Deep pyramidal residual networks with inception sub-structure in image classification
Pages 5885 - 5906
Abstract
Deep convolutional neural networks (DCNNs) have shown remarkable performance in image classification tasks in recent years. In the network structure of DPRN, as the network depth increases, the number of convolutional kernels also increases linearly or nonlinearly. On the one hand, in the DPRN block, the size of the receptive field is only 3 × 3, which results in insufficient network ability to extract feature map information of different filter sizes. On the other hand, the number of convolution kernels in the second 1x1 convolution will be multiplied by a coefficient relative to the first convolution, which can cause overfitting to some extent. In order to overcome these weaknesses, we introduce the inception-like structure on the basis of the DPRN network which is called by pyramid inceptional residual networks (PIRN). In addition, we also discuss the performance of PIRN network with squeeze and excitation (SE) mechanism and regularization term. Furthermore, some results in network performance are discussed when adding a stochastic depth networkto the PIRN model. Compared to DPRN, PIRN achieved better results on the CIFAR10, CIFAR100, and Mini-ImageNet datasets. In the case of using zero-padding, the multiplicative PIRN with SE mechanism achieves the best result of 95.01% on the CIFAR10 dataset. Meanwhile, on the CIFAR100 and Mini-ImageNet datasets, the additive PIRN network with a network depth of 92 achieves the best results of 76.06% and 65.86%, respectively. According to the experimental results, our method has achieved better accuray than that of DPRN with same network settings which demonstrate its effectiveness in generalization ability.
References
[1]
LeCun Y., Bottou L., Bengio Y. and Haffner P., Gradient-based learning applied to document recognition, Proceedings of the IEEE 86(11) (1998), 1, 2278–2324.
[2]
Krizhevsky A., Sutskever I. and Hinton G.E., ImageNet Classification with Deep Convolutional Neural Networks. In NIPS (2012), 1,2.
[3]
Zeiler M.D. and Fergus R., Visualizing and understandingconvolutional networks. In ECCV, (2014), 1, 2.
[4]
Simonyan K. and Zisserman A., Very deep convolutional networks forlarge-scale image recognition. In ICLR (2015), 1,2,3,4,8.
[5]
Szegedy C., Liu W., Jia Y., Sermanet P., Reed S., Anguelov D., Erhan D., Vanhoucke V. and Rabinovich A., Going deeper with convolutions. In CVPR (2015), 1,2,8.
[6]
He K., Zhang X., Ren S. and Sun J., Deep residual learning for imagerecognition. In CVPR (2016), 1,2,3,4,5,6,7,8.
[7]
He K., Zhang X., Ren S. and Sun J., Identity mappings in deepresidual networks. In ECCV (2016), 1,2,3,4,5,6,7,8.
[8]
Szegedy C., Ioffe S. and Vanhoucke V., Inception-v4,inception-resnet and the impact of residual connections on learning.In ICLR Workshop, (2016), 1,8.
[9]
Lin M., Chen Q. and Yan S., Network in network, Computing Research Repository (CoRR) arXiv:1312.4400.
[10]
Veit A., Wilber M. and Belongie S., Residual networks behave likeensembles of relatively shallow networks. In NIPS (2016), 1,2,3,4.
[11]
Liu S., et al., A novel scene classification model combining ResNetbased transfer learning and data augmentation with a filter, Neurocosmputing 338(21) (2019), 191–206.
[12]
Lu Z., Bai Y., Chen Y., Su C., Lu S., Zhan T., Hong X. and Wang S., The classification of gliomas based on a Pyramid dilated convolution resnet model, Pattern Recognition Letters 133 (2020), 173–179.
[13]
Han D., Kim J. and Kim J., Deep Pyramidal Residual Networks, In CVPR 2017 arXiv:1610.02915.
[14]
Arora S., Bhaskara A., Ge R. and Ma T., Provable bounds for learning some deep representations, Proceedings of the 31st International Conference on Machine Learning 2014 arXiv:1310.6343.
[15]
Ioffe S. and Szegedy C., Batch normalization: Accelerating deepnetwork training by reducing internal covariate shift. InProceedings of the 32nd International Conference on Machine Learning, (2015), 448–456.
[16]
Szegedy C., Vanhoucke V., Ioffe S. and Shlens J., Rethinking the Inception Architecture for Computer Vision, arXiv:1512.00567.
[17]
Krizhevsky A., Sutskever I. and Hinton G.E., Imagenet classification with deep convolutional neural networks, Advances in Neural Information Processing Systems (2012), 1097–1105.
[18]
Zagoruyko S. and Komodakis N., Wide Residual Networks, In Proceedings of the British Machine Vision Conference (BMVC) 87 (2016), 1–12.
[19]
Huang G., Sun Y., Liu Z., Sedra D. and Weinberger K., Deep networks with stochastic depth. In ECCV (2016), 1,2,4,7,8.
[20]
Pierre Foret, Ariel Kleiner, Hossein Mobahi and Behnam Neyshabur, Sharpness-Aware Minimization for Efficiently Improving Generalization, In ICLR, 2021.
[21]
Chollet F., Xception: Deep learning with depthwise separable convolutions. arXiv:1610.02357v2, 2016. 1.
[22]
Wan L., Zeiler M., Zhang S., Cun Y.L. and Fergus R., Regularization of neural networks using dropconnect. In S. Dasgupta, D. Mcallester, eds.: Proceedings of the 30th International Conference on Machine Learning (ICML13). Volume 28., JMLRWorkshop and Conference Proceedings (May 2013), 1058–1066.
[23]
Yamada Y., Iwamura M., Akiba T. and Kise K., Shakedrop regularization for deep residual learning. arXiv:1802.02375v3 (2020).
