 Research
 Open Access
A novel image denoising method based on spherical coordinates system
 Degan Zhang^{1, 2},
 Xuejing Kang^{1, 2}Email author and
 Jinghui Wang^{1, 2}
https://doi.org/10.1186/168761802012110
© Zhang et al; licensee Springer. 2012
 Received: 28 December 2011
 Accepted: 16 May 2012
 Published: 16 May 2012
Abstract
In this article, a novel image denoising method is proposed. This method is based on spherical coordinates system. First, spherical transform is redefined in wavelet domain, and the properties of the spherical transform in wavelet domain are listed. Then, a new adaptive threshold in spherical coordinate system is presented. It has been proved based on Besov space norm theory. After that, a novel curve shrinkage function is proposed to overcome the limitation of the traditional shrinkage functions. The new function can reach and exceed the true value and enhance the edge of the image. Finally, the multiscale product in wavelet domain is introduced to spherical coordinates system. This article names the multiscale product in spherical coordinates system as MultiScale Norm Product. The experimental results compared the improved algorithm with other methods from the peak signaltonoise ratio, mean square error, and running time. The results indicate that improved algorithm is simple and effective.
Keywords
 wavelet translation
 image denoising
 multiscale product
 shrinkage function
 spherical coordinates.
1. Introduction
Noise produced in image acquisition and transmission will cause image quality decline. Therefore, image denoising is an important problem in the field of image processing [1–4]. The goal of image denoising is to recover the original image from such a noisy copy. The traditional denoising methods cause details blurring of the reconstructed image. Discrete wavelet transform (DWT) provides a fast, local, sparse, and decorrelates multiresolution analysis of signals. It can reconstruct highquality images [5]. In recent years, wavelet shrinkage algorithm for image denoising has widely been used in different fields [6–8].
The threshold and shrinkage function are the key factors in image denoising. They will determine the effectiveness of image denoising. Donoho [9, 10] presented a method named wavelet shrinkage and showed its obvious efficiency on signal denoising. They proved that wavelet shrinkage was nearly optimal over a wide rage of function classes and error criterions. They also provided an estimated function with smoothness not less than the original. With these advantages, the method began to be widely applied. However, the universal threshold $T=\sigma \sqrt{2\mathsf{\text{ln}}\left(N*M\right)}$ tended to set all the coefficients of image details to zero, especially when N approaches infinite. At the mean while, the standard deviation of noise must be preestimated. Bias existed between estimated value and real value will reduce the effect of image denoising. DeVore [11] and Feng et al. [12] believe the image space belongs to Besov space. In 1998, Chambolle et al. [13] proposed a threshold in Besov space.
The classic shrinkage functions are the hard and soft threshold functions [9, 10]. Many researchers have proposed improved methods to optimize the classic algorithm. Khare et al. [14] proposed a multiscale adaptive softthreshold function and achieved a good denoising effect. However, in this method, the wavelet coefficients were processed directly. It will cause distortion of the reconstructed image. Wang [15] introduced a hyperbolic shrinkage function. This function avoided processing the wavelet coefficients directly. However, the hyperbolic function could not achieve the true value of the coefficients and cause edge blurring. Li et al. [16] presented an improved softthreshold function. This function is more flexible than the classic soft and hard thresholds. However, the starting point of this function is origin. There will be some noise information mistakenly processed as signal information.
Bao and Zhang [17] and Meng et al. [18] used multiscale products to control the shrinkage of the wavelet coefficients. Multiscale products can fully use the correlation of wavelet coefficients in different scales. Inspired by this idea, multiscale norm product (MSNP) is defined in this article to control the shrinkage of the wavelet coefficients.
This article proposes a new threshold after analyzing the relationship between wavelet coefficients norm, Besov space norm, and radial component in the spherical coordinates, and a new curve shrinkage function is proposed to achieve good denoising effect. The coefficients estimated by new function can reach and exceed the real value. It will avoid the edge blurring and enhance the coefficients of contours. Experimental results show that the improved method in this article retains the advantage of spherical coordinates. This new method cannot only be provided with effectively denoising, but also have better preserve image details and enhance the image edge information.
