Impulsive noise rejection method for compressed measurement signal in compressed sensing
 Parichat Sermwuthisarn^{1},
 Duangrat Gansawat^{2},
 Vorapoj Patanavijit^{3} and
 Supatana Auethavekiat^{1}Email author
https://doi.org/10.1186/16876180201268
© Sermwuthisarn et al; licensee Springer. 2012
Received: 15 September 2011
Accepted: 20 March 2012
Published: 20 March 2012
Abstract
The Lorentzian norm of robust statistics is often applied in the reconstruction of the sparse signal from a compressed measurement signal in an impulsive noise environment. The optimization of the robust statistic function is iterative and usually requires complex parameter adjustments. In this article, the impulsive noise rejection for the compressed measurement signal with the design for image reconstruction is proposed. It is used as the preprocessing for any compressed sensing reconstruction given that the sparsified version of the signal is obtained by utilizing octavetree discrete wavelet transform with db8 as the mother wavelet. The presence of impulsive noise is detected from the energy distribution of the reconstructed sparse signal. After the noise removal, the noise corrupted coefficients are estimated. The proposed method requires neither complex optimization nor complex parameter adjustments. The performance of the proposed method was evaluated on 60 images. The experimental results indicated that the proposed method effectively rejected the impulsive noise. Furthermore, at the same impulsive noise corruption level, the reconstruction with the proposed method as the preprocessing required much lower measurement rate than the modelbased Lorentzian iterative hard thresholding.
Keywords
1. Introduction
where s and Ψ are a ksparse signal and an N × N orthogonal basis matrix, respectively. k is the number of nonzero elements or a sparsity level. Without loss of generality, Ψ is defined as an identity matrix in this article and x is equivalent to s.
where e is the additive noise.
where ε and u_{ p }are the error bound and the L_{ P } norm of u, respectively. The error bound is set based on the noise characteristics, such as bounded noise, Gaussian noise, finite variance noise, etc. [5–14]. L_{0} norm in Equation (4) is relaxed to L_{1} norm in the reconstruction by Basis Pursuit Denoising (BPDN), whereas it is replaced by heuristic rules in the reconstruction by greedy algorithms.
where τ is a regularization parameter.
where y_{ i } and ${\phi}_{i}^{T}$ are the i th element of y and the i th row of Φ, respectively. The Lorentzianbased Iterative Hard Thresholding (LIHT) approach is proposed as the fast reconstruction method in [22]. Iterative Hard Thresholding (IHT) is used in the place of BP to increase the speed of LBP. However, it faces the same problem as IHT in that it requires high measurement rate in order to acquire successful reconstruction [13]. Consequently, LIHT is suitable for very sparse signals.
The noise tolerance can be increased by including prior knowledge. One of the popular knowledge is the model of a sparse signal [23–29], such as the wavelettree structure. Modelbased reconstruction methods have three benefits: (1) the reduction of the number of measurements, (2) the increase in robustness, and (3) the faster reconstruction.
Even though robust statistic provides the tolerance against impulsive noise, its optimization problem is often difficult. In this article, the impulsive noise rejection method for image data is proposed. It is used as the preprocessing to estimate the noisefree y. It iteratively applies the heuristic rule that is based on the energy distribution of the image data in wavelet domain to detect the existence of the impulsive noise. Octavetree discrete wavelet transform (DWT) is used to transform signals to sparse domain in this article. In an image, most energy should be contained in the thirdlevel subband. The existence of the impulsive noise leads to the high ratio of the energy outside the thirdlevel subband to the total energy. The rejection and the estimation of the noise corrupted elements are made possible by the following fact. In most images, the ksparse signal s can successfully be reconstructed even though some elements in y are removed, because the image data are redundant. The proposed rejection method requires only two parameters: the energyratio threshold and the rejectionratio threshold. These two parameters are easily adjusted and are evaluated for the optimal values as presented in the "Experimental" section.
