A B C D E F G H I J K L M N O P Q R S T U V W X Y Z All
Parasuraman, Kumar
- Security Based Speaker Verification for Lip-Password using Learning Multi-Boosted HMMS
Authors
1 Department of Information Technolgy in Manonmaniam Sundaranar University, Tirunelveli, TamilNadu, IN
2 Centre for Information Technology and Engineering, Manonmaniam Sundaranar University, Tirunelveli, Tamilnadu, IN
Source
Digital Image Processing, Vol 7, No 8 (2015), Pagination: 234-241Abstract
Lip password is composed of a password embedded with motions of lip and point out the characteristic of lip motion. To provides security of a speaker verification system by using private password and behavioral biometrics of a lip motion simultaneously. The target speaker saying wrong password then rejected and the target speaker saying correct password then detected. Here a Hidden Markov Model (HMM) learning approach based on multi boosted scheme is presented for a security speaker system. This method first extract the visual features and to characterize each frame. The lip password segmentation algorithm is used for the segmentation of lip sequences. Hidden Markov Models with boosting learning framework contains random subspace method and data sharing scheme. Finally, the lip-password is verified based on verification results provided by all the subunit learned from HMM based multi-boosted scheme and it will check whether the password is spoken by the speaker with the already-recorded password or not.Keywords
Lip Motion, HMM, GMM, RSM, DSS.- Graph Cut Based Method for Automatic Lung Segmentation for Tuberculosis by using Screening Method in Chest Radiographs
Authors
1 Manonmaniam Sundaranar University, Tirunelveli, TamilNadu, IN
2 Centre for Information Technology and Engineering, Manonmaniam Sundaranar University, Tirunelveli, Tamilnadu, IN
3 Centre for Information Technology and Engineering, Manonmaniam Sundaranar University, Tirunelveli, Tamilnadu, IN
Source
Digital Image Processing, Vol 7, No 9 (2015), Pagination: 285-291Abstract
In medical imaging technique tuberculosis is an important challenging approach. Most of the peoples affected by the tuberculosis and tuberculosis are a very big disease after the HIV in India. The mortality rate of the peoples is high by affecting tuberculosis. Chest radiographs are also called as chest x ray or CXR. By using graph cut segmentation method is used to extract the lung region and texture and shape features are classified by using binary classifier. The postero anterior is used to automatically detect the tuberculosis. The existing smear microscopy is slow and unreliable. The ROC curve is used to illustrate the performance of the binary classifier. Three terms are classified as follows: Lung segmentation, feature computation, classification. Automated nodule detection is more nature applications of decision support/automation for CXR and CT.
Keywords
CXR, CR, Radiography, HIV, TB.- A Novel Approach for MRI Brain Image Segmentation using Local Independent Projection Model
Authors
Source
Digital Image Processing, Vol 8, No 7 (2016), Pagination: 237-243Abstract
Brain tumor segmentation is an important process for tumor identification and assists to planning for further treatment. Although several brain tumor segmentation methods are existing, still efficient brain tumor segmentation is challenging in medical field. To achieve the high detection accuracy of tumor part with lower error rate, we have lots of enhancing techniques for the tumor segmentation methods. In this paper, we propose a Novel approach for brain Image Segmentation using Local Independent Projection Model for MRI images. The main objective of this paper is to develop a system for brain segmentation based on local independent projection classification. The proposed system has 2 stages such as training and testing and it has 4 steps such as pre-processing, feature extraction, segmentation and post processing. Preprocessing is done before starting process. In feature extraction the related features from the input data to be retrieved. This project proposed the patch based method used for feature extraction. Then apply the local independent projection classification. The segmentation of brain tumor can be assumed as a multiclass classification problem. Resolving this problem by One-Versus-All (OvA) strategy In this strategy, a classifier is trained for each class to differentiate a class from all other classes. In this classification first construct the dictionary based on original samples in training set. Then present the sparse representation using locally linear representation. Dictionary construction is performed by using manually labeled original samples in a training set. In order to achieve classification scores, Softmax regression model is used. By using learned as well as without learned softmax regression model, classification accuracy was tested. Finally calculate the classification score computation. After post processing to get the final results.