A Comparative Study On Diverse Fuzzy Logic Techniques In Segmenting The Color Images

Vol. 2  Issue 2
Year: 2015
Issue:Apr-Jun
Title:A Comparative Study On Diverse Fuzzy Logic Techniques In Segmenting The Color Images
Author Name:P. Ramesh and S. Thilagamani
Synopsis:
During the past several decades, there were tremendous developments in the field of Image Segmentation. Due to the extreme thrust in enhancing the quality of the image segmentation process, numerous segmentation techniques have evolved. Segmentation in color images is quite difficult due to the uncertainties that exist at the boundary. Fuzzy logic is an ideal concept which is well suited in such cases. It is an approach of computation which is based on the degrees of truth rather than the Boolean logic through which the modern computer works. Fuzzy techniques are highly popular due to its rapid extension of fuzzy set theory and are mainly based on the binary valued membership. Thus the Fuzzy techniques applied in image processing can efficiently manage the ambiguities present in the images. Hence this comparative analysis mainly indicates the working methodologies of different collections of Fuzzy logic techniques in image segmentation and this in turn helps the image processing researchers to innovate more advanced techniques in Fuzzy concept and solve the problem in hand.

Guassian and Speckle Noise Removal From Ultrasound Images Using Bivariate Shrinkage By Dual Tree Complex Wavelet Transform

Vol. 2  Issue 2
Year: 2015
Issue:Apr-Jun
Title:Guassian and Speckle Noise Removal From Ultrasound Images Using Bivariate Shrinkage By Dual Tree Complex Wavelet Transform
Author Name:Devanand Bhonsle,Vivek Chandra and G. R. Sinha
Synopsis:
This paper introduces bivariate thresholding based Dual Tree Complex Wavelet Transform (DTCWT) technique to remove both Gaussian and speckle noise signals. Since both types of noises are different in nature hence it is difficult to remove them by using single filter. In this paper DTCWT approach is used to denoise ultrasound images. DTCWT based filter removes Additive White Gaussian Noise (AWGN) effectively. Since speckle noise is multiplicative in nature; it is converted into logarithmic transform before applying wavelet transform. Bivariate shrinkage (soft thresholding) function is used.

Better Scheme For Multimedia Compression Using Random Discrete Fractional Fourier Transform For Jpeg 2000 Standard

Vol. 2  Issue 1
Year: 2015
Issue:Jan-Mar
Title:Better Scheme For Multimedia Compression Using Random Discrete Fractional Fourier Transform For Jpeg 2000 Standard
Author Name:Deepak Sharma, Udita Arora, Prakram Suri and Rishabh Tripathi
Synopsis:
Rapid growth of high quality multimedia (HD) and exchange of data over internet with less storage space and fast processing attracted researchers in the area of compression. Compression is the technique of reducing the image size without degrading the quality of the image. In this work, a commuting matrix with random discrete Fourier transform (DFT) eigenvectors is first constructed. A Random Discrete Fractional Fourier Transform (RDFRFT) kernel matrix with random DFT eigenvectors and eigenvalues is then utilized in image compression. The RDFRFT has an important feature that the magnitude and phase of its transform output are both random. Later, a compression scheme based on random discrete fractional Fourier transform is compared with Discrete Cosine Transform (DCT) and discrete wavelet transform (DWT) based image compression schemes. The given image is subdivided and RDFRFT is applied for each subdivided image to transformed coefficients and reverse order of RDFRFT is applied for reconstruction of original images. The performance of compression scheme based on RDFRFT shows better performance over DFRFT, DCT and DWT based scheme for any multimedia contents. The performance of the proposed scheme is observed on JPEG standard image for prime evaluation parameters such as Peak Signal-to-Noise Ratio (PSNR), Mean Square Error (MSE) and Compression Ratio (CR) for RDFRFT, DCT and DWT based scheme on MATLAB software platform. In addition, the proposed scheme has following advantages: it shows the same computation complexity as DFRFT based system and the feature of additional security can also be incorporated with RDFRFT which is not very significant in case of FRFT and DFRFT based system.

