebraic formulation are very attractive for image processing. The necessary changes for adapting the WGM algorithm as a solution of total variation problem are derived. The performance of the proposed method in enhancing the noisy images against noise additions is compared with l 1 , non-convex tight regularization algorithm and finite element method (FEM). The effectiveness of the proposed method is confirmed by the improved structural similarity index (SSIM) and peak signal to noise ratio (PSNR) based image quality metrics. A GMM based image classification system is developed for determining the efficiency of the proposed method based on accuracy of the classification. The improved classification accuracy of proposed method confirmed its effectiveness. Image classification is performed as a quality metric for analyzing the performance of proposed algorithm. For sparse reconstruction, the proposed method gives performance comparable to most of the algorithms used for comparison and has much reduced run time.

关键词： Image Denoising Total Variation Methods Wavelet Galerkin Method Wavelet Galerkin Denoising

来源：评论

学校读者我要写书评

暂无评论

A deep learning approach for Malayalam morphological analysis at character level

引用

Procedia Computer Science 2018年 132卷 47-54页

作者： Premjith B Soman K.P. M Anand Kumar Center for Computational Engineering and Networking (CEN) Amrita School of Engineering Coimbatore Amrita Vishwa Vidyapeetham India

Morphological analysis is one of the fundamental tasks in computational processing of natural languages. It is the study of the rules of word construction by analysing the syntactic properties and morphological information. In order to perform this task, morphemes have to be separated from the original word. This process is termed as sandhi splitting. Sandhi splitting is important in the morphological analysis of agglutinative languages like Malayalam, because of the richness in morphology, inflections and sandhi. Due to sandhi, many morphological changes occur at the conjoining position of morphemes. Therefore, determining the morpheme boundaries becomes a tough task, especially in languages like Malayalam. In this paper, we propose a deep learning approach for learning the rules for identifying the morphemes automatically and segmenting them from the original word. Then, individual morphemes can be further analysed to identify the grammatical structure of the word. Three different systems were developed for this analysis using Recurrent Neural Network (RNN), Long Short-Term Memory (LSTM) and Gated Recurrent Units (GRU) and obtained accuracies 98.08%, 97.88% and 98.16% respectively.

关键词： Malayalam morphological analyser Sandhi splitting Deep learning RNN LSTM GRU

来源：评论

学校读者我要写书评

暂无评论

Deep AlexNet with Reduced Number of Trainable Parameters for Satellite Image Classification

引用

Procedia Computer Science 2018年 143卷 931-938页

作者： Anju Unnikrishnan Sowmya V Soman K P Center for Computational Engineering and Networking (CEN) Amrita School of Engineering Coimbatore Amrita Vishwa Vidyapeetham India

The emergence of machine learning algorithms enhanced the effectiveness of satellite image applications. On account of the high variability in-born in satellite data, a large portion of the present classification proposals are not sensible for handling satellite datasets. In this paper, we assessed profound learning systems in view of Convolutional Neural Networks for the precise order of multispectral remote sensing information. The Normalized Difference Vegetation Index (NDVI) is a solitary parameter for detecting landcover by utilizing the red and near-infrared band (NIR) information of the electromagnetic spectrum. NDVI is used to break down remote detecting images and survey the presence or absence of live green vegetation. The experiment is conducted on new publicaly available SAT-4 dataset, where the classes involved are types of vegetation. As NDVI computation require just two band data, it takes the benefit of both RED and NIR band information to classify diverse land covers. In the present work, the AlexNet architecture with two band information along with the reduced number of filters were trained and high-level features obtained from tested model managed to classify different land cover classes in the dataset. The proposed architecture is compared against the benchmark and results are estimated in terms of accuracy, precision and total number of trainable parameters. The proposed design brings about the aggregate diminishment of trainable parameters, while retaining high accuracy and precision, when compared against the existing architecture, which utilizes four bands.

关键词： Satellite Image Classification Convolutional networks SAT-4 Normalized Difference Vegetation Index Trainable Parameters Land Cover

来源：评论

学校读者我要写书评

暂无评论

A Better Digital Filtering Technique for Estimation of SPO2 and Heart Rate from PPG Signals

A Better Digital Filtering Technique for Estimation of SPO2 ...

