Fast DCNN based on FWT, intelligent dropout and layer skipping for image retrieval
Copyright © 2017 Elsevier Ltd. All rights reserved..
Deep Convolutional Neural Network (DCNN) can be marked as a powerful tool for object and image classification and retrieval. However, the training stage of such networks is highly consuming in terms of storage space and time. Also, the optimization is still a challenging subject. In this paper, we propose a fast DCNN based on Fast Wavelet Transform (FWT), intelligent dropout and layer skipping. The proposed approach led to improve the image retrieval accuracy as well as the searching time. This was possible thanks to three key advantages: First, the rapid way to compute the features using FWT. Second, the proposed intelligent dropout method is based on whether or not a unit is efficiently and not randomly selected. Third, it is possible to classify the image using efficient units of earlier layer(s) and skipping all the subsequent hidden layers directly to the output layer. Our experiments were performed on CIFAR-10 and MNIST datasets and the obtained results are very promising.
Medienart: |
E-Artikel |
---|
Erscheinungsjahr: |
2017 |
---|---|
Erschienen: |
2017 |
Enthalten in: |
Zur Gesamtaufnahme - volume:95 |
---|---|
Enthalten in: |
Neural networks : the official journal of the International Neural Network Society - 95(2017) vom: 01. Nov., Seite 10-18 |
Sprache: |
Englisch |
---|
Beteiligte Personen: |
ElAdel, Asma [VerfasserIn] |
---|
Links: |
---|
Themen: |
Deep Convolution Neural Network |
---|
Anmerkungen: |
Date Completed 23.04.2018 Date Revised 10.12.2019 published: Print-Electronic Citation Status MEDLINE |
---|
doi: |
10.1016/j.neunet.2017.07.015 |
---|
funding: |
|
---|---|
Förderinstitution / Projekttitel: |
|
PPN (Katalog-ID): |
NLM275158322 |
---|
LEADER | 01000naa a22002652 4500 | ||
---|---|---|---|
001 | NLM275158322 | ||
003 | DE-627 | ||
005 | 20231225005041.0 | ||
007 | cr uuu---uuuuu | ||
008 | 231225s2017 xx |||||o 00| ||eng c | ||
024 | 7 | |a 10.1016/j.neunet.2017.07.015 |2 doi | |
028 | 5 | 2 | |a pubmed24n0917.xml |
035 | |a (DE-627)NLM275158322 | ||
035 | |a (NLM)28843091 | ||
035 | |a (PII)S0893-6080(17)30178-8 | ||
040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
041 | |a eng | ||
100 | 1 | |a ElAdel, Asma |e verfasserin |4 aut | |
245 | 1 | 0 | |a Fast DCNN based on FWT, intelligent dropout and layer skipping for image retrieval |
264 | 1 | |c 2017 | |
336 | |a Text |b txt |2 rdacontent | ||
337 | |a ƒaComputermedien |b c |2 rdamedia | ||
338 | |a ƒa Online-Ressource |b cr |2 rdacarrier | ||
500 | |a Date Completed 23.04.2018 | ||
500 | |a Date Revised 10.12.2019 | ||
500 | |a published: Print-Electronic | ||
500 | |a Citation Status MEDLINE | ||
520 | |a Copyright © 2017 Elsevier Ltd. All rights reserved. | ||
520 | |a Deep Convolutional Neural Network (DCNN) can be marked as a powerful tool for object and image classification and retrieval. However, the training stage of such networks is highly consuming in terms of storage space and time. Also, the optimization is still a challenging subject. In this paper, we propose a fast DCNN based on Fast Wavelet Transform (FWT), intelligent dropout and layer skipping. The proposed approach led to improve the image retrieval accuracy as well as the searching time. This was possible thanks to three key advantages: First, the rapid way to compute the features using FWT. Second, the proposed intelligent dropout method is based on whether or not a unit is efficiently and not randomly selected. Third, it is possible to classify the image using efficient units of earlier layer(s) and skipping all the subsequent hidden layers directly to the output layer. Our experiments were performed on CIFAR-10 and MNIST datasets and the obtained results are very promising | ||
650 | 4 | |a Journal Article | |
650 | 4 | |a Deep Convolution Neural Network | |
650 | 4 | |a Fast Beta wavelet transform | |
650 | 4 | |a Image classification & retrieval | |
650 | 4 | |a Intelligent dropout | |
700 | 1 | |a Zaied, Mourad |e verfasserin |4 aut | |
700 | 1 | |a Amar, Chokri Ben |e verfasserin |4 aut | |
773 | 0 | 8 | |i Enthalten in |t Neural networks : the official journal of the International Neural Network Society |d 1996 |g 95(2017) vom: 01. Nov., Seite 10-18 |w (DE-627)NLM087746824 |x 1879-2782 |7 nnns |
773 | 1 | 8 | |g volume:95 |g year:2017 |g day:01 |g month:11 |g pages:10-18 |
856 | 4 | 0 | |u http://dx.doi.org/10.1016/j.neunet.2017.07.015 |3 Volltext |
912 | |a GBV_USEFLAG_A | ||
912 | |a GBV_NLM | ||
951 | |a AR | ||
952 | |d 95 |j 2017 |b 01 |c 11 |h 10-18 |