scholarly journals Sparse Representation Based Multi Object Tracking using GPU

2019 ◽  
Vol 9 (2S) ◽  
pp. 585-591
Keyword(s):  
Sparse Representation ◽  
Object Tracking ◽  
Matching Pursuit ◽  
Video Frame ◽  
Processing Unit ◽  
Computational Overhead ◽  
Proposed Model ◽  
Graphical Processing ◽  
Low Dimensional ◽  
State Of Art

This work proposes a sparse based representation for tracking multi object for the longer sequence of video frame. Object of interest are first identified and then represented with set of low dimensional feature covariance matrix. These feature of different object are kept in a dictionary. In order to classify the object, sparse based Orthogonal matching pursuit(OMP) algorithm is used. Furthermore, towards reducing the computational overhead, proposed model is implemented on a graphical processing unit enhanced with the multi threaded resource for parallelization of the task. Experimental results shows that proposed method out perform as compared with the state of art in identifying the objects.

Mixed-mode database miner classifier: Parallel computation of graphical processing unit mining

2021 ◽  
pp. 002072092098849
Author(s):  
Soumya Ranjan Nayak ◽  
S Sivakumar ◽  
Akash Kumar Bhoi ◽  
Gyoo-Soo Chae ◽  
Pradeep Kumar Mallick
Keyword(s):  
Credit Card ◽  
Mixed Mode ◽  
Processing Time ◽  
Gpu Computing ◽  
Graphical Processing Unit ◽  
Computational Time ◽  
Processing Unit ◽  
Large Set ◽  
Minimal Processing ◽  
Graphical Processing

Graphical processing unit (GPU) has gained more popularity among researchers in the field of decision making and knowledge discovery systems. However, most of the earlier studies have GPU memory utilization, computational time, and accuracy limitations. The main contribution of this paper is to present a novel algorithm called the Mixed Mode Database Miner (MMDBM) classifier by implementing multithreading concepts on a large number of attributes. The proposed method use the quick sort algorithm in GPU parallel computing to overcome the state of the art limitations. This method applies the dynamic rule generation approach for constructing the decision tree based on the predicted rules. Moreover, the implementation results are compared with both SLIQ and MMDBM using Java and GPU with the computed acceleration ratio time using the BP dataset. The primary objective of this work is to improve the performance with less processing time. The results are also analyzed using various threads in GPU mining using eight different datasets of UCI Machine learning repository. The proposed MMDBM algorithm have been validated on these chosen eight different dataset with accuracy of 91.3% in diabetes, 89.1% in breast cancer, 96.6% in iris, 89.9% in labor, 95.4% in vote, 89.5% in credit card, 78.7% in supermarket and 78.7% in BP, and simultaneously, it also takes less computational time for given datasets. The outcome of this work will be beneficial for the research community to develop more effective multi thread based GPU solution in GPU mining to handle large set of data in minimal processing time. Therefore, this can be considered a more reliable and precise method for GPU computing.

Concise CNN model for face expression recognition

2021 ◽  
pp. 1-9
Author(s):  
Harshadkumar B. Prajapati ◽  
Ankit S. Vyas ◽  
Vipul K. Dabhi
Keyword(s):  
Network Architecture ◽  
Expression Recognition ◽  
Human Machine Interaction ◽  
Face Expression ◽  
Computational Overhead ◽  
Average Accuracy ◽  
Proposed Model ◽  
Face Expression Recognition ◽  
Expression Classification ◽  
Machine Interaction

Face expression recognition (FER) has gained very much attraction to researchers in the field of computer vision because of its major usefulness in security, robotics, and HMI (Human-Machine Interaction) systems. We propose a CNN (Convolutional Neural Network) architecture to address FER. To show the effectiveness of the proposed model, we evaluate the performance of the model on JAFFE dataset. We derive a concise CNN architecture to address the issue of expression classification. Objective of various experiments is to achieve convincing performance by reducing computational overhead. The proposed CNN model is very compact as compared to other state-of-the-art models. We could achieve highest accuracy of 97.10% and average accuracy of 90.43% for top 10 best runs without any pre-processing methods applied, which justifies the effectiveness of our model. Furthermore, we have also included visualization of CNN layers to observe the learning of CNN.