引用

2018 International Conference on Inventive Research in Computing Applications, ICIRCA 2018

作者： Anagha, S. Suyampulingam, A. Ramachandran, K.I. Department of Electrical and Electronics Engineering Amrita School of Engineering Amrita Vishwa Vidyapeetham Amrita University Coimbatore India Amrita Vishwa Vidyapeetham Amrita University Center for Computational Engineering and Networking Amrita School of Engineering Coimbatore India

ISBN: (纸本)9781538624562

Peripheral arterial disease (PAD) is defined as the deficiency of oxygen rich blood to organs especially to lower extremities. PAD is caused due to shrinking of arteries and it is very difficult to identify the disease in its beginning stage. PAD is often diagnosed when it develops complications and in worst case it may even lead to leg amputation. This indicates that early detection of PAD is essential. But the existing method to detect PAD is time consuming, inefficient and requires assistance of expertise. Blood oxygen saturation (SPO2) level can serve as early warning of PAD. A simulation model to detect PAD based on SPO2 levels is presented in this paper. Non-invasive measurement of SPO2 can be done using photoplethysmography (PPG). Pre-processing of the PPG signals are done. Gaussian filter and Hamming filter are implemented. Signal to noise ratio (SNR) of the filtered and unfiltered signals are done. An algorithm to compute SPO2 values from PPG signals and to detect PAD is presented. In addition a novel algorithm is developed to compute heart rate from PPG signal. Error analysis on the SPO2 and heart rate values of both Gaussian and hamming filtered signals is also done in this paper. © 2018 IEEE.

关键词： Photoplethysmography

来源：评论

学校读者我要写书评

暂无评论

Multivariate extension of phase synchronization improves the estimation of region-to-region source space functional connectivity

引用

Brain Multiphysics 2021年 2卷

作者： Bruña, Ricardo Pereda, Ernesto Laboratory for Cognitive and Computational Neuroscience Canter for Biomedical Technology Technical University of Madrid Pozuelo de Alarcón Madrid Spain Department of Experimental Psychology Faculty of Psychology Complutense University of Madrid Pozuelo de Alarcón Madrid Spain Networking Research Center on Bioengineering Biomaterials and Nanomedicine (CIBER-BBN) Madrid Spain Electrical Engineering and Bioengineering Group Departamento de Ingeniería Industrial Instituto Universitario de Neurociencia (IUNE) e ITB University of La Laguna San Cristobal de La Laguna Tenerife Spain

The estimation of functional connectivity (FC) from noninvasive electrophysiological data recorded from sensors outside the skull requires transforming these data into a source space. As the number of sensors is much lower than the number of electrophysiological sources, the brain activity is usually parcellated into anatomical regions, and the FC between each pair of regions is then estimated. In this work, we generate a set of simulated scenarios with different configurations and coupling levels between synthetic time series. Then, this simulated brain activity is converted into simulated MEG sensor-space data and reconstructed back into the source space. Last, we estimated the FC between different regions using different approaches commonly used in the literature and compared them with a novel approach. Our results show that this novel approach, based on using all the information in each region, clearly outperforms classical approaches based on a representative time series. The proposed approach is more sensitive to the level of coupling and the extent of the area synchronized, and the resulting estimate better reflects the underlying FC. Based on these results, we strongly discourage using a representative time series to summarize large brain areas' activity when calculating FC. Statement of significance: While it is now well established that mechanical instabilities play an important role for cortical folding in the developing human brain, the mechanisms on the cellular scale leading to those macroscopic structural changes remain insufficiently understood. Here, we demonstrate that a two-field mechanical model coupling cell division and migration with volume growth is capable of capturing the spatial and temporal distribution of the cell density and the corresponding cortical folding pattern observed in the human fetal brain. The presented model provides a platform to obtain important insights into the cellular mechanisms underlying normal cortical folding and,

关键词： Functional connectivity Multivariate phase synchronization Source reconstruction

来源：评论

学校读者我要写书评

暂无评论

Improved Epoch Extraction from Telephonic Speech Using Flat-Top Cosine Windows

Improved Epoch Extraction from Telephonic Speech Using Flat-...

引用

IEEE India Conference (INDICON)

作者： B. Ganga Gowri Soman. K.P. Govind. D. Center for Computational Engineering & Networking (CEN) Amrita Vishwa Vidyapeetham India

ISBN: (纸本)9781538682364

Epoch extraction from the telephonic speech signal is challenging due to the attenuation of the low frequency information. The present work focuses on the strengthening of the low frequency information in the speech signal by utilizing the properties of Hilbert envelope and filter out the vocal tract resonances by an FIR filter with the center frequency at 0Hz and is based on the flat-top cosine window. The properties of the flat-top cosine window are effectively exploited in the extraction of epoch information from the Hilbert envelope obtained from the speech signal. The proposed method shows improvement in identification rate and identification accuracy when compared with the performance of ZFF of the Hilbert of speech signal for both clean and telephonic speech signal.

