R. Bhavani

Classification of MRI brain images using k-nearest neighbor and artificial neural network

Magnetic resonance imaging (MRI) is often the medical imaging method of choice when soft tissue delineation is necessary. This paper presents a new approach for automated diagnosis based on classification of the magnetic resonance images (MRI). The proposed method consists of two stages namely feature extraction and classification. In the first stage, we have obtained the features related to MRI images using discrete wavelet transformation (DWT). Wavelet transform based methods are a well known tool for extracting frequency space information from non-stationary signals. The features extracted using DWT of magnetic resonance images have been reduced, using principal component analysis (PCA), to the more essential features. In the classification stage, two classifiers have been developed. The first classifier is based on feed forward back propagation artificial neural network (FP-ANN) and the second classifier is based on k-nearest neighbor (k-NN). The features hence derived are used to train a neural network based binary classifier, which can automatically infer whether the image is that of a normal brain or a pathological brain, suffering from brain lesion. A classification with a success of 90% and 99% has been obtained by FP-ANN and k-NN, respectively. This result shows that the proposed technique is robust and effective compared with other recent work.

Dual transform based steganography using wavelet families and statistical methods

Steganography is the discipline of exchanging top secret information by embedding it into a multimedia carrier. The ultimate aim, here is to hide the very existence of the embedded information within seemingly innocuous carriers. The proposed method extracts either Discrete Wavelet Transform (DWT) or Integer Wavelet Transform (IWT) coefficients of both cover image and secret image. After that two extracted coefficient values are embedded by fusion processing technique. Then the stego image is obtained by applying various combinations of DWT and IWT on both images. In this method, we concentrated for perfecting the visual effect of the stego image and robustness against the various attacks by using different wavelet families. Finally performance evaluation is done on dual transform steganography using wavelet families and statistical methods. In our method achieved acceptable imperceptibility and certain robustness.

A robust QR-Code video watermarking scheme based on SVD and DWT composite domain

Nowadays, Digital video is one of the popular multimedia data exchanged in the internet. Commercial activity on the internet and media require protection to enhance security. The 2D Barcode with a digital watermark is a widely interesting research in the security field. In this paper propose a video watermarking with text data (verification message) by using the Quick Response (QR) Code technique. The QR Code is prepared to be watermarked via a robust video watermarking scheme based on the (singular value decomposition)SVD and (Discrete Wavelet Transform)DWT. In addition to that logo (or) watermark gives the authorized ownership of video document. SVD is an attractive algebraic transform for watermarking applications. SVD is applied to the cover I-frame. The extracted diagonal value is fused with logo (or) watermark. DWT is applied on SVD cover image and QR code image. The inverse transform on watermarked image and add the frame into video this watermarked (include logo and QR code image) the video file sends to authorized customers. In the reverse process check the logo and QR code for authorized ownership. These experimental results can achieved acceptable imperceptibility and certain robustness in video processing.

Enhancing k-means and kernelized fuzzy c-means clustering with cluster center initialization in segmenting MRI brain images

Clustering is the process of organizing data objects into a set of disjoint classes called clusters. The objective of this paper is to develop an enhanced k-means and kernelized fuzzy c-means for a segmentation of brain magnetic resonance images. Performance of iterative clustering algorithms which converges to numerous local minima depend highly on initial cluster centers. In general the clustering algorithm chooses the initial centers in random manner. In this paper we propose a new center initialization algorithm for measuring the initial centers of the proposed clustering algorithms. This algorithm is based on maximum measure of the distance function which is found for cluster center detection process. More recently clustering is an effective tool in segmenting medical images for further treatment plan. In order to solve the problems of clustering performance affected by initial centers of clusters, this paper introduces a specialised center initialization method for executing the proposed algorithms in segmenting medical images. Experiments are performed with real brain images to access the performance of the proposed methods. Further the validity of clustering results are obtained using silhouette method and compares the results with the results of original k-means and fuzzy c-means clustering algorithms. The experimental results show the superiority of the proposed clustering results.

A combined classifier kNN-SVM in gait-based biometric authentication system

The objective of this paper is to develop an efficient authentication system with reduced search space to recognise individuals based on their gait when they enter into surveillance area. To achieve this objective: 1 we have split the database into two based on gender and then the search is restricted to the identified gender database; 2 based on one gaitcycle, we have selected gait representing static and dynamic features which are invariant to various covariate factors such as wearing coats and carrying; 3 we used the decisions of both k-nearest neighbour kNN and support vector machine SVM by decision level fusion. Experimental results evaluated on the benchmark CASIA B gait dataset shows superior performance in terms of correct classification rate and it shows robustness to variations in clothing and carrying conditions.