2. Traditional wavelet shrinkage algorithm
where ε_{ i } is independent and identically distributed (i.i.d) as N(0,1), σ refers to the noise level.
The idea of the wavelet shrinkage denoising is large wavelet coefficients reserve the useful information of image, while smaller coefficients contain the noise information which should be abandoned. The steps of the wavelet denoising algorithm are

To decompose noising image in wavelet domain;

Use threshold T and shrinkage function to process the highfrequency coefficients.
where w(x, y) denotes wavelet coefficients, w*(x, y) is wavelet coefficients after processing, T is the threshold, sign(·)is sign function.

To process the lowfrequency and highfrequency coefficients by inverse wavelet transform and reconstruct the image.
3. Spherical transform redefined in wavelet domain
Skowronski [19] firstly introduced spherical coordinates system into wavelet domain. Wang [15] extended this idea. However, he did not propose the properties of spherical transform in wavelet domain, neither define a proper threshold in spherical coordinates system.
This article redefines the spherical transform in wavelet domain and lists its properties. Set f denotes a twodimensional image, ${C}_{j,k}^{\left(i\right)},\phantom{\rule{1em}{0ex}}i=1,2,3$ is the highfrequency parts of the image f after wavelet decomposing.
Skowronski [19] only expressed the Haar coefficients in spherical coordinates and the radial component came from the variance of a block. Compared with the define of Skowronski, in this article, the wavelet coefficients are mapped to the components in spherical coordinate system. This define is reversible and exactly establishes onetoone correspondence between wavelet coefficients and components in spherical coordinate system. The properties of spherical transform in wavelet domain lists as follows:
Property 1: The image energy in the three highfrequency parts is completely mapped to radial component R of spherical coordinate system. The phase angle components θ, φ only contain the location information.
From Property 1, the energy in highfrequency parts is completely mapped to radial component. Therefore, we just need to process the radial component in spherical coordinate system. This idea can avoid processing the wavelet coefficients directly. It reduces the image distortion in a certain extent.
Property 2: Radial component R and the phase angle components θ, φ are mutually independent. The small angle does not always correspond to little energy.
Property 3: Spherical transform can further reduce the correlation among coefficients in the same scale.
Property 4: Spherical transform does not change the correlation among coefficients in different scales.
4. A novel threshold in spherical coordinates
DeVore [11] and Feng et al. [12] have proved the image space belong to Besov space. Chambolle et al. [13] proposed a threshold in Besov space. Based on Chambolle threshold, this article presents a novel shrinkage threshold which is optimal in spherical coordinates. The new threshold does not require preestimated noise intensity. It avoids the incomplete denoising problem caused by inaccurate estimates.
where N denotes image pixel, R is radial component in spherical coordinates.
 (a)
First, to establish equivalent relationship between Besov space norm ${\u2225F\u2225}_{{B}_{q}^{\beta}\left({L}_{p}\left(\mathrm{\Omega}\right)\right)}$ and wavelet coefficients norm $\left{C}_{j,k}^{\left(i\right)}\right,\phantom{\rule{1em}{0ex}}i=1,2,3$.
 (b)
Second, to establish equivalent relationship between the threshold T in Besov space and wavelet coefficients norm $\left{C}_{j,k}^{\left(i\right)}\right$, i = 1,2,3.
 (c)
Finally, to establish equivalent relationship between wavelet coefficients norm $\left{C}_{j1,k}^{\left(i\right)}\right$ and radial component norm R in spherical coordinates system.
