Ch. Hima Bindu

SVD based image watermarking with firefly algorithm

This paper presents an image watermarking scheme based on the singular value decomposition (SVD) on non-overlapping blocks of both the host image and the watermark image. The singular values of each non-overlapped block of binary watermark are embedded into the singular values of corresponding non-overlapped block of the host image using the Multiple Scaling Factors (MSFs) of firefly algorithm. Finally, the objective function is to be computed to know the best scaling factor for embedding. In addition to this, various types of attacks (like compression, gaussian filtering, sharpening, scaling and cropping etc..) are applied on embedded image to show the robustness of the proposed watermarking scheme. The proposed watermarking scheme gives better optimization, less computations, robustness and good at visual quality. For this, the Structure Similarity Index measure (SSIM) is used to indicate the measure of imperceptibility and robustness of watermarking scheme.

Image fusion with Biorthogonal Wavelet Transform based on maximum selection and region energy

Image Fusion plays major research role in the fields of image processing. Image Fusion is a method of combining the relevant information from a set of images, into a single image, where in the resultant fused image will be more informative and complete than any of the input images. Specifically it serves best in medical diagnosis i.e. Computed Tomography (CT), Magnetic Resonance Image (MRI) scans provide different types of information, by fusion can get accurate information for better clinical diagnosis. The Biorthogonal Wavelet Transform (BWT) is one of the most widely used transform method for fusion. Here this paper discusses the Biorthogonal wavelet transform based image fusion with absolute maximum selection rule and energy based fusion rule. The proposed method analysed both qualitatively and quantitatively among various fusion methods.

Medical image fusion using content based automatic segmentation

Image fusion is a process of combining complementary information from multi modality images of the same patient in to an image. Hence the resultant image consists of more informative than the individual images alone. In this paper, a novel feature level image fusion is proposed. In feature level fusion, source images are segmented into regions and features like pixel intensities, edges or texture are used for fusion. The feature level image fusion with region based would be more meaningful than the pixel based fusion methods. The proposed fusion method contains three steps. Firstly, the multi modal images are segmented into regions using automatic segmentation process. Secondly the images are fused according to region based fusion rule. Finally the regions are merged together to acquire final fused image. The performance of the proposed method can be evaluated with fusion symmetry, peak signal to noise ratio both quantitatively and qualitatively.

Automatic Region Segmentation and Variance Based Multimodal Medical Image Fusion

In this technical paper, multimodal medical image fusion using automatic segmentation and variance is proposed. Image fusion is used to mix more images of different modalities into a single image. The fused image consists of more information than the individual images alone. The way of fusion process on region based image fusion. Initially the images are automatically segmented into regions using 3-D doctor software. These region wise statistical properties are used in the process to make accurate decision on fusion. At last the fused image is merging of all the isolated regions. The efficiency of the algorithm is evaluated with quantitative parameters like fusion symmetry and region cross correlation coefficient.

QPSK signal reconstruction using compressive sensing algorithms

Now a days Compressed Sensing (CS) is one of the crucial task in signal processing. To reconstruct the signal efficiently, we can follow the Nyquist criteria by finding solutions to undetermined linear systems. These systems have less no of equations with more unknowns. This paper concentrates on the use of sub-Nyquist technique such as CS which detect the spectrum usage without the prior information of sparsity. Narrowband QPSK signal is recovered by using OMP (Orthogonal Matching Pursuit), CoSaMP (Compressive Sampling Matching Pursuit) and l-1 minimization algorithms. Experimental results show that the comparative analysis of the above algorithms in terms of MSE(Mean Square Error) and SNR(Signal to Noise Ratio). All simulations are carried out using MATLAB.

A novel medical image fusion with lαβ color transformation

For clinical application, the medical images play a vital role. The various multimodal medical images like MRI(magnetic resonance imaging), CT(computed tomography), PET(positron emission tomography), SPECT(single photon emission computed tomography) etc., represent various functional information of the body. The purpose of the proposed work is to acquire more information of the various clinical multi modal images in single image with good visual perception and more quality (called process of fusion). This paper proposes a novel image fusion with new color transformation. Non Subsampled Contourlet Transform (NSCT) is used in this work. At first, one of the RGB color images is transformed into lαβ color model. Later NSCT is applied on source images to obtain low & high frequency coefficients. Here, low frequency coefficients are processed using phase congruency model and high frequency coefficients are processed using Sum modified laplacian (SML) based directive contrast model. At last, the proposed method is compared with existing method [14]. The effectiveness of proposed method is is applied for assessing using various performance measures like normalized mutual information and structural similarities metrics.

MRI–PET Medical Image Fusion Technique by Combining Contourlet and Wavelet Transform

This paper proposes the application of the hybrid Multiscale transform in medical image fusion. The multimodality medical image fusion plays an important role in clinical applications which can support more accurate information for physicians to diagnosis diseases. In this paper, a new fusion scheme for Magnetic Resonance Images (MRI) and Positron Emission Tomography (PET) images based on hybrid transforms is proposed. PET/MRI medical image fusion has important clinical significance. Medical image fusion is the important step after registration, which is an integrative display method of two images. The PET image indicates the brain function and a low spatial resolution; MRI image shows the brain tissue anatomy and contains no functional information. Hence, a perfect fused image should contains both more functional information and more spatial characteristics with no spatial and color distortion. Firstly, the image is decomposed into high and low frequency subband coefficients with discrete wavelet transform (DWT). On these coefficients apply contourlet transform individually before going for fusion process. Later the fusion process is performed on contourlet components for each subband, for fusion the spatial frequency method is used. Finally, the proposed algorithm results are compared with different Multiscale transform techniques. According to simulation results, the algorithm holds useful information from source images.