Abstract
Noise present in an image is an active challenge in the field of Digital Image Processing. Most computer vision applications such as object detection and edge detection are heavily influenced by the presence of noise. The presence of noise mandates pre-processing. Determining the noise level (whether high or low) is a demanding field in current times. Salt and pepper noise is the kind of noise that is commonly caused by dirt debris at the capturing tool and appears as black and white dots in an image. This work proposes a model to identify the level of salt and pepper noise in an image using a deep convolutional neural network (CNN or ConvNet). Once the noise levels are calculated, an appropriate de-noising filter can be applied. The proposed model achieves 98 percent of classification accuracy for salt and pepper quantification task using a data-set proposed in this work, which contains images having a different level of salt and pepper noise.
Similar content being viewed by others
References
R.J. Beaton, Quantitative models of image quality, in Proceedings of the Human Factors Society Annual Meeting, Vol. 27. No. 1 (Sage, Los Angeles, 1983)
R. Verma, J. Ali, A comparative study of various types of image noise and efficient noise removal techniques. Int. J. Adv. Res. Comput. Sci. Softw. Eng. 3(10), 66 (2013)
A.J. Ahumada, Computational image quality metrics: a review. SID Dig. 24, 66 (1993)
Y. Lee, S. Kassam, Generalized median filtering and related nonlinear filtering techniques. IEEE Trans. Acoust. Speech Signal Process. 33(3), 66 (1985)
N. Gallagher, G. Wise, A theoretical analysis of the properties of median filters. IEEE Trans. Acoust. Speech Signal Process. 29, 6 (1981)
MSRA, Salient Object Image Data Set. https://mmcheng.net/en/msra10k/. Accessed 14 Oct 2021
M.-M. Cheng et al., Global contrast based salient region detection. IEEE Trans. Pattern Anal. Mach. Intell. 37(3), 66 (2014)
M.-M. Cheng, et al., Efficient salient region detection with soft image abstraction, in Proceedings of the IEEE International Conference on Computer Vision (2013)
A. Borji et al., Salient object detection: a benchmark. IEEE Trans. Image Process. 24(12), 66 (2015)
S.-J. Ko, Y.H. Lee, Center weighted median filters and their applications to image enhancement. IEEE Trans. Circuits Syst. 38(9), 66 (1991)
H.-L. Eng, K.-K. Ma, Noise adaptive soft-switching median filter. IEEE Trans. Image Process. 10(2), 66 (2001)
L. Hou, et al., Image deblurring in the presence of salt-and-pepper noise, in IEEE International Conference on Image Processing (ICIP) (IEEE, 2017)
Adam, et al., Method of adaptive pixel averaging for impulse noise reduction in digital images, in 2018 Baltic URSI Symposium (URSI) (IEEE, 2018)
M.R. Khammar, et al., Removal of high density salt and pepper noise from image and video based on optimal decision based algorithm, in 2014 2nd International Conference on Electronic Design (ICED) (IEEE, 2014)
A. Shams-Baboli, A.A. Shams-Baboli, A modified nonlinear filtering technique for removal of high density salt and pepper noise, in 2015 9th Iranian Conference on Machine Vision and Image Processing (MVIP) (IEEE, 2015)
B. Fu et al., A convolutional neural networks denoising approach for salt and pepper noise. Multimedia Tools Appl. 78(21), 66 (2019)
Y. Xing et al., Deep CNN for removal of salt and pepper noise. IET Image Process. 13(9), 66 (2019)
L. Liang et al., Convolutional neural network with median layers for denoising salt-and-pepper contaminations. Neurocomputing 442, 66 (2021)
S.C. Kumain, K. Kumar, VBNC: voting based noise classification framework using deep CNN, in International Conference on Deep Learning, Artificial Intelligence and Robotics (Springer, Cham, 2019)
S. Dey et al., Median filter aided CNN based image denoising: an ensemble approach. Algorithms 14(4), 109 (2021)
K. Radlak, L. Malinski, B. Smolka, Deep learning based switching filter for impulsive noise removal in color images. Sensors 20(10), 66 (2020)
J.H. Chuah, et al., Detection of Gaussian noise and its level using deep convolutional neural network, in TENCON 2017-2017 IEEE Region 10 Conference (IEEE, 2017)
Model Architectures: LeNet, AlexNet, VGG, GoogLeNet, ResNet and more. https://medium.com/@sidereal/cnns-architectures-lenet-alexnet-vgg-googlenet-resnet-and-more-666091488df5. Accessed 14 Oct 2021
N. Tajbakhsh et al., Convolutional neural networks for medical image analysis: full training or fine tuning? IEEE Trans. Med. Imaging 35(5), 66 (2016)
How to Choose an Activation Function for Deep Learning. https://machinelearningmastery.com/choose-an-activation-function-for-deep-learning/. Accessed 13 Oct 2021
Gentle Introduction to the Adam Optimization Algorithm for Deep Learning. https://machinelearningmastery.com/adam-optimization-algorithm-for-deep-learning/. Accessed 13 Oct 2021
Confusion Matrix in Machine Learning. https://www.geeksforgeeks.org/confusion-matrix-machine-learning/. Accessed 13 Oct 2021
Funding
The authors have not disclosed any funding.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors have not disclosed any competing interests..
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Kumain, S.C., Kumar, K. Quantifying Salt and Pepper Noise Using Deep Convolutional Neural Network. J. Inst. Eng. India Ser. B 103, 1293–1303 (2022). https://doi.org/10.1007/s40031-022-00729-3
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s40031-022-00729-3