Understanding the Convolutional Neural Network & it’s Research Aspects in Deep Learning
Journal Title: International Journal for Research in Applied Science and Engineering Technology (IJRASET) - Year 2017, Vol 5, Issue 6
Abstract
Convolutional Neural Network (CNN) is the most important Deep Neural Network (DNN) architecture to implement the Deep Learning’s application of data and pattern representation in an effective and efficient manner. It uses the idea of animal’s visual cortex organization to achieve connectivity pattern between its neurons. A receptive field is a restricted part of the space where a respond to stimuli are done by the individual cortical neurons. The main motivation behind the development of convolutional networks is the biological processes and CNNs are considered as the multilayer perceptrons’ variations that are designed for the purpose of providing the minimal usage of pre-processing. The major applications of convolutional neural network include image recognition, natural language processing, recommender systems and video recognition. We have tried to put an honest effort in this paper to analyze the Convolutional Neural Network (CNN) and the various developments made in its area of research.
Authors and Affiliations
Manish Kumar Singh, Prof. G S Baluja, Dr. Dinesh Prasad Sahu
Keywords
Related Articles
- EP ID EP24512
- DOI -
- Views 355
- Downloads 9
Improving Engineering Properties of Clayey Soil with Addition of Cement and Coir Fiber
In recent days, improving engineering properties of soil with various waste product such as(hair fiber, coir fiber) etc. We found that these by-products can improve and maintain the properties of soil. Soil reinforcemen...
Elimination of Leakage Current in PV Based Current Source Inverter
This paper deals with the implementation of four leg inverter to eliminate the earth leakage current. In the PV based system, some amount of leakage current induces on the inverter output terminals due to high switching...
An Improved Data Transmission Protocol for VANETS Using Link State Information
Efficient data delivery in vehicular ad hoc networks (VANETs) with high mobility is a challenging issue due to dynamic topology changes and unstable wireless links. The opportunistic routing protocols can improve the re...
Sensor Based Human Gait Recognition for Drunk State
Many research studies have examined human activity recognition and body posture and gait detection. Relatively fewer studies have investigated specifically into shaky, abnormal gait or how drunken person move in a low l...
Simulation Development of Linux Firewall
Open-source systems are not going to replace offerings from commercial vendors. They do, however, offer an increasingly viable alternative. Open-source solutions should be evaluated side-by-side with commercial as part...