Abstract
Abstract This paper is a study to improve the classification efficiency of rotating objects by using deep neural net-works to which a deep learning algorithm was applied. For the cl assification experiment of rotating ob jects, COIL-20 is used as data and total 3 types of classifiers are co mpared and analyzed. 3 types of classifiers used in the study include PCA classifier to derive a feature va lue while reducing the dimension of data b yusing Principal Component Analysis and classify by using euclid ean distance, MLP classifier of the way of re-ducing the error energy by using error back-propagation algorit hm and finally, deep learning applied DBN classifier of the way of increasing the probability of observin g learning data through pre-training and re-ducing the error energy through fine-tuning. In order to identi fy the structure-specific error rate of the deepneural networks, the experiment is carried out while changing the number of hidden layers and number o f hidden neurons. The classifier using DBN showed the lowest erro r rate. Its structure of deep neural net-works with 2 hidden layers showed a high recognition rate by mov ing parameters to a location helpful fo rrecognition.Key Words : PCA, MLP, DBN, Pre-training, Deep learningReceived: Jun. 15, 2015Revised : Sep. 16, 2015Accepted: Sep. 16, 2015
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: Journal of Korean Institute of Intelligent Systems
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