关键词： Telephonic speech Hilbert envelope Flat-top cosine window Voice Signal Cosine EPOCH Hilbert windows windows center frequency

来源：评论

学校读者我要写书评

暂无评论

Effect of Decolorized Images In Scene Classification Using Deep Convolution Features

引用

Procedia Computer Science 2018年 143卷 954-961页

作者： Nikhil Damodaran V Sowmya D Govind K P Soman Center For Computational Engineering and Networking (CEN) Amrita School of Enginerring Coimbatore Amrita Vishwa Vidyapeetham India

Scene classification is considered as an imperative issue for computer vision and has got extensive consideration in the recent past. Due to recent developments in high performance computing units such as GPUs, popularly known deep learning algorithm namely, convolutional neural networks (CNNs), exploits huge datasets to give powerful models. The paper proposes the use of transfer learning technique, by which a pre-trained model known as Places-CNN is used to generate feature vectors for each scene image of the dataset. The scene-classification experiments are conducted on the Oliva Torralba (OT) scene dataset, which consists of eight outdoor scene categories. The features were extracted from the fully connected layer of the pre-trained Places CNN architecture. The deep features were extracted from the input color images and the grayscale images converted using two different techniques based on singular value decomposition (SVD). The results obtained from classification experiments show that, models trained on SVD-Decolorized and Modified-SVD decolorized images give comparable performance to the input color images. Unlike the color images, which use three planes (RGB) of information, the grayscale images use only one plane of information. The grayscale images were able to retain the required shape and texture information from the original RGB images and, thus sufficient to categorize the classes of scene images.

关键词： CNN SVM SVD MLP Transfer Learning

来源：评论

学校读者我要写书评

暂无评论

Real time Optimal Operation of an Indian Microgrid using Convex Optimisation

Real time Optimal Operation of an Indian Microgrid using Con...

引用

2018 International Conference on Circuits and Systems in Digital Enterprise Technology, ICCSDET 2018

作者： Kumar, Anu G. Sindhu, M.R. Kumar, Sachin Dept. of Electrical Electronics Engg. Amrita School of Engineering Amrita Vishwa Vidyapeetham Coimbatore India Center for Computational Engg. Networking Amrita School of Engineering Amrita Vishwa Vidyapeetham Coimbatore India

ISBN: (纸本)9781538605769

This paper proposes a convex optimization method with an objective to minimize the operating cost of the microgrid (MG). The operating cost is minimised by optimal operation of various distributed generation (DG) units with renewable energy resources such as photovoltaic panel (PV) and Wind turbine (WT). Residential loads are connected to the microgrid, and the microgrid is connected to the main grid. The DG units are selected based on cost rate and load demand at a given time. The power is exchanged with the utility grid based on the power availability in the microgrid. The convex optimization based computation techniques are tested on the typical micro-grid and the simulation results are compared with other optimization techniques © 2018 IEEE.

关键词： Distributed power generation

来源：评论

学校读者我要写书评

暂无评论

A Comparative Analysis of Total Variation and Least Square Based Hyperspectral Image Denoising Methods

A Comparative Analysis of Total Variation and Least Square B...

引用

International Conference on Communications and Signal Processing

作者： Naveen Varghese Jacob V Sowmya K P Soman Center For Computational Engineering and Networking (CEN) Amrita Vishwa Vidyapeetham Coimbatore India

Hyperspectral image (HSI) with high spectral resolution will be always degraded by the noise accumulation. Therefore, image denoising is a fundamental preprocessing technique which improves the precision of successive processes like image classification, unmixing etc. In this paper, we compare least square (LS) weighted regularization in spectral domain with spatial least square and total variation (TV) denoising techniques. These methods are experimented on real, and noise simulated hyperspectral image datasets. The contrast and edges of the image are well preserved in the spectral LS. The image contrast varies in spatial LS, and edge informations are lost in TV. The experimental results show that, the spectral LS is superior to other two techniques in terms of visual interpretation, Signal-to-Noise Ratio (SNR) and Structural Similarity (SSIM) Index.

关键词： Noise reduction Signal to noise ratio Noise measurement Noise level Hyperspectral sensors Visualization TV

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：