Convolutional Neural Networks (CNNs) have emerged one of the most powerfully used tools in deep learning, particularly in image and pattern recognition tasks. CNNs are being revolutionized through deep learning resulting in a dominant architecture in computer vision as well as other various domains. This paper provides a comprehensive survey of CNN-based deep learning, its methodologies, regularizations, and computational efficiencies. We also address the challenges and future directions of CNN. This paper provides overview of CNNs. CNNs are differentiated into several layers: input layer, convolutional layer, activation layer, pooling layer, fully connected layer, and the output layer. Over time CNNs evolved significantly. Key methods of training CNNs include data augmentation, transfer layering, and batch normalization as well as additional techniques like dropout and early stopping to improve generalization. CNN has real world applications in security, medical imaging, traffic management, defence, agriculture and pattern recognition. This paper emphasizes the significance of CNN architecture, model evolution, training techniques, and practical use cases. Despite overall development, CNNs continue to evolve and will be even more intelligent and efficient as AI technology advances in future.

CNN (convolutional neural networks), artificial intelligence, real world applications, natural language processing (NLP), rectified linear unit

🔒 This is a subscription article

Full text is available to subscribers and institutional members. Please choose an option below to access it.

1. McCulloch WS, Pitts W. A logical calculus of the ideas immanent in nervous activity. Bull Math Biophys. 1943 Dec;5(4):115–33.
2. Rumelhart DE, Hinton GE, Williams RJ. Learning representations by back-propagating errors. Nature. 1986 Oct 9;323(6088):533–6.
3. Waibel A, Hanazawa T, Hinton G, Shikano K, Lang KJ. Phoneme recognition using time-delay neural networks. In: Backpropagation. New Jersey: Psychology Press; 2013. p. 35–61.
4. Zhang W, Tanida J, Itoh K, Ichioka Y. Shift-invariant pattern recognition neural network and its optical architecture. In: Proceedings of annual conference of the Japan Society of Applied Physics. 1988 Aug;564.
5. LeCun Y, Boser B, Denker JS, Henderson D, Howard RE, Hubbard W, et al. Backpropagation applied to handwritten zip code recognition. Neural Comput. 1989 Dec;1(4):541–51.
6. Aihara K, Takabe T, Toyoda M. Chaotic neural networks. Phys Lett A. 1990 Mar 12;144(6–7):333–40.
7. Specht DF. A general regression neural network. IEEE Trans Neural Netw. 1991 Nov 1;2(6):568–76.
8. LeCun Y, Bottou L, Bengio Y, Haffner P. Gradient-based learning applied to document recognition. Proc IEEE. 2002 Aug 6;86(11):2278–324.
9. Li Z, Liu F, Yang W, Peng S, Zhou J. A survey of convolutional neural networks: Analysis, applications, and prospects. IEEE Trans Neural Netw Learn Syst. 2021 Jun 10;33(12):6999–7019.
10. Krizhevsky A, Sutskever I, Hinton GE. Imagenet classification with deep convolutional neural networks. Adv Neural Inf Process Syst. 2012;25.

Convolutional Neural Networks (CNNs) have emerged one of the most powerfully used tools in deep learning, particularly in image and pattern recognition tasks. CNNs are being revolutionized through deep learning resulting in a dominant architecture in computer vision as well as other various domains. This paper provides a comprehensive survey of CNN-based deep learning, its methodologies, regularizations, and computational efficiencies. We also address the challenges and future directions of CNN. This paper provides overview of CNNs. CNNs are differentiated into several layers: input layer, convolutional layer, activation layer, pooling layer, fully connected layer, and the output layer. Over time CNNs evolved significantly. Key methods of training CNNs include data augmentation, transfer layering, and batch normalization as well as additional techniques like dropout and early stopping to improve generalization. CNN has real world applications in security, medical imaging, traffic management, defence, agriculture and pattern recognition. This paper emphasizes the significance of CNN architecture, model evolution, training techniques, and practical use cases. Despite overall development, CNNs continue to evolve and will be even more intelligent and efficient as AI technology advances in future.

CNN (convolutional neural networks), artificial intelligence, real world applications, natural language processing (NLP), rectified linear unit

🔒 This is a subscription article

Full text is available to subscribers and institutional members. Please choose an option below to access it.

1. McCulloch WS, Pitts W. A logical calculus of the ideas immanent in nervous activity. Bull Math Biophys. 1943 Dec;5(4):115–33.
2. Rumelhart DE, Hinton GE, Williams RJ. Learning representations by back-propagating errors. Nature. 1986 Oct 9;323(6088):533–6.
3. Waibel A, Hanazawa T, Hinton G, Shikano K, Lang KJ. Phoneme recognition using time-delay neural networks. In: Backpropagation. New Jersey: Psychology Press; 2013. p. 35–61.
4. Zhang W, Tanida J, Itoh K, Ichioka Y. Shift-invariant pattern recognition neural network and its optical architecture. In: Proceedings of annual conference of the Japan Society of Applied Physics. 1988 Aug;564.
5. LeCun Y, Boser B, Denker JS, Henderson D, Howard RE, Hubbard W, et al. Backpropagation applied to handwritten zip code recognition. Neural Comput. 1989 Dec;1(4):541–51.
6. Aihara K, Takabe T, Toyoda M. Chaotic neural networks. Phys Lett A. 1990 Mar 12;144(6–7):333–40.
7. Specht DF. A general regression neural network. IEEE Trans Neural Netw. 1991 Nov 1;2(6):568–76.
8. LeCun Y, Bottou L, Bengio Y, Haffner P. Gradient-based learning applied to document recognition. Proc IEEE. 2002 Aug 6;86(11):2278–324.
9. Li Z, Liu F, Yang W, Peng S, Zhou J. A survey of convolutional neural networks: Analysis, applications, and prospects. IEEE Trans Neural Netw Learn Syst. 2021 Jun 10;33(12):6999–7019.
10. Krizhevsky A, Sutskever I, Hinton GE. Imagenet classification with deep convolutional neural networks. Adv Neural Inf Process Syst. 2012;25.