Dual Wavelet Transform in Color Image Steganography Method

Steganography is the Art and Science of hiding the information and remarkable cover media. So, as not to arouse on Eaviesdroppers suspicion. In this paper, secret image is hide into two different domains like as IWT (Integer Wavelet Transform) and DWT (Discrete Wavelet Transform). The cover image and secret image co-efficient values are embedded by 512*512 using fusion process techniques. We applied various combinations of DWT and IWT on both images and obtained a good quality stego images. The both domain gives more secure with secret key and certain robustness of our algorithm. ThisDual Wavelet Transform Used in Color Image Steganography Method(DWTSM)model provides high capacity and security. This proposed algorithm is tested with image quality parameters and compared to other algorithms. The experimental Results show that dual based approach achieved high capacity and high security of our system and also improves the performance of steganography system. The proposed algorithm achieved high PSNR ratio and other parameter values achieved the optimal solution and this method compared to other combination of transform.

A HIGH SECURE AND ROBUST IMAGE STEGANOGRAPHY USING DUAL WAVELET AND BLENDING MODEL

Steganography is an ability of concealing informati on inside the cover in such a way it looks like sim ple cover though it has concealed information. There are many techniques to carry out steganography on electroni c media, most especially audio and image files. In this meth od, we proposed a high secure steganography scheme hiding a 256◊256 size gray secret image into a 512◊512 size gray cover image with different combination of Discrete Wavelet Transform (DWT) and Integer Wavelet Transform (IWT). Pixel Value Adjustment (PVA) is first performed on cover image. The secret image values are scrambled by using Arnold transform. The DWT /IWT is applied on both cover and scrambled secret image. Blending process is applied to both images and compute Inverse DWT/IWT on the same to get the stego image. The extraction model is actually the reverse proce ss of the embedding model. Different combination of DWT/IWT transform is performed on the scrambled secret image and cover image to achieved high security and robustness. Hybrid transform combination approach and case analysis provided the various hiding environment. E xperimental results and case study provided the ste go-image with perceptual invisibility, high security and cer tain robustness.

SVM based biometric authorization system by video analysis of human gait

Biometric Systems to recognize authorized person when they enter into a surveillance area has received growing attention in modern era. In this paper human gait is used as a discriminative feature for authorization. Initially background modeling is done from a video sequence and the foreground moving objects in the individual frames are segmented using the background subtraction algorithm. Then gait representing spatial, temporal, and wavelet features are extracted and fused for training and testing the multiclass support vector machine model (SVM). The proposed system is evaluated using side view videos of Chinese National Laboratory of Pattern Recognition (NLPR) gait database and experimental results demonstrate the effectiveness of our approach.

Dual Wavelet Transform Used in Color Image Steganography Method

Steganography is an art of activity the data within the cover like the simplest way that, its like straightforward cover, though its hidden information. In this paper, secret image is activity into two completely different domains like as IWT (Integer Wavelet Transform) and DWT (Discrete Wavelet Transform). The cover image and secret image coefficient values square measure embedded by 512*512 exploitation fusion process techniques. We tend to applied numerous combinations of DWT and IWT on each images and obtained a decent quality stego images. The each domain offers safer with secret key and sure robustness of our algorithm. This dual wavelet transform utilized in Color Image Steganography Method (DWTSM) in Blue Channel (B-Channel) provides high capability and security. This proposed algorithm is tested with image quality parameters and compared to other algorithms. The experimental Results show that dual based approach achieved high capability and high security of our system in B-Channel and conjointly improves the performance of steganography system. The proposed algorithm achieved high PSNR quantitative relation 45 to 55 and other parameter values achieved the best solution and this methodology compared to different combination of transform.

Gait based Gender Identification using Statistical Pattern Classifiers

Gait based gender identification has received a great attention from researchers in the last decade due to its potential in different applications. This will help a human identification system to focus only on the identified gender related features, which can improve search speed and efficiency of the retrieval system by limiting the subsequent searching space into either a male database or female database. In this paper after preprocessing, four binary moment features and two spatial features are extracted from human silhouette. Then the extracted features are used for training and testing two different pattern classifiers kNearest Neighbor (kNN) and Support Vector Machine(SVM). Experimental results show superior performance of our approach to the existing gender classifiers. To evaluate the performance of the proposed algorithm experiments have been conducted on NLPR database.