基于组合ResNet和InceptionNet的神经网络分类研究
Research on Neural Network Classification Based on Combined ResNet and InceptionNet
DOI: 10.12677/CSA.2022.126168, PDF,  被引量   
作者: 李翔宇：天津工业大学计算机科学与技术学院，天津
关键词: ResNet模型；InceptionNet模型；图像分类；ResNet Model； InceptionNet Model； Image Classification
摘要: 针对基础的图像分类问题，我们分别使用主流的卷积神经网络进行训练，之后将ResNet网络和InceptionNet网络模块化，提取出ResNetBlock和InceptionBlock，建立新的卷积神经网络组合两个独立模块，并使用一些列的调参方法，和经典的卷积神经网络进行比较，验证的识别率高于原来的卷积神经网络，并且损失函数能降至更低的水平。
Abstract: For the basic image classification problem, we use the mainstream convolutional neural networks for training, and then modularize the ResNet network and the InceptionNet network, extract the ResNetBlock and InceptionBlock, build a new convolutional neural network to combine two independent modules, and use a series of parameter tuning methods, compared with the classic convolutional neural network, to verify that the recognition rate is higher than the original convolutional neural network, and the loss function can be reduced to a lower level.
文章引用：李翔宇. 基于组合ResNet和InceptionNet的神经网络分类研究[J]. 计算机科学与应用, 2022, 12(6): 1674-1684. https://doi.org/10.12677/CSA.2022.126168

参考文献

[1] 卢宏涛, 张秦川. 深度卷积神经网络在计算机视觉中的应用研究综述[J]. 数据采集与处理, 2016, 31(1): 1-17.
[2] Albawi, S., Mohammed, T.A. and Al-Zawi, S. (2017) Understanding of a Convolutional Neural Network. 2017 International Conference on Engineering and Technology (ICET), Antalya, 21-23 August 2017, 1-6. [Google Scholar] [CrossRef] 
[3] Liu, L., Shen, C. and van den Hengel, A. (2016) Cross-Convolutional-Layer Pooling for Image Recognition. IEEE Transactions on Pattern Analysis and Machine Intelli-gence, 39, 2305-2313. [Google Scholar] [CrossRef] 
[4] He, K., Zhang, X., Ren, S., et al. (2016) Deep Residual Learning for Image Recognition. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Ve-gas, 27-30 June 2016, 770-778. [Google Scholar] [CrossRef] 
[5] Szegedy, C., Liu, W., Jia, Y., et al. (2014) Going Deeper with Con-volutions. 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Boston, 7-12 June 2015, 1-9. [Google Scholar] [CrossRef] 
[6] Hafner, D., Davidson, J. and Vanhoucke, V. (2017) Tensor-Flow Agents: Efficient Batched Reinforcement Learning in TensorFlow. arXiv:1709.02878.
[7] Ho-Phuoc, T. (2018) CIFAR10 to Compare Visual Recognition Performance between Deep Neural Networks and Humans. arXiv:1811.07270.
[8] Sharma, N., Jain, V. and Mishra, A. (2018) An Analysis of Convolutional Neural Networks for Image Classification. Procedia Computer Science, 132, 377-384. [Google Scholar] [CrossRef] 
[9] Xiao, H., Rasul, K. and Vollgraf, R. (2017) Fashion-MNIST: A Novel Image Dataset for Benchmarking Machine Learning Algorithms. arXiv:1708.07747.
[10] Cherry, J.M., Adler, C., Ball, C., et al. (1998) SGD: Saccharomyces Genome Database. Nucleic Acids Research, 26, 73-79. [Google Scholar] [CrossRef] [PubMed]
[11] Ward, R., Wu, X. and Bottou, L. (2019) Adagrad Stepsizes: Sharp Con-vergence over Nonconvex Landscapes. Proceedings of the 36th International Conference on Machine Learning, Long Beach, 9-15 June 2019, 6677-6686.
[12] Zeiler, M.D. (2012) Adadelta: An Adaptive Learning Rate Method. arXiv:1212.5701.
[13] 杨观赐, 杨静, 李少波, 胡建军. 基于Dopout与ADAM优化器的改进CNN算法[J]. 华中科技大学学报(自然科学版), 2018, 46(7): 122-127. [Google Scholar] [CrossRef] 
[14] 蒋正锋, 廖群丽. 基于多参数融合优化的深度神经网络设计研究[J]. 现代计算机, 2021, 27(31): 13-24.
[15] Simonyan, K. and Zis-serman, A. (2014) Very Deep Convolutional Networks for Large-Scale Image Recognition. arXiv:1409.1556.
[16] Krizhevsky, A., Sutskever, I. and Hinton, G. (2012) ImageNet Classification with Deep Convo-lutional Neural Networks. Advances in Neural Information Processing Systems, 25, No. 2.
[17] Lecun, Y., Bottou, L., Bengio, Y. and Haffner, P. (1998) Gradient-Based Learning Applied to Document Recognition. Proceedings of the IEEE, 86, 2278-2324. [Google Scholar] [CrossRef] 
[18] 任进军, 王宁. 人工神经网络中损失函数的研究[J]. 甘肃高师学报, 2018, 23(2): 61-63.
[19] 解天舒. 基于卷积神经网络的Dropout方法研究[D]: [硕士学位论文]. 成都: 电子科技大学, 2021.[CrossRef] 
[20] 刘建伟, 赵会丹, 罗雄麟, 许鋆. 深度学习批归一化及其相关算法研究进展[J]. 自动化学报, 2020, 46(6): 1090-1120. [Google Scholar] [CrossRef] 
[21] 蒋明威, 邓明舟, 李振亚. 结合全局与局部池化的多幅图像识别算法[J]. 信息通信, 2019(8): 9-10.
[22] 吕国豪, 罗四维, 黄雅平, 等. 基于卷积神经网络的正则化方法[J]. 计算机研究与发展, 2014, 51(9): 1891-1900.
[23] 刘海章, 黄大池. 浅析激活函数SoftMax的设计与实现[J]. 西部广播电视, 2021, 42(17): 201-206.
[24] Lin, T.Y., Maire, M., Belongie, S., et al. (2014) Microsoft Coco: Common Objects in Context. European Conference on Computer Vision, Zurich, 6-12 September 2014, 740-755. [Google Scholar] [CrossRef]