Real-Time Multi Oriented Ancient Script Recognition using Single Layer Hierarchical Graph Neuron (SLHGN)

2018 ◽  
Author(s):  
Benny Benyamin Nasution ◽  
Asad Iqbal Khan ◽  
Anang Hudaya Muhamad Amin
Keyword(s):  
Real Time ◽  
Single Layer ◽  
Script Recognition ◽  
Hierarchical Graph

scholarly journals Cross-Layer Framework for Multiuser Real Time H.264/AVC Video Encoding and Transmission over Block Fading MIMO Channels Using Outage Probability

2014 ◽  
Vol 2014 ◽  
pp. 1-10 ◽  
Author(s):  
Slavche Pejoski ◽  
Venceslav Kafedziski
Keyword(s):  
Real Time ◽  
Fading Channels ◽  
Video Transmission ◽  
Single Layer ◽  
Rate Distortion ◽  
Wireless Channel ◽  
Video Encoding ◽  
Cross Layer ◽  
Mimo Channels ◽  
Channel Errors

We present a framework for cross-layer optimized real time multiuser encoding of video using a single layer H.264/AVC and transmission over MIMO wireless channels. In the proposed cross-layer adaptation, the channel of every user is characterized by the probability density function of its channel mutual information and the performance of the H.264/AVC encoder is modeled by a rate distortion model that takes into account the channel errors. These models are used during the resource allocation of the available slots in a TDMA MIMO communication system with capacity achieving channel codes. This framework allows for adaptation to the statistics of the wireless channel and to the available resources in the system and utilization of the multiuser diversity of the transmitted video sequences. We show the effectiveness of the proposed framework for video transmission over Rayleigh MIMO block fading channels, when channel distribution information is available at the transmitter.

scholarly journals Object detection in real time based on improved single shot multi-box detector algorithm

2020 ◽  
Vol 2020 (1) ◽  
Author(s):  
Ashwani Kumar ◽  
Zuopeng Justin Zhang ◽  
Hongbo Lyu
Keyword(s):  
Neural Networks ◽  
Object Detection ◽  
Real Time ◽  
Convolutional Neural Networks ◽  
Single Layer ◽  
Single Shot ◽  
Convolutional Network ◽  
Detection Techniques ◽  
Computational Performance ◽  
Speed Up

Abstract In today’s scenario, the fastest algorithm which uses a single layer of convolutional network to detect the objects from the image is single shot multi-box detector (SSD) algorithm. This paper studies object detection techniques to detect objects in real time on any device running the proposed model in any environment. In this paper, we have increased the classification accuracy of detecting objects by improving the SSD algorithm while keeping the speed constant. These improvements have been done in their convolutional layers, by using depth-wise separable convolution along with spatial separable convolutions generally called multilayer convolutional neural networks. The proposed method uses these multilayer convolutional neural networks to develop a system model which consists of multilayers to classify the given objects into any of the defined classes. The schemes then use multiple images and detect the objects from these images, labeling them with their respective class label. To speed up the computational performance, the proposed algorithm is applied along with the multilayer convolutional neural network which uses a larger number of default boxes and results in more accurate detection. The accuracy in detecting the objects is checked by different parameters such as loss function, frames per second (FPS), mean average precision (mAP), and aspect ratio. Experimental results confirm that our proposed improved SSD algorithm has high accuracy.

Hybrid Modeling of Melt Pool Geometry in Additive Manufacturing Using Neural Networks

2021 ◽  
Author(s):  
Kevontrez Jones ◽  
Zhuo Yang ◽  
Ho Yeung ◽  
Paul Witherell ◽  
Yan Lu
Keyword(s):  
Neural Networks ◽  
Finite Element ◽  
Additive Manufacturing ◽  
Real Time ◽  
Hybrid Model ◽  
Computational Models ◽  
Single Layer ◽  
Element Model ◽  
Additive Process ◽  
Melt Pool

Abstract Laser powder-bed fusion is an additive manufacturing (AM) process that offers exciting advantages for the fabrication of metallic parts compared to traditional techniques, such as the ability to create complex geometries with less material waste. However, the intricacy of the additive process and extreme cyclic heating and cooling leads to material defects and variations in mechanical properties; this often results in unpredictable and even inferior performance of additively manufactured materials. Key indicators for the potential performance of a fabricated part are the geometry and temperature of the melt pool during the building process, due to its impact upon the underlining microstructure. Computational models, such as those based on the finite element method, of the AM process can be used to elucidate and predict the effects of various process parameters on the melt pool, according to physical principles. However, these physics-based models tend to be too computationally expensive for real-time process control. Hence, in this work, a hybrid model utilizing neural networks is proposed and demonstrated to be an accurate and efficient alternative for predicting melt pool geometries in AM, which provides a unified description of the melting conditions. The results of both a physics-based finite element model and the hybrid model are compared to real-time experimental measurements of the melt pool during single-layer AM builds using various scanning strategies.

