TY - GEN
T1 - Multi-Scale Convolutional Neural Network for Remote Sensing Scene Classification
AU - Alhichri, Haikel
AU - Alajlan, Naif
AU - Bazi, Yakoub
AU - Rabczuk, Timon
N1 - Funding Information:
The authors extend their appreciation to the Distinguished Scientist Fellowship Program (DSFP) at King Saud University for funding this work.
Publisher Copyright:
© 2018 IEEE.
PY - 2018/10/18
Y1 - 2018/10/18
N2 - In recent years the problem of scene classification in remote sensing has attracted a considerable amount of attention. Solution for this important problem based on deep convolutional neural networks (CNN) are currently state-of-the-art. So far all CNNs used images of fixed size (typically 224× 224 which commonly used in other fields of computer vision). In this paper, we propose a multi-scale deep CNN architecture that can accept variable image sizes. We achieve this by using multiple CNN, that share some or all parameters, followed by a merge layer, fully connected layers, and finally a softmax layer for classification. In each epoch we train the network with a batch of images of all scales. We have implemented this architecture using three SqueezeNet CNNs trained on three different scales of scene images. Then we carried out experiments on three well know datasets, namely UC Merced, KSA, and AID datasets. Preliminary results show that this multi-scale CNN do converge just as the traditional single-scale training, and leads to better testing accuracy.
AB - In recent years the problem of scene classification in remote sensing has attracted a considerable amount of attention. Solution for this important problem based on deep convolutional neural networks (CNN) are currently state-of-the-art. So far all CNNs used images of fixed size (typically 224× 224 which commonly used in other fields of computer vision). In this paper, we propose a multi-scale deep CNN architecture that can accept variable image sizes. We achieve this by using multiple CNN, that share some or all parameters, followed by a merge layer, fully connected layers, and finally a softmax layer for classification. In each epoch we train the network with a batch of images of all scales. We have implemented this architecture using three SqueezeNet CNNs trained on three different scales of scene images. Then we carried out experiments on three well know datasets, namely UC Merced, KSA, and AID datasets. Preliminary results show that this multi-scale CNN do converge just as the traditional single-scale training, and leads to better testing accuracy.
KW - Convolutional neural networks (CNN)
KW - Deep learning
KW - Deep neural networks
KW - Remote sensing
KW - Scene classification
UR - http://www.scopus.com/inward/record.url?scp=85057107188&partnerID=8YFLogxK
U2 - 10.1109/EIT.2018.8500107
DO - 10.1109/EIT.2018.8500107
M3 - Conference contribution
AN - SCOPUS:85057107188
T3 - IEEE International Conference on Electro Information Technology
SP - 113
EP - 117
BT - 2018 IEEE International Conference on Electro/Information Technology, EIT 2018
PB - IEEE Computer Society
T2 - 2018 IEEE International Conference on Electro/Information Technology, EIT 2018
Y2 - 3 May 2018 through 5 May 2018
ER -