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
Sumathi, S.
- Palatal Rugae and Dental Work Information-Tools of Human Identification
Authors
1 Sri Sairam Engineering College, Chennai-44, IN
Source
Biometrics and Bioinformatics, Vol 4, No 7 (2012), Pagination: 318-323Abstract
Dental biometrics is used in forensic dentistry to identify or verify persons. This paper presents a method for human identification based on dental work information along with palatal rugae recognition which is considered to be unique for each person just like the finger prints. The proposed method works with three main processing steps: segmentation (feature extraction), creation of a dental code, and matching. In the segmentation step, seed points of the dental works are detected by thresholding. The final segmentation is obtained with a snake (active contour) algorithm. The dental code is defined from the position (upper or lower), the size of the dental works, and distance between neighboring dental works. The matching stage is performed with the Edit distance (Levenshtein distance). The palatal rugae is also considered. By superimposing the rugae patterns of unknown with those in database the identification is made easier.
- Abduction of Newborn Infants Using Footprint Recognition
Authors
1 Sri Sairam Engineering College, Chennai-44, IN
Source
Biometrics and Bioinformatics, Vol 4, No 7 (2012), Pagination: 324-329Abstract
Identify or authenticate the person using different biometric traits. Whenever sensational crimes such as abduction of infants or baby switching occurs, need for biometric identification technique increases. For this issue here we propose a novel online newborn personal authentication system based on footprint recognition. A newborn footprint database is established to examine the performance of the system and promising experimental results demonstrate the effectiveness of the proposed system. The entire process can be implemented in software. For the software implementation, there are two steps one is image acquisition which is done by capturing the newborn footprint images using a digital camera and the other is preprocessing the obtained image which involves various stages like orientation, normalization, scale normalization, ROI extraction and recognized by ordinal code and hamming distance. These experiments were conducted on a personal computer using Matlab R2009a. This method represents a reliable, expeditious and cost efficient method for establishing probable personal identity which is thus encouraged by Federal Bureau of Investigation (FBI).