The proposed method and the impulsive noise cancellation method in [30] are similar as they have two stages: the noise detection and the signal estimation stages. Both methods detect impulsive noise iteratively. However, they are different in a number of aspects. Only a few are mentioned here. The proposed method detects the impulsive noise via the energy distribution of the projected sparse signal. Its estimation stage is separated from its detection stage. The estimation is performed only once after the noise detection has been completed. On the other hand, the method in [30] detects the noise via the difference between the original noisy and the estimated signals; consequently, its estimation stage is integrated into the same loop as its detection stage. The estimation is performed iteratively.
The remainder of this article is organized as follows. Section 2 addresses a brief review of CS, the reconstruction by Orthogonal Matching Pursuit (OMP) and OMP with Partially Known Support (OMPPKS). Section 3 describes the proposed impulsive noise rejection method. The block processing and the vectorization are also given. In Section 4, the proposed method is evaluated. The conclusion is given in Section 5.
2. Background
2.1. Compressed sensing
CS is based on the assumption of the sparse property of signals and incoherency between the basis of the sparse domain and the basis of measurement vectors [1–3]. CS has three major steps: the construction of ksparse representation, the measurement, and the reconstruction. The first step is the construction of the ksparse representation, where k is the sparsity level of the sparse signal. Most natural signals can be made sparse by applying orthogonal transforms such as wavelet transform, Fast Fourier Transform, or Discrete Cosine Transform. This step is represented as Equation (2).
where φ_{ i } and ψ_{ j } are the i th and the j th column in Φ and Ψ, respectively. If the elements in Φ and Ψ are correlated, the coherence is large. Otherwise, it is small. From linear algebra, it is known that $\mu \left(\mathbf{\Phi},\mathbf{\Psi}\right)\phantom{\rule{0.3em}{0ex}}\in \phantom{\rule{2.77695pt}{0ex}}\left[1,\sqrt{N}\right]$[2]. In the measurement process, the error (due to hardware noise, transmission error, etc.) may occur. The error is added into the compressed measurement signal as described in Equation (3).
where δ_{ k } is the krestricted isometry constant of Φ. RIP is used to ensure that all the subsets of k columns taken from Φ are nearly orthogonal. It should be noted that Φ has more columns than rows; thus, Φ cannot exactly be orthogonal [2].
The reconstruction by L_{1}minimization as in BP is stable but slow. Greedy algorithms increase the reconstruction speed by applying heuristic rules. In OMP [31], the heuristic rule is created based on the assumption that y has the large correlation to the bases corresponding to the nonzero elements (or the elements with large magnitude) of s. OMP selects the bases of the nonzero elements according to the correlation and estimates the values of the nonzero elements by the least squared method. The selection is iterated until the certain condition is reached. The reconstruction by greedy algorithms has a fast runtime, but lacks stability and uniform guarantee. RIP is not seriously considered in the greedy algorithms [12].
2.2 Orthogonal matching pursuit
OMP is a wellknown reconstruction algorithm [31]. It was developed from matching pursuit [32] using different method to estimate the magnitude of the nonzero elements in s. Instead of projecting the residual signal onto the selected basis, it estimates the magnitude of the nonzero elements by solving the least squared error between the projection of the reconstructed s and y. OMP has the advantage of simple and fast implementations. The algorithm is as follows.
Input:

The M × N measurement matrix, $\mathbf{\Phi}=\left[\begin{array}{cccc}\hfill {\phi}_{1}\hfill & \hfill {\phi}_{\mathsf{\text{2}}}\hfill & \hfill ...\hfill & \hfill {\phi}_{N}\hfill \end{array}\right]$

The Mdimension compressed measurement signal, y

The sparsity level of the sparse signal, k
Output:

The reconstructed signal, ŝ

The set containing k indexes of nonzero elements in ŝ, Λ_{ k }= {λ_{1}, λ_{2}, ..., λ_{ k }}
 (a)Initialize the residual (r _{0}), the index set (Λ_{0}) and the iteration counter (t) as follows.${\mathbf{r}}_{0}=\mathbf{y},\phantom{\rule{0.3em}{0ex}}{\mathrm{\Lambda}}_{0}=\varnothing ,\phantom{\rule{0.3em}{0ex}}t=1$
 (b)Find the index λ_{ t } of the measurement basis that has the highest correlation to the residual in the previous iteration, r _{t1}.${\lambda}_{t}=arg\underset{j=1,...,N}{max}\left\u3008{\mathbf{r}}_{t1},{\phi}_{j}\u3009\right\phantom{\rule{0.3em}{0ex}}$
 (c)
Augment the index set and the matrix of chosen bases: Λ_{ t }= Λ_{t1}∪{λ_{ t } } and ${\mathbf{\Phi}}_{t}=\left[{\mathbf{\Phi}}_{t1}\phantom{\rule{1em}{0ex}}{\phi}_{{\lambda}_{t}}\right]\phantom{\rule{0.3em}{0ex}}$, where Φ _{0} is an empty matrix.
 (d)Solve the following least squared problem to obtain the new reconstructed signal, z _{ t }.${\mathbf{z}}_{t}=arg\phantom{\rule{0.2em}{0ex}}\underset{z}{min}{\u2225\mathbf{y}{\mathbf{\Phi}}_{t}\mathbf{z}\u2225}_{2}$
 (e)Calculate the new approximation, a _{ t }, that best describes y. Then, calculate the residual of the tth iteration, r _{ t }.$\begin{array}{c}{\mathbf{a}}_{t}={\mathbf{\Phi}}_{t}{\mathbf{z}}_{t}\\ {\mathbf{r}}_{t}=\mathbf{y}{\mathbf{a}}_{t}\end{array}$
 (f)
Increment t by one.
 (g)
If t > k, terminate; otherwise, go to step (b).
The reconstructed signal, ŝ, has nonzero elements at the indexes listed in Λ_{ k }. The value of the λ_{ j } th elements in ŝ equals to the j th element of z_{ k }(j = 1,2,...,k). The termination criterion can be changed from t > k to that r_{t1}is less than the predefined threshold.
2.3. OMP with partially known support
OMPPKS [28] is adapted from the classical OMP [31]. The partially known support gives a priori information to determine which subbands in the sparse signal structure are more important than the others and should be selected as nonzero elements. It has the characteristic of OMP that the requirement of RIP is not as severe as BP's [6]. It has a fast implementation but may fail to reconstruct the signal (lacks stability). It requires very low measurement rate. It is different from Treebased OMP (TOMP) [24] in that the subsequent basis selection of OMPPKS does not consider the previously selected bases, while TOMP sequentially compares and selects the next good wavelet subtree and the group of related atoms in the wavelet tree.
Input:

The M × N measurement matrix, $\mathbf{\Phi}=\left[\begin{array}{cccc}\hfill {\phi}_{1}\hfill & \hfill {\phi}_{\mathsf{\text{2}}}\hfill & \hfill ...\hfill & \hfill {\phi}_{N}\hfill \end{array}\right]$

The Mdimension compressed measurement signal, y

The set containing the indexes of the bases in LL_{3} subbands, Γ = {γ_{1}, γ_{2}, ..., γ_{Γ}}

The sparsity level of the sparse signal, k
Output:

The reconstructed signal, ŝ

The set containing k indexes of the nonzero element in ŝ, Λ_{ k }= {λ_{1}, λ_{2}, ..., λ_{ k }}
Procedure:
 (a)Select every basis in the LL_{3} subband.$t=\left\mathrm{\Gamma}\right$${\mathrm{\Lambda}}_{t}=\mathrm{\Gamma}$${\mathbf{\Phi}}_{t}=\left[\begin{array}{cccc}\hfill {\phi}_{{\gamma}_{1}}\hfill & \hfill {\phi}_{{\gamma}_{2}}\hfill & \hfill ...\hfill & \hfill {\phi}_{{\gamma}_{t}}\hfill \end{array}\right]$
 (b)Solve the least squared problem to obtain the new reconstructed signal, z _{ t }.${\mathbf{z}}_{t}=arg\phantom{\rule{0.2em}{0ex}}\underset{z}{min}{\u2225\mathbf{y}{\mathbf{\Phi}}_{t}\mathbf{z}\u2225}_{2}$
 (c)Calculate the new approximation, a _{ t }, and find the residual (error, r _{ t }). a _{ t }is the projection of y on the space spanned by Φ _{ t }.$\begin{array}{c}{\mathbf{a}}_{t}={\mathbf{\Phi}}_{t}{\mathbf{z}}_{t}\\ {\mathbf{r}}_{t}=\mathbf{y}{\mathbf{a}}_{t}\end{array}$
 (a)
Increment t by one, and terminate if t > k.
 (b)
Apply steps (b)(g) of OMP described in Section 2.2 to find the remaining kΓ nonzero elements of ŝ.
The reconstructed sparse signal, ŝ, has the indexes of nonzero elements listed in Λ_{ k }. The value of the λ_{ j } th element of ŝ equals to the j th element of z_{ k }.
3. Proposed method
The proposed impulsive noise rejection method is described in this section. Block processing and the vectorization of the wavelet coefficients are addressed before a description of the noise rejection method. The block processing is applied to reduce the computation cost. The proposed noise rejection method is applied before the reconstruction and divided into two stages. In the first stage, the algorithm to detect impulsive noise is applied. Then, OMPPKS is applied to estimate the information that is lost due to the impulsive noise. The algorithm to detect the impulsive noise and the estimation of the missing information are described in Sections 3.2 and 3.3, respectively.
3.1. Block processing and the vectorization of the wavelet coefficients
The waveletdomain image in Figure 3b is divided into blocks along its rows as shown in Figure 3c. In Figure 3c, the image has eight rows; consequently, it is divided into eight blocks. Each row in Figure 3c is considered as a sparse signal in this article.
The signal can be made more sparse by the wavelet shrinkage thresholding [33]. In the wavelet shrinkage thresholding, all the coefficients in the LL_{3} subband are preserved, while coefficients outside the LL_{3} subband with magnitude less than the wavelet shrinkage threshold are set to zero. Note that not all coefficients outside the LL_{3} subband are set to zero. Since only the small coefficients in highfrequency subband are set to zero, most distinct edges in the image are preserved. The sparsifying transformation by the wavelet shrinkage thresholding has little distinct visual degradation if the wavelet shrinkage threshold is selected properly.
3.2. The detection of the impulsive noise
 (1)
The reconstructed ŝ has most of its energy inside the thirdlevel subband.
 (2)
The reconstruction is unlikely to be successful because too many elements in y have been removed.
According to the stopping criteria, there are two thresholds that need to be defined. The threshold in the first criterion is used to indicate the amount of the energy that is allowed to be leaked out of the thirdlevel subband. The amount of the energy is measured as the ratio to the total energy. The threshold is defined as the energyratio threshold, η. The threshold in the second criterion is required to ensure that there is sufficient information left for the reconstruction. This threshold is called rejectionratio threshold, T, which is defined as the ratio between the numbers of the removed elements to the size of y(M). Thus, the maximum number of the elements that can be removed is TM. The optimum values of η and T are investigated in Section 4.2.
At each iteration, the noisecorrupted elements are removed and the size of the available measurement signal becomes smaller. Hence, it is required that the reconstruction algorithm is still effective at low measurement rate. OMPPKS is adopted by including the algorithm for the detection and the removal of impulsive noise as follows.
Input:

The M × N measurement matrix, $\mathbf{\Phi}=\left[\begin{array}{cccc}\hfill {\phi}_{1}\hfill & \hfill {\phi}_{\mathsf{\text{2}}}\hfill & \hfill ...\hfill & \hfill {\phi}_{N}\hfill \end{array}\right]$

The Mdimension compressed measurement signal, y

The sparsity level of the sparse signal, k

The number of wavelet coefficients in the thirdlevel subband, l_{3}

The energyratio threshold, η

The rejectionratio threshold, T
Output:

The number of impulsive noise corrupted elements, n_{ δ }

The set containing the n_{ δ }indexes of the impulsive noisecorrupted elements, ${\varsigma}_{\delta}=\left\{{\varpi}_{1},\phantom{\rule{2.77695pt}{0ex}}\phantom{\rule{0.3em}{0ex}}{\varpi}_{2},\phantom{\rule{0.3em}{0ex}}\phantom{\rule{2.77695pt}{0ex}}...,\phantom{\rule{2.77695pt}{0ex}}{\varpi}_{{n}_{\delta}}\right\}$
 (a)
Initialize t = 0, n_{ δ } = 0, ς_{ δ } = ∅, y _{ t }= y, Φ_{ t }= Φ.
 (b)
Apply OMPPKS to reconstruct ŝ from y _{ t }and Φ _{ t }.
 (c)Calculate the energyratio (ER).$\mathsf{\text{ER}}=\frac{\sum _{i={l}_{3}+1}^{N}{{\mathit{\u015d}}_{i}}^{2}}{\sum _{j=1}^{N}{{\mathit{\u015d}}_{j}}^{2}},$
 (d)
Terminate if ER < η.
 (e)
Assign the elements in y _{ t }having the maximum magnitude as the impulsive noise. α_{ m } ($m=1,2,\dots ,{n}_{{\delta}_{t}};\phantom{\rule{2.77695pt}{0ex}}{n}_{{\delta}_{t}}$is the number of the elements having the maximum magnitude in y _{ t }) are defined as the indexes of the recently assigned impulsive noise elements. Note that α_{ m } are the indexes of y. In case that there are more than one element having the maximum magnitude (${n}_{{\delta}_{t}}>1$), all of them are to be removed in step (i).
 (f)
Increment n_{ δ } by ${n}_{{\delta}_{t}}$ and add α_{ m } to ς_{ δ } .
 (g)
Terminate if n_{ δ } ≥ TM.
 (h)
Set t = t+1.
 (i)
y _{ t }is assigned the value of y after the noise elements (the elements with the indexes in ς_{ δ } ) are removed from y. Φ _{ t }is assigned the value of Φ after the rows corresponding to the noise elements are removed from Φ.
 (j)
Go to step (b).
If the algorithm is terminated in step (g), then the removal of impulsive noise is unsuccessful. Too many elements have been removed and it is unlikely that there is sufficient information to reconstruct ŝ and estimate the missing information in the next stage.
It should be noted that the proposed algorithm is applicable to images because image data have some degree of redundancy. The rejectionratio threshold, T, can be set quite large. For the signal data that have low degree of redundancy, the value of T has to be very small. In this case, the reconstruction is unlikely to succeed if every information in y is not used.
3.3. Estimation of the missing information
The outputs from the detection stage and y are used as the inputs of this stage. The noisecorrupted elements, specified in ς_{ δ } , are removed. After the noise removal, the size of the compressed measurement signal y is smaller than the size of the original y; consequently, the reconstruction methods requiring high measurement rate may fail to reconstruct ŝ. It is necessary to estimate the values of the removed elements to preserve the measurement rate. In the proposed method, the values are estimated such that they comply with other noiseless elements. The estimation algorithm is as follows.
Input:

The Mdimension compressed measurement signal, y

The number of impulsive noisecorrupted elements, n_{ δ }

The set containing the n_{ δ }indexes of the impulsive noisecorrupted elements, ${\varsigma}_{\delta}=\left\{{\varpi}_{1},\phantom{\rule{2.77695pt}{0ex}}\phantom{\rule{0.3em}{0ex}}{\varpi}_{2},\phantom{\rule{0.3em}{0ex}}\phantom{\rule{2.77695pt}{0ex}}...,\phantom{\rule{2.77695pt}{0ex}}{\varpi}_{{n}_{\delta}}\right\}$
Output:

The estimated noisefree y, ŷ
 (a)
Define y_{ s } as y with its ϖ_{ i } th (i = 1, 2, ..., n_{ δ } ) elements removed. Define Φ_{ s } as Φ with its ϖ_{ i } th (i = 1, 2, ..., n_{ δ } ) rows removed.
 (b)
Apply OMPPKS to reconstruct ŝ _{s} from y_{ s } and Φ _{ s }.
 (c)Define ỹ = Φŝ _{s} and estimate the i th elements in ŷ as follows.${\mathit{\u0177}}_{i}=\left\{\begin{array}{cc}\hfill {y}_{i}\hfill & \hfill ;i\notin {\varsigma}_{\delta}\hfill \\ \hfill {\mathit{\u1ef9}}_{i}\hfill & \hfill ;i\in {\varsigma}_{\delta}\hfill \end{array}\right.,$
where the subscript i indicates the i th elements of the signal and i = 1, 2, ..., M.
After this process, the impulsive noisecorrupted elements in y are replaced by values complying with noisefree elements. Conventional CS reconstruction methods can be applied to reconstruct ŝ from the impulsive noisefree ŷ.
4. Experiment and discussion
4.1. Experimental environment
Octavetree DWT was used to transform test images to sparse domain. The mother wavelet used in the implementation was Daubechies 8 (db8). The wavelet shrinkage thresholding [33] was applied to make the signal more sparse. The probability of impulsive noise is denoted as p; p ∈ {0, 0.05, 0.10, 0.15, 0.20}. The magnitude of impulsive noise was set relative to the maximum magnitude in y(y_{max}). The measurement matrix was Hadamard matrix. Each wavelet image was divided into 256 blocks of 1 × 256. The sparsity rates (k/N) of blocks in an image were intentionally varied to demonstrate that one set of thresholds was applicable for various sparsity rates. The average sparsity rate in each test image was set to 0.1. The measurement rate (M/N) of an image was the rate averaged over every block in the image. The average measurement rates used in the experiment were 0.2, 0.3, 0.4, 0.5, and 0.6.
The experiment consists of two parts: (1) the evaluation of the two thresholds (η and T) and the minimum size of the detectable impulsive noise given in Section 4.2 and (2) the performance evaluation of the proposed method given in Section 4.3.
4.2. Evaluation of the two thresholds and the minimum size of the detectable impulsive noise
The percent of inaccurate noise rejection of the proposed method
η  The magnitudes of impulsive noise  

