Channelized Noise Augmentation to Endorse DICOM Medical Image for Diagnosing
Kamred Udham Singh, Achintya Singhal · SSRN Electronic Journal · 2018
Advance medical imaging technology viz. Color Doppler Imaging, Positron Emission Tomography (PET), CT scan, MRI, Neuroimaging Informatics Technology Initiative etc. generates DICOM (Digital Imaging and Communications in Medicine) images, which allows significant enlargement or contraction of the images by medical practitioners to diagnose the patient condition. DICOM images also contain patient information that differs it from other medical image formats such as .bmp, jpg, .tiff etc.Teleradiology allows transmitting these images over a communication network for second opinion/diagnostics. The images transmitted over the network are susceptible to theft and various attacks yielding the problems of copyright, privacy, legal and ethical issues. To address these challenges the researchers investigated various watermarking scheme but due to the sensitivity of DICOM images and data confined in it; no notable state of the art contributions on color DICOM images watermarking is found. In this paper, we propose a non-blind watermarking scheme based on Discrete Wavelet Transform (DWT) and LU factorization. The combination of DWT-LU helps in retaining the sensitivity of the data, and improvise the robustness of the scheme by embedding multifarious watermarks into the colored DICOM medical images. This approach and the embedding technique aids authentication of these medical images even post image format conversion and compression.