Color image denoising with wavelet thresholding based on human visual system model

Huang, Kai; Wu, Zhen; Fung, George S. K.; Chan, F.H.Y.

doi:10.1016/j.image.2004.08.002

Cited by 43 publications

(22 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…There has been significant research on image quality metrics [21,44,45], which often aim at the creation and optimization of encoding/ compression/decoding algorithms such as MEPG2 and MPEG4. In our enhancement algorithm, we want to produce the best looking image rather than achieve luminance and color fidelity.…”

Section: Color Image Quality Metricsmentioning

confidence: 99%

“…Numerous methods are available in the literature for texture details enhancement [2][3][4][5][6]. These methods can be classified into denoising-based enhancement and contrastbased enhancement.…”

Section: Introductionmentioning

confidence: 99%

“…These methods can be classified into denoising-based enhancement and contrastbased enhancement. Deniosing-based approaches [4,6] give good performances upon gray-level images, while they normally cannot work well for color image enhancement. There are also many algorithms based on contrast-based enhancement, which has two major groups: (1) spatial uniform ones: the linear contrast stretch approach (also called windowing and leveling) and the histogram equalization approach are two of the most widely used spatial uniform contrast-based enhancement technique and (2) spatial nonuniform ones: adaptive histogram equalization (AHE), contrast-limited adaptive histogram equalization (CLAHE), and multiscale enhancement [2] are most representative spatial non-uniform contrast-based enhancement methods.…”

Section: Introductionmentioning

confidence: 99%

“…We have been investigating the image processing techniques based on HVS properties [3,21,22]. In this paper, based on HVS models, a novel color image enhancement algorithm, Human visual system Controlled Color Image Enhancement and Evaluation (HCCIEE) algorithm, is presented.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Natural color image enhancement and evaluation algorithm based on human visual system

Huang

Wang

2006

Computer Vision and Image Understanding

Self Cite

152

View full text Add to dashboard Cite

Section: Color Image Quality Metricsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Natural color image enhancement and evaluation algorithm based on human visual system

Huang

Wang

2006

Computer Vision and Image Understanding

Self Cite

152

View full text Add to dashboard Cite

“…Real-life digital data sets are often contaminated by noise, either because of the data acquisition process or because of a naturally occurring phenomena [1]. In order to obtain the best possible estimate from noisy data, a robust, efficient and reliable denoising method is required.…”

Section: Introductionmentioning

confidence: 99%

An ICI Based Algorithm for Fast Denoising of Video Signals

2014

View full text Add to dashboard Cite

In this paper, we have proposed a fast method for video denoising using the modified intersection of confidence intervals (ICI) rule, called fast ICI (FICI) method. The goal of the new FICI based video denoising method is to maintain an acceptable quality level of the denoised video estimate, and at the same time to significantly reduce denoising execution time when compared to the original ICI based method. The methods are tested on real-life video signals and their performances are analyzed and compared.It is shown that the FICI method outperforms the ICI method in terms of the execution time reduction by up to 96% (or up to 25 times). However, practical application demands dictate the choice of the video denoising method. If one wants fast denoising method with decent denoising results, the FICI based video denoising method is a better choice. The original ICI method, however, should be used in applications where significant noise suppression is an imperative regardless the computational complexity. Key words: adaptive and temporal filtering, edge preserving, intersection of confidence intervals (ICI) rule, video denoisingAlgoritam za brzo uklanjanje šuma iz video signala temeljen na ICI postupku. U ovom smo radu predložili brzi postupak za uklanjanje šuma iz video signala koristeći modificirano pravilo presjecišta intervala pouzdanosti (eng. intersection of confidence intervals -ICI), nazvano brzim ICI (eng. fast ICI -FICI) postupkom. Cilj novog FICI postupka za uklanjanje šuma iz video signala jest da se, uz zadržavanje prihvatljive razine kvalitete odšuml-jenog video signala, značajno smanji vrijeme izvršavanja algoritma u usporedbi s izvornim ICI postupkom. Postupci su testirani na realnim video signalima, a njihove su performanse analizirane i usporeîene.Pokazano je da FICI postupak ima do 96% kraće vrijeme izvršavanja (odnosno kraće i do 25 puta) u usporedi s izvornim ICI postupkom. Meîutim, zahtjevi praktične primjene odreîuju izbor postupka za uklanjanje šuma iz video signala. Ukoliko je potrebno brzo izvršavanje s pristojnim performansama uklanjanja šuma, FICI postupak je bolji izbor. Meîutim, u aplikacijama kojima je imperativ značajno suzbijanje šuma bez obzira na računsku složenost, trebao bi se koristiti izvorni ICI postupak.Ključne riječi: adaptivno i vremensko filtriranje, očuvanje rubova, metoda presjecišta intervala pouzdanosti, uklanjanje šuma iz video signala

show abstract