1.25 y_{max}  2.5 y_{max}  3 y_{max}  5 y_{max}  10 y_{max}  
0.01  9.09  8.41  8.40  8.39  8.34 
0.02  4.02  1.71  1.76  1.70  1.72 
0.03  5.28  0.60  0.60  0.54  0.54 
0.04  8.46  0.35  0.33  0.28  0.28 
0.05  12.00  0.24  0.13  0.17  0.15 
0.1  33.04  1.22  0.30  0.16  0.10 
0.15  50.07  5.21  1.68  0.16  0.14 
0.2  61.62  13.98  5.35  0.30  0.18 
0.25  68.34  24.28  12.64  0.94  0.26 
0.3  73.98  36.53  22.03  2.00  0.60 
Table 1 also indicated the relationship of η to the percentage of inaccurate rejection. The inaccurate rejection was the result of (1) the rejection of the noisefree elements and (2) the failure to reject the noisecorrupted elements. When η was too small, the energyratio criterion was too strict and the proposed method did not accept even the correct energy distribution of ŝ; consequently, it started to remove the elements uncorrupted by noise. In the opposite case, when η was too large, the energyratio criterion became too lax and the proposed method accepted even the incorrect energy distribution of ŝ; consequently, it failed to remove the noisecorrupted elements. The range of η giving less than 1% of inaccurate rejection was larger, when the magnitude of the impulsive noise was larger. This was because the effect of the impulsive noise to the energy distribution became more distinct and easier to detect when the size of the noise was larger. When the magnitude of the impulsive noise was at least 2.5 y_{max}, the values of η giving less than 1% inaccurate rejection were 0.03, 0.04, and 0.05. Among the three values, the values of η = 0.05 gave the most accurate rejection.
Because the benefit of CS is the capability of compressing the signal to very small size, the measurement rate should be kept low. It is recommended that T be selected such that it is applicable even at low measurement rate. In the following section, T was set to 0.4 to ensure the high probability of successful reconstruction. The value of η was set to 0.05 as it was the optimal value (Table 1).
4.3. Performance evaluation
 (1)
OMPPKS
 (2)
OMPPKS with the proposed rejection method as the preprocessing (OMPPKS+R)
 (3)
Modelbased LIHT (MLIHT) which is the LIHT that is forced to consider the elements in LL_{3} subband as nonzero elements.
 (4)
MLIHT with the proposed rejection method as the preprocessing (MLIHT+R)
The Lorentzian parameter and the number of iteration for MLIHT were 0.25 and 100, respectively. The values of η and T were 0.05 and 0.4, respectively. There were 256 y's in an image and y_{max} was chosen as the maximum magnitude among 256 y's in the image. The magnitude of impulsive noise varied according to the Gaussian pdf with the mean of 7 y_{max} and the standard deviation of y_{max}.
The performance is evaluated based on the PSNR of the reconstructed images, the computation time and the visual quality of the reconstructed images.
At p = 0.05, the effect of adding the proposed method as the preprocessing to MLIHT was minimal; however, at higher p, the addition of the proposed method (the dashed red line) resulted in higher PSNR than the reconstruction by MLIHT alone (the red line). When p was 0.15 or higher, MLIHT was no longer an effective reconstruction method, but MLIHT+R was still effective. It indicated that the addition of the proposed method increased the robustness against p to MLIHT.
It should be noted that even though MLIHT was based on LIHT which was designed to be robust against impulsive noise. MLIHT+R provided less PSNR than OMPPKS+R, because MLIHT required the higher measurement rate. Figure 9 indicated that MLIHT+R was as effective as OMPPKS when the measurement rate was 0.6 and it should become better at the higher measurement rate. However, the improvement by increasing the measurement rate is not recommended because it leads to the large size of y and eliminates the benefit of CS.
 (a)
Apply the proposed rejection method to the first block. Define β as the smallest magnitude of the noise corrupted elements in the first block.