The noise image f is decomposed on multiscale wavelet in this article. The highfrequency parts ${C}_{j1,k}^{\left(i\right)},\phantom{\rule{1em}{0ex}}i=1,2,3$ are mapped to θ, φ and radial component R in spherical coordinates system. Based on Properties 1 and 2, threshold T just needs to process radial component R. So, the equivalent relationship between radial component R and highfrequency ${C}_{j1,k}^{\left(i\right)}$ needs to be established.
where i = 1,2,3 k ≥ 0, j ∈ {0,1,....2 ^{ k } 1}^{2}.
where N denotes the number of image pixel.
 (1)
After the third or more decomposition of wavelet transform, the scaling factor is ${\eta}_{k}\le \frac{1}{16}$, therefore, it can simplify the process and advance the operation speed.
 (2)
Do not need to estimate the standard deviation of noise before image denoising, which will avoid the error existed between estimated value and real value and reduce the complexity of the algorithm.
5. MSNP in spherical coordinates system
 (1)
In the wavelet domain, with the increase of decomposition level, the wavelet coefficients of effective signal will increase or remain unchanged. While the wavelet coefficients of noise signal will decrease.
 (2)
Spherical transform reduces the relevance of scaling interior. However, it does not change the correlation of wavelet coefficients in different scales (Figure 2).
where K is the decomposing scales. The MSNP has the following properties:
Property 5: The MSNP in spherical coordinates must be positive value.
Property 6: "The larger (effective signal) the greater, the smaller (noise signal) the smaller". With the increase of the decomposition level K, we will get the conclusion ${p}_{{2}^{k}}\left(\mathsf{\text{signal}}\right)\xbb{p}_{{2}^{k}}\left(\mathsf{\text{noise}}\right)$.
The MSNP is controlled by the new threshold proposed in Section 4 in the image denoising. It can enlarge the distance between effective signal and noise signal. It avoids the disadvantage of the traditional methods that smaller wavelet coefficients are wrongly removed. Therefore, MSNP will effectively remove the additive noise without loss of the important features of the original image.
6. A new curve shrinkage function in spherical coordinates system
From the promise 1 mentioned in Section 5, the maximum amplitude wavelet coefficients must be the effective signal.
At present, the improve algorithms about shrinkage function are infinitely close to the true value. However, they cannot reach the true value, which will cause the edge blurry of reconstructed image.
where $\rho =1+\frac{1}{5\pi}\left\mathsf{\text{arctan}}\left(\leftR\left(i,j\right)\rightT\right)\right$ is the adjust factor of the curve shrinkage function, N, M are the image dimensions, i = 1,2,....,N, j = 1,2,....,M.
 (a)
When R (i, j) = T, it is easy to verify R*(i, j) = 0. So, the new function is continues at T.
 (b)When R (i, j) > T, the proposed function meets the following inequality:$R\left(i,j\right)T\le \leftR\left(i,j\right)\right{\left(1+\mathrm{log}\left(1+\leftR\left(i,j\right)\rightT\right)\right)}^{1}\cdot T\le R\left(i,j\right)$(28)
 (c)
As R (i, j)(1+ log(1+R(i, j)T))^{1}·T is nonlinearity function, and log(·)is a monotone increasing function, when R(i, j)→∞, the conclusion R*(i, j)→ R(i, j) will be established. Therefore, with the increase of R (i, j), the new curve function will infinite approximate to true value.
 (d)
The adjust factor $\rho =1+\frac{1}{5\pi}\left\mathsf{\text{arc}}\mathsf{\text{tan}}\left(\leftR\left(i,j\right)\rightT\right)\right$, ρ∈[1, 1.1] in the new function will raise the approach speed of the formula R*(i, j)→ R(i, j). Arctan(·) is a monotone continuous function, which can ensure R*(i, j) smooth approach to R (i, j). ρ∈[1, 1.1], as the signal amplitude increases, the estimated signal can reach and exceed the real value. Large radial component has a certain enhancement, and this increase is limited to < 110% of the original energy. Therefore, it can effectively preserve the edge of the signal, and avoid the local enhancement caused by the unexpected.
