Romain Mavudila - Academia.edu (original) (raw)

Papers by Romain Mavudila

Chemical Physics Letters, 2019

This is a PDF file of an article that has undergone enhancements after acceptance, such as the ad... more This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

MRI images are often subject to noise (artifacts). We evaluated the performance of DT-CWT combine... more MRI images are often subject to noise (artifacts). We evaluated the performance of DT-CWT combined to Bayesian MAP Estimator to restore those images. We chose the images from two of four sequences commonly used in coronal and axial MRI with and without contrast agent. The image with contrast agent was used as a reference, and the image without where we added artificially noise was subjected to de-noising by forward Dual-Tree Wavelet Transform (DT-CWT) combined to Bayesian MAP estimator. A test was submitted to radiologists for an assessment of the de-noised images to compare the proposed algorithm to other effective techniques from the recent literature using MRI images. Our approach contributed effectively to the MRI images de-noising, with better results. In general, wavelets and Bayesian estimator contributed effectively to the denoising process and to other image processing methods, but ranging from classic to complex wavelet transforms, the results gradually improved.

This paper introduces a new reversible and blind scheme based on the analysis of Multi-band Dual ... more This paper introduces a new reversible and blind scheme based on the analysis of Multi-band Dual Tree Wavelet domain in medical image watermarking domain.Different from many other watermarking schemes with wavelet transform are used on one side in which the studies have dealt only in the dyadic case, in the other hand, in multi-band only on the classical discrete wavelet (DWT), for this suggested approach: security issues for medical information and technical solution to protect these data in medical information are examined. A solution is proposed, which consist in watermarking image areas that are not relevant for the diagnosis, considering advantages of multi-bands wavelet analysis. Applications of the method were examined for MRI images watermarking, inserting data in different sub-bands with private key provides high security for the embedded data using discrete wavelet and complex wavelet (CWT) domain.Experimental results obtained from several watermarked medical images indicate imperceptibility, in each case; the CWT provides higher capacity than the real DWT and DWT domain, by increasing the number of bands. The system is transparent to the user and allows image integrity control; in addition it provides information on the location of potential alterations and evaluation of image modifications which is of major importance in medico-legal framework. The proposed method generalizes the findings of previous works.

Chemical Physics Letters, 2019

This is a PDF file of an article that has undergone enhancements after acceptance, such as the ad... more This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

MRI images are often subject to noise (artifacts). We evaluated the performance of DT-CWT combine... more MRI images are often subject to noise (artifacts). We evaluated the performance of DT-CWT combined to Bayesian MAP Estimator to restore those images. We chose the images from two of four sequences commonly used in coronal and axial MRI with and without contrast agent. The image with contrast agent was used as a reference, and the image without where we added artificially noise was subjected to de-noising by forward Dual-Tree Wavelet Transform (DT-CWT) combined to Bayesian MAP estimator. A test was submitted to radiologists for an assessment of the de-noised images to compare the proposed algorithm to other effective techniques from the recent literature using MRI images. Our approach contributed effectively to the MRI images de-noising, with better results. In general, wavelets and Bayesian estimator contributed effectively to the denoising process and to other image processing methods, but ranging from classic to complex wavelet transforms, the results gradually improved.

This paper introduces a new reversible and blind scheme based on the analysis of Multi-band Dual ... more This paper introduces a new reversible and blind scheme based on the analysis of Multi-band Dual Tree Wavelet domain in medical image watermarking domain.Different from many other watermarking schemes with wavelet transform are used on one side in which the studies have dealt only in the dyadic case, in the other hand, in multi-band only on the classical discrete wavelet (DWT), for this suggested approach: security issues for medical information and technical solution to protect these data in medical information are examined. A solution is proposed, which consist in watermarking image areas that are not relevant for the diagnosis, considering advantages of multi-bands wavelet analysis. Applications of the method were examined for MRI images watermarking, inserting data in different sub-bands with private key provides high security for the embedded data using discrete wavelet and complex wavelet (CWT) domain.Experimental results obtained from several watermarked medical images indicate imperceptibility, in each case; the CWT provides higher capacity than the real DWT and DWT domain, by increasing the number of bands. The system is transparent to the user and allows image integrity control; in addition it provides information on the location of potential alterations and evaluation of image modifications which is of major importance in medico-legal framework. The proposed method generalizes the findings of previous works.