@Article{cmc.2020.011706,
AUTHOR = {Rongyu Chen, Lili Pan, Cong Li, Yan Zhou, Aibin Chen, Eric Beckman},
TITLE = {An Improved Deep Fusion CNN for Image Recognition},
JOURNAL = {Computers, Materials \& Continua},
VOLUME = {65},
YEAR = {2020},
NUMBER = {2},
PAGES = {1691--1706},
URL = {http://www.techscience.com/cmc/v65n2/39900},
ISSN = {1546-2226},
ABSTRACT = {With the development of Deep Convolutional Neural Networks (DCNNs), the 
extracted features for image recognition tasks have shifted from low-level features to the
high-level semantic features of DCNNs. Previous studies have shown that the deeper the 
network is, the more abstract the features are. However, the recognition ability of deep 
features would be limited by insufficient training samples. To address this problem, this 
paper derives an improved Deep Fusion Convolutional Neural Network (DF-Net) which 
can make full use of the differences and complementarities during network learning and 
enhance feature expression under the condition of limited datasets. Specifically, DF-Net 
organizes two identical subnets to extract features from the input image in parallel, and 
then a well-designed fusion module is introduced to the deep layer of DF-Net to fuse the 
subnet’s features in multi-scale. Thus, the more complex mappings are created and the 
more abundant and accurate fusion features can be extracted to improve recognition 
accuracy. Furthermore, a corresponding training strategy is also proposed to speed up the 
convergence and reduce the computation overhead of network training. Finally, DF-Nets 
based on the well-known ResNet, DenseNet and MobileNetV2 are evaluated on 
CIFAR100, Stanford Dogs, and UECFOOD-100. Theoretical analysis and experimental 
results strongly demonstrate that DF-Net enhances the performance of DCNNs and 
increases the accuracy of image recognition.},
DOI = {10.32604/cmc.2020.011706}
}