7. Algorithm design
Suppose f is the observed noise image. The steps of the improved algorithm:
Step 1. First, the image f is decomposed by wavelet transform.
Step 2. The high frequencies ${C}_{j,k}^{\left(i\right)}\mathsf{\text{,}}i=1,2,3$ of the image at different scales are transformed to spherical coordinates system by formula (4).
Step 3. Calculate the MSNP ${p}_{{2}^{k}}$
 (a)At the first decomposition, the interference brought by noise is strong, the improved curve shrinkage function proposed in this article is used to process the radial components:${R}^{*}\left(i,j\right)=\left\{\begin{array}{c}\rho \cdot \left(\leftR\left(i,j\right)\right{\left(1+\mathsf{\text{log}}\left(1+\leftR\left(i,j\right)\right{T}_{k}\right)\right)}^{1}\cdot {T}_{k}\right)R\left(i,j\right)\ge {T}_{k}\hfill \\ 0\phantom{\rule{18.8em}{0ex}}R\left(i,j\right)<{T}_{k}\hfill \end{array}\right.$(29)
 (b)
At the second or more decomposition of wavelet transform, the new threshold proposed in this article controls the MSNP for image denoising.
Step 5. If the estimated radial components are R*(i, j) = 0, set the angle θ, φ equal to zero.
where ${\left({C}_{j,k}^{\left(1\right)}\right)}^{*}$, ${\left({C}_{j,k}^{\left(2\right)}\right)}^{*}$, ${\left({C}_{j,k}^{\left(3\right)}\right)}^{*}$ are estimated horizontal high frequency, vertical high frequency, and diagonal high frequency, respectively.
Step 7. Finally, to realize denoising and reconstruct the image.
8. Experiments
To illustrate the effectiveness of this improved method, experimentation is done to use the image lena and the Internet of things (IOT) which sizes are 256*256.
8.1. The evaluation criteria
8.2. The choice of decomposition level
Denoising effects evaluation of different decomposing level
Level  PSNR  MSE  Time  

Lena  1  28.0877  100.9974  0.7438 
2  29.6917  69.8088  0.8150  
3  30.1217  63.2281  1.1094  
4  30.1198  63.2558  1.8175  
5  30.1198  63.2558  2.5694  
6  27.2008  123.8798  3.8712  
IOT  1  27.6682  111.2398  0.6987 
2  28.4816  92.2402  0.8150  
3  29.6977  69.1725  1.2526  
4  29.6977  69.1725  1.8725  
5  29.7011  69.6579  2.5694  
6  25.5632  180.6177  4.0925 
From Table 1, we can get the conclusion that the denoising effect is asymptotically stable after three levels. When the scale is greater than six levels, turbulence will generate in PSNR and MSE. Therefore, the best decomposition level is 35.
8.3. The denoising results adding Gaussian white noise
8.4. The comparison of running time
8.5. The denoising results adding salt&pepper noise
9. Conclusion
In this article, we use spherical coordinates system to remove the relevance of coefficients. Based on the Besov space norm theory, this article derives a new threshold in spherical coordinate system. MSNP is defined to make full use of the correlation in different decomposition scales. A new adaptive curve shrinkage function is constructed. This function can reach and exceed the true value. It largely avoids the edge blurry during image denoising. The new threshold controls MSNP for image denoising. Experimental results show that the proposed method visually superiors to other algorithms, and it is more effective in PSNR, MSE, and runtime.
Declarations
Acknowledgements
This study was supported by the National Natural Science Foundation of China (Nos. 61170173, 60773073, and 61001174), the Program for New Century Excellent Talents in University of China (No. NCET090895), the Key project of Ministry of Education of China (No. 208010), and the Tianjin Natural Science Foundation (No. 10JCYBJC00500).