scholarly journals NUMERICAL INVESTIGATION OF THE ACOUSTIC DAMPING OF PLANE ACOUSTIC WAVES BY PERFORATED LINERS WITH BIAS FLOW

2012 ◽  
Vol 19 ◽  
pp. 6-17
Author(s):  
DAN ZHAO ◽  
ZHI YUAN ZHONG
Keyword(s):  
Frequency Domain ◽  
Real Time ◽  
Acoustic Waves ◽  
Method Of Lines ◽  
Single Layer ◽  
Numerical Results ◽  
Domain Model ◽  
Acoustic Energy ◽  
Acoustic Damping ◽  
Bias Flow

Perforated liners are extensively used in aero-engines and gas turbine combustors to suppress combustion instabilities. These liners, typically subjected to a low Mach number bias flow (a cooling flow through perforated holes), are fitted along the bounding walls of a combustor to convert acoustic energy into flow energy by generating vorticity at the rims of the perforated apertures. To investigate the acoustic damping of such liners with bias flow on plane acoustic waves, a time-domain numerical model is developed to compute acoustic wave propagation in a cylindrical duct with a single-layer liner attached. The damping mechanism of the liner is characterized in real-time by using a 'compliance', developed especially for this work. It is a rational function representation of the frequency-domain homogeneous compliance adapted from the Rayleigh conductivity of a single aperture with mean bias flow in the z-domain. The liner 'compliance' model is then incorporated into partial differential equations of the duct system, which are solved by using the method of lines. The numerical results are then evaluated by comparing with the numerical results of Eldredge and Dowling's frequency-domain model. Good agreement is observed. This confirms that the model and the approach developed are suitable for real-time characterizing the acoustic damping of perforated liners.

scholarly journals New Fabrication Techniques for Ring-Array Transducers for Real-Time 3D Intravascular Ultrasound

2009 ◽  
Vol 31 (4) ◽  
pp. 247-256 ◽  
Author(s):  
Edward D. Light ◽  
Victor Lieu ◽  
Stephen W. Smith
Keyword(s):  
Real Time ◽  
Cross Talk ◽  
Insertion Loss ◽  
Nearest Neighbor ◽  
Vena Cava ◽  
Single Layer ◽  
3D Ultrasound ◽  
Round Trip ◽  
Cook Medical ◽  
The Mean

We have previously described miniature 2D array transducers integrated into a Cook Medical, Inc. vena cava filter deployment device. While functional, the fabrication technique was very labor intensive and did not lend itself well to efficient fabrication of large numbers of devices. We developed two new fabrication methods that we believe can be used to efficiently manufacture these types of devices in greater than prototype numbers. One transducer consisted of 55 elements operating near 5 MHz. The interelement spacing is 0.20 mm. It was constructed on a flat piece of copper-clad polyimide and then wrapped around an 11 French catheter of a Cook Medical, Inc. inferior vena cava (IVC) filter deployment device. We used a braided wiring technology from Tyco Electronics Corp. to connect the elements to our real-time 3D ultrasound scanner. Typical measured transducer element bandwidth was 20% centered at 4.7 MHz and the 50 Ω round trip insertion loss was −82 dB. The mean of the nearest neighbor cross talk was −37.0 dB. The second method consisted of a 46-cm long single layer flex circuit from MicroConnex that terminates in an interconnect that plugs directly into our system cable. This transducer had 70 elements at 0.157 mm interelement spacing operating at 4.8 MHz. Typical measured transducer element bandwidth was 29% and the 50 Ω round trip insertion loss was −83 dB. The mean of the nearest neighbor cross talk was −33.0 dB.

scholarly journals Decentralized Identification and Control in Real-Time of a Robot Manipulator via Recurrent Wavelet First-Order Neural Network

2015 ◽  
Vol 2015 ◽  
pp. 1-12 ◽  
Author(s):  
Luis A. Vázquez ◽  
Francisco Jurado ◽  
Alma Y. Alanís
Keyword(s):  
Neural Network ◽  
Real Time ◽  
Robot Manipulator ◽  
Single Layer ◽  
Activation Function ◽  
Neural Structure ◽  
Neural Controller ◽  
First Order ◽  
Parallel Configuration ◽  
And Control

A decentralized recurrent wavelet first-order neural network (RWFONN) structure is presented. The use of a wavelet Morlet activation function allows proposing a neural structure in continuous time of a single layer and a single neuron in order to identify online in a series-parallel configuration, using the filtered error (FE) training algorithm, the dynamics behavior of each joint for a two-degree-of-freedom (DOF) vertical robot manipulator, whose parameters such as friction and inertia are unknown. Based on the RWFONN subsystem, a decentralized neural controller is designed via backstepping approach. The performance of the decentralized wavelet neural controller is validated via real-time results.

Sign in / Sign up

Export Citation Format

Share Document