Open Access System for Information Sharing

Login Library

 

Article
Cited 0 time in webofscience Cited 0 time in scopus
Metadata Downloads

Selective deep convolutional neural network for low cost distorted image classification

Title
Selective deep convolutional neural network for low cost distorted image classification
Authors
Ha, MinhoByun, YounghoonKim, JeonghunLee, JaecheolLee, YoungjooLEE, SUNG GU
Date Issued
Dec-2019
Publisher
Institute of Electrical and Electronics Engineers Inc.
Abstract
Neural networks trained using images with a certain type of distortion should be better at classifying test images with the same type of distortion than generally-trained neural networks, given other factors being equal. Based on this observation, an ensemble of convolutional neural networks (CNNs) trained with different types and degrees of distortions is used. However, instead of simply classifying test images of unknown distortion types with the entire ensemble of CNNs, an extra tiny CNN is specifically trained to distinguish between the different types and degrees of distortions. Then, only the dedicated CNN for that specific type and degree of distortion, as determined by the tiny CNN, is activated and used to classify a possibly distorted test image. This proposed architecture, referred to as a \textit{selective deep convolutional neural network (DCNN)}, is implemented and found to result in high accuracy with low hardware costs. Detailed simulations with realistic image distortion scenarios using three popular datasets show that memory, MAC operations, and energy savings of up to 93.68%, 93.61%, and 91.92%, respectively, can be achieved with almost no reduction in image classification accuracy. The proposed selective DCNN scores up to 2.18x higher than the state-of-the-art DCNN model when evaluated using NetScore, a comprehensive metric that considers both CNN performance and hardware cost. In addition, it is shown that even higher hardware cost reduction can be achieved when selective DCNN is combined with previously proposed model compression techniques. Finally, experiments conducted with extended types and degrees of image distortion show that selective DCNN is highly scalable.
URI
http://oasis.postech.ac.kr/handle/2014.oak/100060
ISSN
2169-3536
Article Type
Article
Citation
IEEE Access, vol. 7, no. 1, page. 133030 - 133042, 2019-12
Files in This Item:
There are no files associated with this item.

qr_code

  • mendeley

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

Related Researcher

Researcher

 LEE, SUNG GU
Dept of Electrical Enginrg
Read more

Views & Downloads

Browse