Authors’ Affiliations
References
 Frandes M, Magnin I, Prost R: Wavelet thresholdingbased denoising method of listmode MLEM algorithm for compton imaging. IEEE Trans Nuclear Sci 2011, 58(3 Part 1):714723.View ArticleGoogle Scholar
 Yacov HO, Shaked D: A discriminative approach for wavelet denoising. IEEE Trans Image Process 2008, 17(4):443457.MathSciNetView ArticleGoogle Scholar
 Chen G, Qian S: Denoising of hyper spectral imagery using principal component analysis and wavelet shrinkage. IEEE Trans Geosci Remote Sens 2011, 49(3):973980.View ArticleGoogle Scholar
 Zhang D: A new approach and system for attentive mobile learning based on seamless migration. Appl Intell 2012, 36(1):7589. 10.1007/s1048901002450View ArticleMATHGoogle Scholar
 Yu H, Zhao L, Wang H: Image denoising using trivariate shrinkage filter in the wavelet domain and joint bilateral filter in the spatial domain. IEEE Trans Image Process 2009, 18(10):23642369.MathSciNetView ArticleGoogle Scholar
 Zhou D, Wang Q, Tian Q, Lin Q, Fu W: Wavelet analysis and its application in denoising the spectrum of hyperspectral image. Spectrosc Spectral Anal 2009, 29(7):19411945.Google Scholar
 Parrilli S, Poderico M, Angelino CV, Verdoliva L: A nonlocal SAR image denoising algorithm based on LLMMSE wavelet shrinkage. IEEE Trans Geosci Remote Sensing 2012, 50(2):606616.View ArticleGoogle Scholar
 Zhang D, Zhang X: A new serviceaware computing approach for mobile application with uncertainty. Appl Math Inf Sci 2012, 6(1):921.Google Scholar
 Donoho DL: Denoising by softthresholding. IEEE Trans Inf Theory 1995, 41(3):613627. 10.1109/18.382009MathSciNetView ArticleMATHGoogle Scholar
 Donoho DL, Johnstone IM: Ideal spatial adaptation via wavelet shrinkage. Biometrika 1994, 81: 425455. 10.1093/biomet/81.3.425MathSciNetView ArticleMATHGoogle Scholar
 DeVore RA: Nonlinear Approximation. Cambridge University Press, London; 1998:51150.Google Scholar
 Feng X, Jiang D, Xu G: Combining variation and wavelet transform for image zooming. Chin J Comput 2008, 31(2):340345.MathSciNetView ArticleGoogle Scholar
 Chambolle A, DeVore R, Lee N, Lucier B: Nonlinear wavelet image processing: variational problems, compression, and noise removal through wavelet shrinkage. IEEE Trans Image Process 1998, 7(3):319335.MathSciNetView ArticleMATHGoogle Scholar
 Khare A, Tiwary US, Pedrycz W, Jeon M: Multilevel adaptive thresholding and shrinkage technique for denoising using Daubechies complex wavelet transform. Imag Sci J 2010, 58(6):340358. 10.1179/136821910X12750339175826View ArticleGoogle Scholar
 Wang S, Zhou Y, Zou D: Nonlinear wavelet image compression coding algorithms based on spherical coordinates. J Infrared Millimet Waves 2001, 20(5):387390.Google Scholar
 Li C, Sun J, Kang Y: Adaptive image denoising by a new thresholding function. WiCOM 2010, 5600682, 1(1):15.Google Scholar
 Bao P, Zhang L: Noise reduction for magnetic resonance images via adaptive multiscale products thresholding. IEEE Trans Med Imag 2003, 22(9):10891099. 10.1109/TMI.2003.816958View ArticleGoogle Scholar
 Meng J, Pan Q, Zhang H: Optimal threshold determination for multiscale product in wavelet denoising. ISCIT 2005, 1566924, 1(1):570573.Google Scholar
 Skowronski J: A nonlinear transform for subband image coding. Signal Process 1999, 77: 289307. 10.1016/S01651684(99)000419View ArticleMATHGoogle Scholar
Copyright
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.