A Hybrid Robust Watermarking And Encryption For Medical Images Using GA Optimized DWT And SVD

Vol. 2  Issue 1
Year: 2015
Issue:Jan-Mar
Title:A Hybrid Robust Watermarking And Encryption For Medical Images Using GA Optimized DWT And SVD
Author Name:CH. Venugopal Reddy and P. Siddaiah
Synopsis:
Watermarking secures the integrity of the medical images and has been used. One specific limitation for the methods that can be used in water marking medical images is that water marking procedure should not result in degradation of the image. In this work, a dual security approach comprising water marking and encryption is implemented to provide authenticity and protect the integrity of the medical image. A multi objective optimization function that can preserve the structural integrity of the medical image while maintaining robustness and imperceptibility is formulated. Genetic Algorithm (GA) optimization approach is used to optimize the multi objective function. The water mark is implemented using hybrid approach comprising Discrete Wavelet Transform (DWT) and Singular Value Decomposition (SVD). The encryption is affected using RSA & AES encryption algorithms. A graphical user interface (GUI) is designed to present the proposed approach in the form of usable tool to help the user in the embedding and extracting process. The water mark is subjected to different types of noise attacks and its robustness and integrity are tested.

Bone Thickness Computation At Low Resolution IN-VIVO CT Images And Classification Through Ann

Vol. 2  Issue 1
Year: 2015
Issue:Jan-Mar
Title:Bone Thickness Computation At Low Resolution IN-VIVO CT Images And Classification Through Ann
Author Name:Jayachitra, and M. Usha 
Synopsis:
Osteoporosis is a bone disease affecting the bone structure and strength and raising the risk of fractures. Osteoporosis is a bone condition that makes bones thinner and more fragile because of reduced bone density. Osteoporosis may be diagnosed directly through the use of a bone scan that measures bone mineral density (BMD). The micro-architectural quality of Trabecular bone is an important factor of bone quality for evaluating fracture risks under clinical conditions. A new algorithm is implemented for computing TB thickness at a low resolution which is achievable in IN-VIVO images. Here the authors have proposed intercept based algorithm to compute its thickness, and robustness to bone strength. Through this algorithm, the true axis of an object orthogonally intersects a minimum intercept line. By calculating centroids of an image, the thickness is calculated. Detected Image can be classified through artificial neural network classifier.

Integrated Registration And Segmentation For Finding Defective Areas In Kidneys

Vol. 2  Issue 1
Year: 2015
Issue:Jan-Mar
Title:Integrated Registration And Segmentation For Finding Defective Areas In Kidneys
Author Name:Baby D Dayana, N. Meffiya, and P. Latha 
Synopsis:
Kidney failure is the major threat in medical field. The kidneys dysfunction and disease can be filtered using the parameter Glomerular Filtration Rate(GFR). Finding the affected part is the most critical phenomenon in medical field. This methods gives the exact area detection in kidney so that the diagnosis and treatment is easily carried out. The aim of the project is to find the Glomerular Filtration Rate(GFR) and according to it, the affected part in kidney is found out. It deals with renal failure and it applies the concept of Registration and segmentation and classification .

SCR Technique For Tumour Operation Using Biomedical Image Processing And Its Implementation In TMS320C6713

Vol. 2  Issue 1
Year: 2015
Issue:Jan-Mar
Title:SCR Technique For Tumour Operation Using Biomedical Image Processing And Its Implementation In TMS320C6713
Author Name:S. Lysiya Merlin, M. Nethravathi, K. P. Harini and N. Santhiyakumari
Synopsis:
One of the major issues in the field of biomedical research is dealt with operating a tumor tissue successfully without causing any damage to the other good tissues. 'SCR (store, compare, recognize) technique for tumor operation using biomedical image processing and its implementation in TMS320C6713' is an advanced process for identifying and tracking the images of the tumor by using 'Aphelion Dev Software development module'. This is especially used for locating the tumor found in highly sensitive organs like brain and heart. In this project, two stages have been involved. The first stage involves the conversion of raw binary information scanned from a CT scan image of the infected organ to colour image. This processed image helps the surgeons to find the tumor sizes and locate them. In second stage, the operation of the tumor analyzed in stage I is recorded using a camera in a neck worn pendent. The recorded video frames are compared with the stored processed image and its histogram of stage I. As a result of this, the tumor tissues are segmented from the good tissues in the video. Now the output video after processing has only the cropped images of the tumor eliminating the good tissues. Using Parallel Processing technique in TMS320C6713 processor, the computational time can be reduced. Thus the surgeons can operate the identified tumor safely without affecting the other good tissues in a short span of time.