Design of a Cost-Effective Deep Convolutional Neural Network–Based Scheme for Diagnosing Faults in Smart Grids

2019 ◽  
Author(s):  
Hossein Hassani ◽  
Maryam Farajzadeh-Zanjani ◽  
Roozbeh Razavi-Far ◽  
Mehrdad Saif ◽  
Vasile Palade
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Smart Grids ◽  
Cost Effective ◽  
Deep Convolutional Neural Network

scholarly journals Sleep apnea detection using deep learning

2019 ◽  
Vol 13 (4) ◽  
pp. 261-266 ◽  
Author(s):  
Hnin Thiri Chaw ◽  
Sinchai Kamolphiwong ◽  
Krongthong Wongsritrang
Keyword(s):  
Neural Network ◽  
Sleep Apnea ◽  
Convolutional Neural Network ◽  
Cost Effective ◽  
Deep Convolutional Neural Network ◽  
Sleep Study ◽  
Proposed Model ◽  
Apnea Detection ◽  
Sleep Recording

Sleep apnea is the cessation of airflow at least 10 seconds and it is the type of breathing disorder in which breathing stops at the time of sleeping. The proposed model uses type 4 sleep study which focuses more on portability and the reduction of the signals. The main limitations of type 1 full night polysomnography are time consuming and it requires much space for sleep recording such as sleep lab comparing to type 4 sleep studies. The detection of sleep apnea using deep convolutional neural network model based on SPO2 sensor is the valid alternative for efficient polysomnography and it is portable and cost effective. The total number of samples from SPO2 sensors of 50 patients that is used in this study is 190,000. The performance of the overall accuracy of sleep apnea detection is 91.3085% with the loss rate of 2.3 using cross entropy cost function using deep convolutional neural network.

LENS CLASSIFICATION ACCORDING TO THE TYPE OF LIGHT SPOT USING A NEURAL NETWORK

2020 ◽  
Vol 2020 (4) ◽  
pp. 4-14
Author(s):  
Vladimir Budak ◽  
Ekaterina Ilyina
Keyword(s):  
Neural Network ◽  
Light Intensity ◽  
Convolutional Neural Network ◽  
Deep Convolutional Neural Network ◽  
Beam Angle ◽  
New Model ◽  
Technical Parameters ◽  
Transfer Training ◽  
Trained Network

The article proposes the classification of lenses with different symmetrical beam angles and offers a scale as a spot-light’s palette. A collection of spotlight’s images was created and classified according to the proposed scale. The analysis of 788 pcs of existing lenses and reflectors with different LEDs and COBs carried out, and the dependence of the axial light intensity from beam angle was obtained. A transfer training of new deep convolutional neural network (CNN) based on the pre-trained GoogleNet was performed using this collection. GradCAM analysis showed that the trained network correctly identifies the features of objects. This work allows us to classify arbitrary spotlights with an accuracy of about 80 %. Thus, light designer can determine the class of spotlight and corresponding type of lens with its technical parameters using this new model based on CCN.

Sign in / Sign up

Export Citation Format

Share Document