scholarly journals A multiresolution spatial parameterization for the estimation of fossil-fuel carbon dioxide emissions via atmospheric inversions

2014 ◽  
Vol 7 (5) ◽  
pp. 1901-1918 ◽  
Author(s):  
J. Ray ◽  
V. Yadav ◽  
A. M. Michalak ◽  
B. van Bloemen Waanders ◽  
S. A. McKenna
Keyword(s):  
Spatial Variability ◽  
Carbon Dioxide Emissions ◽  
Fossil Fuel ◽  
Matching Pursuit ◽  
Regional Scale ◽  
Reconstruction Algorithm ◽  
Spatiotemporal Variability ◽  
Realistic Representation ◽  
Non Gaussian ◽  
Low Dimensional

Abstract. The characterization of fossil-fuel CO2 (ffCO2) emissions is paramount to carbon cycle studies, but the use of atmospheric inverse modeling approaches for this purpose has been limited by the highly heterogeneous and non-Gaussian spatiotemporal variability of emissions. Here we explore the feasibility of capturing this variability using a low-dimensional parameterization that can be implemented within the context of atmospheric CO2 inverse problems aimed at constraining regional-scale emissions. We construct a multiresolution (i.e., wavelet-based) spatial parameterization for ffCO2 emissions using the Vulcan inventory, and examine whether such a~parameterization can capture a realistic representation of the expected spatial variability of actual emissions. We then explore whether sub-selecting wavelets using two easily available proxies of human activity (images of lights at night and maps of built-up areas) yields a low-dimensional alternative. We finally implement this low-dimensional parameterization within an idealized inversion, where a sparse reconstruction algorithm, an extension of stagewise orthogonal matching pursuit (StOMP), is used to identify the wavelet coefficients. We find that (i) the spatial variability of fossil-fuel emission can indeed be represented using a low-dimensional wavelet-based parameterization, (ii) that images of lights at night can be used as a proxy for sub-selecting wavelets for such analysis, and (iii) that implementing this parameterization within the described inversion framework makes it possible to quantify fossil-fuel emissions at regional scales if fossil-fuel-only CO2 observations are available.

Robust object tracking based on sparse representation and incremental weighted PCA

2016 ◽  
Vol 76 (2) ◽  
pp. 2039-2057 ◽  
Author(s):  
Xiaofen Xing ◽  
Fuhao Qiu ◽  
Xiangmin Xu ◽  
Chunmei Qing ◽  
Yinrong Wu
Keyword(s):  
Sparse Representation ◽  
Object Tracking

Image Denoising Using Sparse Representation and Principal Component Analysis

2021 ◽  
pp. 2250033
Author(s):  
Maryam Abedini ◽  
Horriyeh Haddad ◽  
Marzieh Faridi Masouleh ◽  
Asadollah Shahbahrami
Keyword(s):  
Principal Component Analysis ◽  
Sparse Representation ◽  
Image Denoising ◽  
Matching Pursuit ◽  
Signal To Noise Ratio ◽  
Principal Component ◽  
Structural Similarity ◽  
Component Analysis ◽  
Block Matching ◽  
Discrete Wavelet

