High‐density impulse noise detection and removal using deep convolutional neural network with particle swarm optimisation

Khaw, Hui Ying; Soon, Foo Chong; Chuah, Joon Huang; Chow, Chee Onn

doi:10.1049/iet-ipr.2018.5776

Cited by 31 publications

(13 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This mechanism shows significantly better results in terms of edge preservation and noise suppression. Khaw et al [74] have used an efficient CNN with particle swarm optimization (PSO) for high-density impulse noise removal. This high-density impulse noise detection and removal model mainly consists of two parts: impulse noise removal and impulse noisy pixel detection for restoration.…”

Section: B Methodologies Of Cnn-based Models (Impulse Noise)mentioning

confidence: 99%

Image De-Noising With Machine Learning: A Review

et al. 2021

View full text Add to dashboard Cite

Images are susceptible to various kinds of noises, which corrupt the pictorial information stored in the images. Image de-noising has become an integral part of the image processing workflow. It is used to attenuate the noises and accentuate the specific image information stored within. Machine learning is an important tool in the image-de-noising workflow in terms of its robustness, accuracy, and time requirement. This paper explores the numerous state-of-the-art machine-learning-based image de-noisers like dictionary learning models, convolutional neural networks and generative adversarial networks for a range of noises like Gaussian, Impulse, Poisson, Mixed and Real-World noises. The motivation, algorithm and framework of different machine learning de-noisers are analyzed. These de-noisers are compared using PSNR as quality assessment metric on some benchmark datasets. The best de-noising results for different noise type is discussed along with future prospects. Among various Gaussian noise de-noisers, GCBD, BRDNet and PReLU network prove to be promising. CNN+LSTM, and MC2RNet are most suitable CNNbased Poisson de-noisers. For impulse noise removal, Blind CNN, and CNN+PSO perform well. For mixed noise removal, WDL, EM-CNN, CNN, SDL, and Mixed CNN are prominent. De-noisers like GRDN and DDFN show accurate results in the domain of real-world de-noising.

show abstract

Section: B Methodologies Of Cnn-based Models (Impulse Noise)mentioning

confidence: 99%

Image De-Noising With Machine Learning: A Review

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Huang et al [31] utilized Laplacian scale mixture (LSM) modeling and nonlocal low-rank regularization to remove mixture noise efficiently. In [32], a deep convolutional neural network (CNN) architecture and particle swarm optimization are adopted to detect impulse noise pixels accurately, and the median Filter is used to clean noise pixels. Wang et al [33] introduced a low-rank prior in small oriented noise-free image patches, then integrated low-rank and sparse matrix recovery to detect and remove nonpointwise random-valued impulse noise simultaneously.…”

Section: Related Workmentioning

confidence: 99%

Noise Removal in Embedded Image With Bit Approximation

Zhang

Tang

et al. 2022

IEEE Trans. Knowl. Data Eng.

View full text Add to dashboard Cite

Stego-images are often contaminated by interchannel noise or active noise attack when communicating on the Web. And it is challenging to restore embedded image from corrupted stego-image. This paper studies a kNN-bit approximation algorithm to remove noises in embedded image. The proposed algorithm distinguishes reliable bits from extracted bits, and estimates pixel values by keeping reliable bits unchanged and correcting unreliable bits. Specifically, the 8 th (highest) unreliable bit of a pixel can be approximated with its nearest neighbor pixels. And then, if an unreliable bit locates at any one of the 5 th ∼ 7 th bits of a pixel, it is adjusted with two nearest neighbors of the pixel, where the pixel is in-between these two nearest neighbors. Finally, for other unreliable bits, each one is approximated by the maximum and minimum possible values of nearest neighbors of its pixel. We conduct experiments for illustrating the efficiency, and demonstrate that the proposed algorithm can recover the embedded images with good visual quality from corrupted stego-images.

show abstract

“…Case II: If N8 has 0 or 255 greater than one then expand window size to 5 × 5 to check for any possible edge 16 figure 1: 5 × 5 window Let the pixel under processing is "v 0 " as shown in fig 1.The adjacent pixels are denoted as v 1 to v 24 . The following rules are utilized to cross-check whether the pixel needs to be restored or not.…”

Section: If Count1 ≥ 3 or Count2 ≥ 3 Then Take A 3 × 3 Windowmentioning

confidence: 99%

“…S. Schulte et.al.utilized the image histogram to find out impulse noise in color images [14].In addition with the histogram, fuzzy K-means clustering along with fuzzy-support vector machine (FSVM) classifier is used in [15].Many histogram-based fuzzy filters and fuzzy median filters perform effectively in terms of detection of impulse noise but proper restoration of noisy pixel is not achieved.It results in high false alarm(FA) rates.Edges present in an image are often detected as impulse noise which gives rise to greater miss detection(MD) rates. In the recent years, various neuro-fuzzy filters along with evolutionary algorithms have evolved to deal with impulse noise and edge detection such as techniques based on Particle swarm optimization(PSO) [16,17],ant colony optimization [18] and bacterial foraging [19,20] by Verma et.al. Neuro-fuzzy filter with optimized intelligent water drop technique presented by Devi and Soranamageswari in [21] and many more.Neuro-fuzzy filters provide fair results but are complex in nature with increased computational time and cost.…”

Section: Introductionmentioning

confidence: 99%

Edge Preserving Fuzzy Filter for Suppression of Impulse Noise in Images

2019

ijeat

View full text Add to dashboard Cite

Impulse noise in images culminates to the loss of valuable information. This paper proposes an efficient fuzzy filter for suppression of impulse noise in images (EFFSIN). Firstly, impulse noise is effectively detected by the proposed algorithm with the utilization of Fuzzy C-Means (FCM) in two stages.Only legitimate pixels are restored after fair distiction of noisy pixel from edge pixel. Proper restoration of noisy pixel following effective detection by proposed work leads to low miss detection(MD) and false alarm (FA) rates. Simulation results depict the efficacy of the proposed filter with reference to peaksignal-to-noise ratio (PSNR) and structural similarity index measure (SSIM).

show abstract

High‐density impulse noise detection and removal using deep convolutional neural network with particle swarm optimisation

Cited by 31 publications

References 26 publications

Image De-Noising With Machine Learning: A Review

Image De-Noising With Machine Learning: A Review

Noise Removal in Embedded Image With Bit Approximation

Edge Preserving Fuzzy Filter for Suppression of Impulse Noise in Images

Contact Info

Product

Resources

About