Limits...
Multi-scale pixel-based image fusion using multivariate empirical mode decomposition.

Rehman Nu, Ehsan S, Abdullah SM, Akhtar MJ, Mandic DP, McDonald-Maier KD - Sensors (Basel) (2015)

Bottom Line: A novel scheme to perform the fusion of multiple images using the multivariate empirical mode decomposition (MEMD) algorithm is proposed.A variety of image fusion quality measures are employed for the objective evaluation of the proposed method.We also report the results of a hypothesis testing approach on our large image dataset to identify statistically-significant performance differences.

View Article: PubMed Central - PubMed

Affiliation: Department of Electrical Engineering, COMSATS Institute of Information Technology, Park Road, Chak Shahzad, Islamabad 44000, Pakistan. naveed.rehman@comsats.edu.pk.

ABSTRACT
A novel scheme to perform the fusion of multiple images using the multivariate empirical mode decomposition (MEMD) algorithm is proposed. Standard multi-scale fusion techniques make a priori assumptions regarding input data, whereas standard univariate empirical mode decomposition (EMD)-based fusion techniques suffer from inherent mode mixing and mode misalignment issues, characterized respectively by either a single intrinsic mode function (IMF) containing multiple scales or the same indexed IMFs corresponding to multiple input images carrying different frequency information. We show that MEMD overcomes these problems by being fully data adaptive and by aligning common frequency scales from multiple channels, thus enabling their comparison at a pixel level and subsequent fusion at multiple data scales. We then demonstrate the potential of the proposed scheme on a large dataset of real-world multi-exposure and multi-focus images and compare the results against those obtained from standard fusion algorithms, including the principal component analysis (PCA), discrete wavelet transform (DWT) and non-subsampled contourlet transform (NCT). A variety of image fusion quality measures are employed for the objective evaluation of the proposed method. We also report the results of a hypothesis testing approach on our large image dataset to identify statistically-significant performance differences.

No MeSH data available.


Related in: MedlinePlus

EMD decomposition of a synthetic trivariate dataset showing mode mixing and mode misalignment due to the empirical nature of EMD.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4481979&req=5

f1-sensors-15-10923: EMD decomposition of a synthetic trivariate dataset showing mode mixing and mode misalignment due to the empirical nature of EMD.

Mentions: The mode mixing and mode misalignment phenomena in EMD are next demonstrated with the aid of a synthetic trivariate signal composed of a combination of sinusoids, as shown in the top row of Figure 1. A 12-Hz sinusoid was introduced in all components, while a 24-Hz sinusoid was present in the components X and Z and a 4-Hz sinusoid in the components X and Y; the X and Z components were also contaminated with WGN. By applying the EMD algorithm to each component separately, different numbers of IMFs were obtained in each case resulting in mode misalignment, i.e., different scales in the same indexed IMFs of different components. This is evident in all of the IMFs of all of the components, as shown in Figure 1. Mode mixing is also clearly visible in IMF 3 of Channels X and Z and IMF 1 of Channels X and Z. No mode mixing was observed in any of the IMFs of Channel Y, however.


Multi-scale pixel-based image fusion using multivariate empirical mode decomposition.

Rehman Nu, Ehsan S, Abdullah SM, Akhtar MJ, Mandic DP, McDonald-Maier KD - Sensors (Basel) (2015)

EMD decomposition of a synthetic trivariate dataset showing mode mixing and mode misalignment due to the empirical nature of EMD.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4481979&req=5

f1-sensors-15-10923: EMD decomposition of a synthetic trivariate dataset showing mode mixing and mode misalignment due to the empirical nature of EMD.
Mentions: The mode mixing and mode misalignment phenomena in EMD are next demonstrated with the aid of a synthetic trivariate signal composed of a combination of sinusoids, as shown in the top row of Figure 1. A 12-Hz sinusoid was introduced in all components, while a 24-Hz sinusoid was present in the components X and Z and a 4-Hz sinusoid in the components X and Y; the X and Z components were also contaminated with WGN. By applying the EMD algorithm to each component separately, different numbers of IMFs were obtained in each case resulting in mode misalignment, i.e., different scales in the same indexed IMFs of different components. This is evident in all of the IMFs of all of the components, as shown in Figure 1. Mode mixing is also clearly visible in IMF 3 of Channels X and Z and IMF 1 of Channels X and Z. No mode mixing was observed in any of the IMFs of Channel Y, however.

Bottom Line: A novel scheme to perform the fusion of multiple images using the multivariate empirical mode decomposition (MEMD) algorithm is proposed.A variety of image fusion quality measures are employed for the objective evaluation of the proposed method.We also report the results of a hypothesis testing approach on our large image dataset to identify statistically-significant performance differences.

View Article: PubMed Central - PubMed

Affiliation: Department of Electrical Engineering, COMSATS Institute of Information Technology, Park Road, Chak Shahzad, Islamabad 44000, Pakistan. naveed.rehman@comsats.edu.pk.

ABSTRACT
A novel scheme to perform the fusion of multiple images using the multivariate empirical mode decomposition (MEMD) algorithm is proposed. Standard multi-scale fusion techniques make a priori assumptions regarding input data, whereas standard univariate empirical mode decomposition (EMD)-based fusion techniques suffer from inherent mode mixing and mode misalignment issues, characterized respectively by either a single intrinsic mode function (IMF) containing multiple scales or the same indexed IMFs corresponding to multiple input images carrying different frequency information. We show that MEMD overcomes these problems by being fully data adaptive and by aligning common frequency scales from multiple channels, thus enabling their comparison at a pixel level and subsequent fusion at multiple data scales. We then demonstrate the potential of the proposed scheme on a large dataset of real-world multi-exposure and multi-focus images and compare the results against those obtained from standard fusion algorithms, including the principal component analysis (PCA), discrete wavelet transform (DWT) and non-subsampled contourlet transform (NCT). A variety of image fusion quality measures are employed for the objective evaluation of the proposed method. We also report the results of a hypothesis testing approach on our large image dataset to identify statistically-significant performance differences.

No MeSH data available.


Related in: MedlinePlus