This study proposes an image denoising algorithm based on sparse representation and Principal Component Analysis (PCA). The proposed algorithm includes the following steps. First, the noisy image is divided into overlapped [Formula: see text] blocks. Second, the discrete cosine transform is applied as a dictionary for the sparse representation of the vectors created by the overlapped blocks. To calculate the sparse vector, the orthogonal matching pursuit algorithm is used. Then, the dictionary is updated by means of the PCA algorithm to achieve the sparsest representation of vectors. Since the signal energy, unlike the noise energy, is concentrated on a small dataset by transforming into the PCA domain, the signal and noise can be well distinguished. The proposed algorithm was implemented in a MATLAB environment and its performance was evaluated on some standard grayscale images under different levels of standard deviations of white Gaussian noise by means of peak signal-to-noise ratio, structural similarity indexes, and visual effects. The experimental results demonstrate that the proposed denoising algorithm achieves significant improvement compared to dual-tree complex discrete wavelet transform and K-singular value decomposition image denoising methods. It also obtains competitive results with the block-matching and 3D filtering method, which is the current state-of-the-art for image denoising.

Robust object tracking based on sparse representation

2010 ◽  
Author(s):  
Shengping Zhang ◽  
Hongxun Yao ◽  
Xin Sun ◽  
Shaohui Liu
Keyword(s):  
Sparse Representation ◽  
Object Tracking

scholarly journals ACCELERATING 3D NON-RIGID REGISTRATION USING GRAPHICS HARDWARE

2008 ◽  
Vol 08 (01) ◽  
pp. 81-98 ◽  
Author(s):  
NICOLAS COURTY ◽  
PIERRE HELLIER
Keyword(s):  
Image Analysis ◽  
Classical Method ◽  
Graphics Hardware ◽  
Processing Unit ◽  
Recursive Filtering ◽  
Rigid Registration ◽  
Guided Surgery ◽  
Graphical Processing Unit Implementation ◽  
Demons Algorithm ◽  
Graphical Processing

There is an increasing need for real-time implementation of 3D image analysis processes, especially in the context of image-guided surgery. Among the various image analysis tasks, non-rigid image registration is particularly needed and is also computationally prohibitive. This paper presents a GPU (Graphical Processing Unit) implementation of the popular Demons algorithm using a Gaussian recursive filtering. Acceleration of the classical method is mainly achieved by a new filtering scheme on GPU which could be reused in or extended to other applications and denotes a significant contribution to the GPU-based image processing domain. This implementation was able to perform a non-rigid registration of 3D MR volumes in less than one minute, which corresponds to an acceleration factor of 10 compared to the corresponding CPU implementation. This demonstrated the usefulness of such method in an intra-operative context.

scholarly journals A Hybrid Optimized LSTM Models for Human Activity Recognition with IOT Devices

2021 ◽  
pp. 182-189
Author(s):  
S. Arokiaraj ◽  
Dr. N. Viswanathan
Keyword(s):  
Deep Learning ◽  
Internet Of Things ◽  
Short Term Memory ◽  
The Other ◽  
Learning Models ◽  
Computational Overhead ◽  
Temporal Features ◽  
Human Movements ◽  
Proposed Model ◽  
Iot Devices

With the advent of Internet of things(IoT),HA (HA) recognition has contributed the more application in health care in terms of diagnosis and Clinical process. These devices must be aware of human movements to provide better aid in the clinical applications as well as user’s daily activity.Also , In addition to machine and deep learning algorithms, HA recognition systems has significantly improved in terms of high accurate recognition. However, the most of the existing models designed needs improvisation in terms of accuracy and computational overhead. In this research paper, we proposed a BAT optimized Long Short term Memory (BAT-LSTM) for an effective recognition of human activities using real time IoT systems. The data are collected by implanting the Internet of things) devices invasively. Then, proposed BAT-LSTM is deployed to extract the temporal features which are then used for classification to HA. Nearly 10,0000 dataset were collected and used for evaluating the proposed model. For the validation of proposed framework, accuracy, precision, recall, specificity and F1-score parameters are chosen and comparison is done with the other state-of-art deep learning models. The finding shows the proposed model outperforms the other learning models and finds its suitability for the HA recognition.

Sign in / Sign up

Export Citation Format

Share Document