scholarly journals Study on threshold selection method of continuous flame images of spray combustion in the low-pressure chamber

2021 ◽  
pp. 101195
Author(s):  
Kai Xie ◽  
Yunjing Cui ◽  
Chunlin Wang ◽  
Gan Cui ◽  
Guanqin Wang ◽  
...  
Keyword(s):  
Low Pressure ◽  
Selection Method ◽  
Pressure Chamber ◽  
Spray Combustion ◽  
Threshold Selection

A New Threshold Selection Method for Separation of Objects and Background

1995 ◽  
Vol 2 (6) ◽  
pp. 444-448 ◽  
Author(s):  
Chang-Heon Woo ◽  
Soo-Yong Kim
Keyword(s):  
Selection Method ◽  
Threshold Selection

Substation Infrared Image Segmentation Based on Novel Threshold Selection Method

2019 ◽  
Vol 39 (8) ◽  
pp. 0810002
Author(s):  
赵庆生 Zhao Qingsheng ◽  
王雨滢 Wang Yuying ◽  
王旭平 Wang Xuping ◽  
郭尊 Guo Zun
Keyword(s):  
Image Segmentation ◽  
Infrared Image ◽  
Selection Method ◽  
Threshold Selection ◽  
Infrared Image Segmentation

scholarly journals Communication Optimization Schemes for Accelerating Distributed Deep Learning Systems

2020 ◽  
Vol 10 (24) ◽  
pp. 8846
Author(s):  
Jaehwan Lee ◽  
Hyeonseong Choi ◽  
Hyeonwoo Jeong ◽  
Baekhyeon Noh ◽  
Ji Sun Shin
Keyword(s):  
Deep Learning ◽  
Selection Method ◽  
Learning System ◽  
Threshold Selection ◽  
L1 Norm ◽  
Optimization Scheme ◽  
Communication Time ◽  
Local Cache ◽  
Hidden Layer ◽  
Data Optimization

In a distributed deep learning system, a parameter server and workers must communicate to exchange gradients and parameters, and the communication cost increases as the number of workers increases. This paper presents a communication data optimization scheme to mitigate the decrease in throughput due to communication performance bottlenecks in distributed deep learning. To optimize communication, we propose two methods. The first is a layer dropping scheme to reduce communication data. The layer dropping scheme we propose compares the representative values of each hidden layer with a threshold value. Furthermore, to guarantee the training accuracy, we store the gradients that are not transmitted to the parameter server in the worker’s local cache. When the value of gradients stored in the worker’s local cache is greater than the threshold, the gradients stored in the worker’s local cache are transmitted to the parameter server. The second is an efficient threshold selection method. Our threshold selection method computes the threshold by replacing the gradients with the L1 norm of each hidden layer. Our data optimization scheme reduces the communication time by about 81% and the total training time by about 70% in a 56 Gbit network environment.

Numerical simulation of turbulent flow in the throttle of the MBIR reactor’s low-pressure chamber

2015 ◽  
Vol 62 (8) ◽  
pp. 560-563
Author(s):  
V. A. Yarunichev ◽  
E. E. Orlova ◽  
Yu. V. Lemekhov ◽  
V. A. Shpanskii
Keyword(s):  
Numerical Simulation ◽  
Turbulent Flow ◽  
Low Pressure ◽  
Pressure Chamber

scholarly journals A new threshold selection method for peak over for nonstationary time series

2016 ◽  
Vol 39 ◽  
pp. 012071
Author(s):  
C R Zhou ◽  
Y F Chen ◽  
S H Gu ◽  
Q Huang ◽  
J C Yuan ◽  
...  
Keyword(s):  
Time Series ◽  
Selection Method ◽  
Nonstationary Time Series ◽  
Threshold Selection

Active Control of Propagating Shock Wave Using Low Pressure Chamber

2003 ◽  
Vol 2003.52 (0) ◽  
pp. 19-20
Author(s):  
Takuya Kuwayama ◽  
Shuuhei Otuki ◽  
Takeshi Yashima ◽  
Kazuhiko Yokota ◽  
Motoyuki Itoh
Keyword(s):  
Shock Wave ◽  
Active Control ◽  
Low Pressure ◽  
Pressure Chamber

Application of a New Kind of Wavelet Threshold De-Noising Method in Partial Discharge Signals

2014 ◽  
Vol 889-890 ◽  
pp. 780-785
Author(s):  
Cheng Long Xu ◽  
Hong Yu ◽  
Bi Qiang Du ◽  
Jun Li ◽  
Ze Kun Liu ◽  
...  
Keyword(s):  
Wavelet Transform ◽  
Simulation Experiment ◽  
Wavelet Coefficient ◽  
Partial Discharge ◽  
Noise Signal ◽  
Selection Method ◽  
Wavelet Coefficients ◽  
Threshold Selection ◽  
Whole Process ◽  
Maximum Value

In order to more effectively remove noise in partial discharge signals, it is proposed a new threshold selection method in this paper. This method firstly takes the signals before the partial discharge starting to happen as only contain noise signal, and then applies a wavelet transform to the only contain noise signal. Secondly record every detail part and the maximum value of wavelet coefficients of last layer approximation part, and take this value as its layer threshold. And then applies a wavelet transform to the partial discharge signals which contains noises. Next is to process wavelet coefficient of each layer using the selected threshold. Finally, the already handled wavelet coefficients is used to reconstruction the signals. The whole process of threshold choosing is automatic without human intervention. Simulation experiment show that compared with the traditional threshold selection method, this method can be better to remove the noise of the partial discharge signals, and it has a strong practical value.

Sign in / Sign up

Export Citation Format

Share Document