 (b)
Move to the next block. Define the compressed measurement of the new block as y _{ curr }
 (c)
Assign the elements in y _{ curr }having the magnitude not less than β as the impulsive noise. Initialize variables in step (a) of Section 3.2 such that they reflect the removal of the elements with the magnitude not less than β.
 (d)
Apply the proposed rejection method to y _{ curr }. If β is larger than the smallest magnitude of the noise corrupted elements in y _{ curr }, set β to this value.
 (e)
If the current block is the last block in the image, terminate. Otherwise, go to step (b).
The assumption of the above algorithm is that the magnitude of impulsive noise in every block is approximately the same (or share the same distribution). The graphs indicated that the computation time of the reconstruction with the proposed rejection method was no more than four times the computation time of the reconstruction without the proposed rejection method.
5. Conclusion
The impulsive noise rejection for CS reconstruction of image data is proposed. The sparsified version of an image is obtained by applying octavetree DWT using db8 as the mother wavelet. The structure of energy distribution in wavelet domain and the capability to reconstruct the signal from an incomplete y are exploited in order to detect the presence of the impulsive noise. After the noisecorrupted elements are removed, the values of the removed elements are estimated. The experimental results of 60 test images indicated that the proposed rejection method improved the robustness against the impulsive noise of the conventional CS reconstruction methods. The robustness of the reconstruction method against both Gaussian and impulsive noises was also investigated.
Declarations
Acknowledgements
The authors would like to thank the reviewers for their valuable comments and suggestions. This research was financially supported by the National Telecommunications Commission Fund (Grant No. PHD/006/2551 to P. Sermwuthisarn and S. Auethavekiat) and the Telecommunications Research Industrial and Development Institute (TRIDI).
Authors’ Affiliations
References
 Donoho DL: Compressive sensing. IEEE Trans Inf Theory 2006, 52(4):12891306.MathSciNetView ArticleGoogle Scholar
 Candes EJ, Romberg J: Sparsity and incoherence in compressive sampling. Inverse Problem 2007, 23(3):969985. 10.1088/02665611/23/3/008MathSciNetView ArticleGoogle Scholar
 Candes EJ, Wakin MB: An introduction to compressive sampling. IEEE Signal Process Mag 2008, 25(2):2130.View ArticleGoogle Scholar
 Marvasti F, Amini A, Haddadi F, Soltanolkotabi M, Khalaj BH, Aldroubi A, Sanei S, Chambers J: A unified approach to sparse signal processing. EURASIP J Adv Signal Process 2012. doi:10.1186/PREACCEPT1686979482577015Google Scholar
 Tropp JA: Just relax: convex programming methods for identifying sparse signals in noise. IEEE Trans Inf Theory 2006, 52(3):10301051.MathSciNetView ArticleGoogle Scholar
 Candes EJ, Romberg J, Tao T: Stable signal recovery from incomplete and inaccurate measurements. Commun Pure Appl Math 2006, 59(8):12071223. 10.1002/cpa.20124MathSciNetView ArticleGoogle Scholar
 Needell D, Vershynin R: Uniform uncertaintity principle and signal reconstruction via regularized orthogonal matching pursuit. Found Comput Math 2008, 9(3):317334.MathSciNetView ArticleGoogle Scholar
 Omidiran D, Wainwright MJ: Highdimensional subset recovery in noise: sparsified measurements without loss of statistical efficiency. Department of Statistics, UC Berkeley, USA; 2008.Google Scholar
 Candès EJ, Tao T: The dantzig selector: Statistical estimation when p is much larger than n . Ann Statist 2007, 35(6):23132351. 10.1214/009053606000001523MathSciNetView ArticleGoogle Scholar
 Needell D, Tropp JA: CoSaMP: iterative signal recovery from incomplete and inaccurate samples. Appl Comput Harmonic Anal 2008, 26(3):301321.MathSciNetView ArticleGoogle Scholar