[24]
Gastaldi X., Shake-Shake regularization, arXiv: 1705.07485 (2017).
[25]
Sutskever I., Martens J., Dahl G. and Hinton G., On the importance of initialization and momentum in deep learning. In Proceedings of the 30th International Conference on Machine Learning (ICML-13), (2013), 1139–1147.
[26]
Gross S. and Wilber M., Training and investigating residual nets (2016).
[27]
He K., Zhang X., Ren S. and Sun J., Delving deep into rectifiers: Surpassing human-level performance on imagenet classification. In ICCV (2015), 7.
[28]
Nair V. and Hinton G.E., Rectified linear units improve restricted boltzmann machines. In Proceedings of the 27th International Conference on Machine Learning (ICML-10), (2010), 807–814.
[29]
Howard A.G., Zhu M., Chen B., Kalenichenko D., Wang W., Weyand T., Andreetto M. and Adam H., Mobilenets: Efficient convolutional neural networks for mobile vision applications (2017).
[30]
Sandler M., Howard A., Zhu M., Zhmoginov A. and Chen L.-C., Mobilenetv2: Inverted residuals and linear bottlenecks. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (2018), 4510–4520.
[31]
Zhang X., Zhou X., Lin M. and Sun. J., Shufflenet: An extremely efficient convolutional neural network for mobile devices. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (2018).
[32]
Ma N., Zhang X., Zheng H.-T. and Sun. J., ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design. In Proceedings of European Conference on Computer Vision (2018).
[33]
Huang G., Liu S., van der Maaten L. and Weinbergeer K.Q., Condensenet: An efficient densenet using learned group convolutions. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (2018).
[34]
Xie S., Girshick R., Dollár P., Tu Z. and He K., Aggregated residual transformations for deep neural networks. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (2017), 5987–5995.
[35]
Tsung-Yi Lin, Piotr Dollár, Ross Girshick, Kaiming He, Bharath Hariharan and Serge Belongie, Feature Pyramid Networks for Object Detection. In Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, HI, USA (2017), 21–26.
[36]
Ionut Cosmin Duta, Liu L., Zhu F. and Shao L., Pyramidal Convolution: Rethinking Convolutional Neural Networks for Visual Recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (2020).
[37]
Sang H., Zhou Q. and Zhao. Y., PCANet: Pyramid convolutional attention network for semantic segmentation. Image and Vision Computing (2020).
[38]
Zheng F., Yan Q., Leung V.C.M., Yu F.R. and Ming Z., HDPCNN: Highway deep pyramid convolution neural network combining word-level and character-level representations for phishing website detection, Computers and Security (2022).
[39]
Chen K., Wang X. and Zhang S., Thorax Disease Classification Based on Pyramidal Convolution Shuffle Attention Neural Network, in IEEE Access (2022), 85571–85581.
[40]
Sun J., Yang K., Chen C., Shen J., Yang Y., Wu X. and Tomas Norton, Wheat head counting in the wild by an augmented feature pyramid networks-based convolutional neural network, Computers and Electronics in Agriculture (2022).
[41]
Chen Y., Zhu Y., Zhang Z., Wang J. and Wang C., Prediction of drug protein interactions based on variable scale characteristic pyramid convolution network, Methods (2023), 42–47.
[42]
Li F., Long Z., He P., Feng P., Guo X., Ren X., Wei B. and Zhao M., Fully Convolutional Pyramidal Networks for Semantic Segmentation, IEEE Access (2020), 229132–229140.
[43]
Yin L., Hong P., Zheng G., Chen H. and Deng W., A Novel Image Recognition Method Based on Dense Net and DPRN, Applied Sciences 12(9) (2022), 4232.
[44]
Zhou Q., Wang J., Guo J., Huang Z., Ding M., Yuchi M. and Zhang X., Anterior chamber angle classification in anterior segment optical coherence tomography images using hybrid attention based pyramidal convolutional network, Biomedical Signal Processing and Control (2021).
[45]
Zhang G., Zheng C., He J. and Yi S., PCT: Pyramid convolutional transformer for parotid gland tumor segmentation in ultrasound images, Biomedical Signal Processing and Control (2023).
[46]
Tan H. and Dong S., Pixel-level concrete crack segmentation usingpyramidal residual network with omni-dimensional dynamicconvolution, Processes 11(2) (2023), 546.
Index Terms
- Deep pyramidal residual networks with inception sub-structure in image classification
Index terms have been assigned to the content through auto-classification.
Recommendations
Improved inception-residual convolutional neural network for object recognition
AbstractMachine learning and computer vision have driven many of the greatest advances in the modeling of Deep Convolutional Neural Networks (DCNNs). Nowadays, most of the research has been focused on improving recognition accuracy with better DCNN models ...
Three-class brain tumor classification using deep dense inception residual network
AbstractThree-class brain tumor classification becomes a contemporary research task due to the distinct characteristics of tumors. The existing proposals employ deep neural networks for the three-class classification. However, achieving high accuracy is ...
Deep Learning Approaches for Image Classification
EITCE '22: Proceedings of the 2022 6th International Conference on Electronic Information Technology and Computer EngineeringDeep learning models can achieve a higher accuracy result compared with traditional machine learning algorithm. It is widely useful in different areas, especially in images classification area. In recent years, because of the improvement of hardware and ...
Comments
Please enable JavaScript to view thecomments powered by Disqus.Information & Contributors
Information
Published In
© 2023 – IOS Press. All rights reserved.
Publisher
IOS Press
Netherlands
Publication History
Published: 01 January 2023
Author Tags
Qualifiers
- Research-article
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- 0Total Citations
- 0Total Downloads
- Downloads (Last 12 months)0
- Downloads (Last 6 weeks)0
Reflects downloads up to 03 Mar 2025