With the use of deep learning networks image processing tasks has improved due to the development of learning feature illustration from images. Generally, in the real world scenario, these images available to classify is prone to noise and other deformities. According to many types of research in the past, the deep neural networks (DNNs) are found effective for image classification problems, but they suffer from the same real-life problem of noise and other deformities in an image. Noise is common occurrences in real life situations and many studies have been carried out in the past few decades with the purpose to remove the effect of noise in the image data. In this paper, the aim was to examine the DNN-based improved noisy image classification model. We have used a hybrid of denoising autoencoder, convolutional denoising autoencoder then using a classifier which is a combination of two different architectures one is Convolutional Neural Network (CNN) and the other is extreme Gradient Boosting (XGBOOST). This technique gives progressively better outcome by incorporating CNN as a trainable element for feature extraction from the image in input and XGBoost used as an identifier at the last stage of the model for outcomes.

Blurry Images, Image Classification, Noisy Images, Supervised Classification, Unsupervised Classification, Image Denoising

[1] Y. LeCun, L. Bottou, Y. Bengio, and P. Haffner, ‘‘Gradient-based learning applied to document recognition,’’ Proc. IEEE, vol. 86, no. 11, pp. 2278--2324, Nov. 1998.
[2] A. Krizhevsky, I. Sutskever, and G. Hinton, ‘‘ImageNet classification with deep convolutional neural networks,’’ in Proc. NIPS, 2012, pp. 1097--1105
[3] J. Schmidhuber, ‘‘Deep learning in neural networks: An overview,’’ Neural Netw., vol. 61, pp. 85--117, Jan. 2015.
[4] Mayur Thakur, Prof. S. K. Pillai” A Review on Various Methods for Classification of Massive Noisy Image”, 2ndInternational Conference on Intelligent Sustainable Systems (ICISS 2019)
[5] Scherer, D., M¨ uller, A., Behnke, S.:” Evaluation of pooling operations in convolutional architectures for object recognition”. In Proc of International Conference on Artificial Neural Networks ,2010.
[6] Mackey, L., Bryan, J., Mo, M.Y.: Weighted classification cascades for optimizing discovery significance in the HIGGSML challenge. In: NIPS 2014 Workshop on High-energy Physics and Machine Learning, pp. 129–134 (2015).
[7] Schroff, F., Kalenichenko, D., Philbin, J.: Facenet: a unified embedding for face recognition and clustering. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 815–823 (2015).
[8] Song, R., Chen, S., Deng, B., Li, L.: eXtreme gradient boosting for identifying individual users across different digital devices. In: Cui, B., Zhang, N., Xu, J., Lian, X., Liu, D. (eds.) WAIM 2016, Part I. LNCS, vol. 9658, pp. 43–54. Springer, Cham (2016). doi:10.1007/978- 3-319-39937-9_4
[9] Bekkerman, R.: The present and the future of the KDD cup competition: an outsider’s perspective
[10] Perona, Pietro, and Jitendra Malik. ”Scale-space and edge detection using anisotropic diffusion.” IEEE Transactions on pattern analysis and machine intelligence , vol. 12, pp 629-639, 1990.
[11] Rudin, Leonid I., and Stanley Osher. ’’Total variation based image restoration with free local constraints.’’ , In Proc of Image Processing, 1994. Proceedings. ICIP-94., IEEE International Conference. Vol. 1. Nov, 1994
[12] Subakan, Ozlem, et al. ’’Feature preserving image smoothing using a continuous mixture of tensors.’’ 2007 IEEE 11th International Conference on Computer Vision. Oct. 2007.
[13] Coifman, Ronald R., and David L. Donoho. Translation-invariant denoising. Springer New York, vol. 103, pp. 125-150, 1995.
[14] Elad, Michael, and Michal Aharon. ’’Image denoising via sparse and redundant representations over learned dictionaries.’’ IEEE Transactions on Image processing , vol. 15, pp. 3736-3745, 2006.
[15] Olshausen, Bruno A., and David J. Field. ’’Sparse coding with an overcomplete basis set: A strategy employed by V1?.’’ Vision research, vol. 37 pp. 3311-3325, 1997.
[16] Mairal, Julien, et al. ’’Online dictionary learning for sparse coding.’’ , In Proc. of the 26th annual international conference on machine learning. ACM, pp. 689-696 June 2009.
[17] Bengio, Yoshua, et al. ’’Greedy layer-wise training of deep networks.’’ , In Proc of Advances in 19 th neural information processing systems, pp. 153-160, Dec 2006.
[18] Glorot, Xavier, Antoine Bordes, and Yoshua Bengio. ’’Deep Sparse Rectifier Neural Networks.’’ , In Proc of the Fourteenth International Conference on Artificial Intelligence and Statistics (AISTATS-11). Vol. 15. pp. 315-323, 2011.
[19] Jain, Viren, and Sebastian Seung. ’’Natural image denoising with convolutional networks.’’ , In Proc of Advances in Neural Information Processing Systems 21. pp. 769-776, 2008.
[20] Burger, Harold C., Christian J. Schuler, and Stefan Harmeling. ’’Image denoising: Can plain neural networks compete with BM3D?.’’ In Proc of Computer Vision and Pattern Recognition (CVPR), 2012 IEEE Conference on. IEEE, June 2012.
[21] Vincent, Pascal, et al. ’’Extracting and composing robust features with denoising autoencoders.’’, In Proc. of the 25th international conference on Machine learning. ACM, pp. 1096-1103, July 2008.
[22] Lecun, Y., Boser, B., Denker, J.S., et al.: Handwritten digit recognition with a back-propagation network. In: Advances in Neural Information Processing Systems, pp. 396–404 (1990)
[23] Masci, Jonathan, et al. ’’Stacked convolutional auto-encoders for hierarchical feature extraction.’’ , In Proc. International Conference on Artificial Neural Networks. Springer Berlin Heidelberg, pp. 52-59, 2011.
[24] Xudie Ren, Haonan Guo, Shenghong Li, Shilin Wang and Jianhua Li. A Novel Image Classification Method with CNN-XGBoost Model.’’, C. Kraetzer et al. (Eds.): IWDW 2017, LNCS 10431, pp. 378–390, 2017.
[25] Sudipta Singha Roy, Mahtab Ahmed and M. A. H. Akhand, ‘‘Classification of Massive Noisy Image Using Autoencoders and Convolutional Neural Network’’, In 2017 8th International Conference on Information Technology (ICIT), pp. 971-979, 2017.
[26] Roy, S. S., Ahmed, M., & Akhand, M. A. H. (2018). Noisy image classification using hybrid deep learning methods. Journal of Information and Communication Technology, 17 (2), 233–269.