 BenHaim Z, Eldar YC, Elad M: Coherencebased nearoracle performance guarantees for sparse estimation under Gaussian noise. In Proc ICASSP. Texas, USA; 2010:3590.Google Scholar
 Needell D: Topics in compressed sensing. In Ph.D. dissertation. Math. Univ. of California, Davis; 2009.Google Scholar
 Blumensath T, Davies ME: Iterative hard thresholding for compressed sensing. Appl Comput Harmonic Anal 2009, 27(3):265274. 10.1016/j.acha.2009.04.002MathSciNetView ArticleGoogle Scholar
 Blumensath T, Davies ME: Normalized iterative hard thresholding: guaranteed stability and performance. IEEE J Sel Topics Signal Process 2010, 4(2):298309.View ArticleGoogle Scholar
 Popilka B, Setzer S, Steidl G: Signal recovery from incomplete mesurements in the presence of outliers. Inverse Problems Imag 2007, 1(4):661672.MathSciNetView ArticleGoogle Scholar
 Carrillo RE, Aysal TC, Barner KE: A theoretical framework for problem requiring robust behavior. In Proc IEEE CAMSAP. Aruba, Dutch Antilles; 2009:25.Google Scholar
 Carrillo RE, Aysal TC, Barner KE: A generalized cauchy distribution framework for problems requiring robust behavior. EURASIP J Adv Signal Process 2010, 2010: 19. Article ID 312989View ArticleGoogle Scholar
 Carrillo RE, Barner KE, Aysal TC: Robust sampling and reconstruction methods for sparse signals in the presence of impulsive noise. IEEE J Sel Topics Signal Process 2010, 4(2):392408.View ArticleGoogle Scholar
 Arce GR, Otero D, Ramirez AB, Paredes J: Reconstruction of sparse signal from l_{1}dimensionality reduced cauchy randomprojections. In Proc IEEE ICASSP. Texas, USA; 2010:4014.Google Scholar
 Ramirez AB, Arce GR, Sadler BM: Fast algorithm for reconstruction of sparse signal from cauchy random projections. In Proc EUSIPCO. Aalborg, Denmark; 2010:432.Google Scholar
 Paredes JL, Arce GR: Compressive sensing signal reconstruction by weighted median regression estimates. IEEE Trans Signal Proces 2011, 59(6):25852601.MathSciNetView ArticleGoogle Scholar
 Carrillo RE, Barner KE: Lorentzian based iterative hard thresholding for compressed sensing. In Proc IEEE ICASSP. Prague, Czech Republic; 2011:3664.Google Scholar
 La C, Do MN: Signal reconstruction using sparse tree representation. In Proc of SPIE Conf on Wavelet App in Sig and Image Proc. Volume 5914. San Diego, USA; 2005:273.Google Scholar
 La C, Do MN: Treebased orthogonal matching pursuit algorithm for signal reconstruction. In Proc IEEE ICIP. Georgia, USA; 2006:1277.Google Scholar
 Duarte MF: Compressed sensing for signal ensembles. In Ph.D. dissertation. Dept. Elect. Eng., Rice Univ., Houston, Texas; 2009.Google Scholar
 He L, Carin L: Exploiting structure in waveletbased Baysian compressive sensing. IEEE Trans Signal Process 2009, 57: 34883497.MathSciNetView ArticleGoogle Scholar
 Baron D, Wakin MB, Duarte MF, Sarvotham S, Baraniuk RG: Distributed compressed sensing. Rice Univ., Dept. Elect. and Com. Eng., Houston, TX; 2006.Google Scholar
 Carrillo RE, Polania LF, Barner KE: Iterative algorithm for compressed sensing with partially known support. In Proc IEEE ICASSP. Texas, USA; 2010:3654.Google Scholar
 Xu M, Lu J: Kclustervalues compressive sensing for imaging. Eurasip J Adv Signal Process 2011, 2011: 75. 10.1186/16876180201175View ArticleGoogle Scholar
 Zahedpour S, Feizi S, Amini A, Ferdosizadeh M, Marvasti F: Impulsive noise cancellation based on soft decision and recursion. IEEE Trans Instrum Meas 2009, 58(8):27802790.View ArticleGoogle Scholar
 Tropp JA, Gilbert AC: Signal recovery from random measurements via orthogonal matching pursuit. IEEE Trans Inf Theory 2007, 53(12):46554666.MathSciNetView ArticleGoogle Scholar
 Mallat SG, Zhang Z: Matching pursuits with timefrequency dictionaries. IEEE Trans Signal Process 1993, 41(12):33973415. 10.1109/78.258082View ArticleGoogle Scholar
 Donoho DL: Denoising by soft thresholding. IEEE Trans Inf Theory 1995, 38(2):613627.MathSciNetView ArticleGoogle 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.