Publications (3)0 Total impact
- [Show abstract] [Hide abstract]
ABSTRACT: Multibiometrics is the usage of more than one physiological or behavioral characteristic to identify an individual. Multibiometrics is advantageous over unibiometrics as it is resilience to spoofing and has low False Acceptance Rate (FAR). However Multibiometrics requires storage of multiple biometric templates for each user, which results in increased risk to user privacy and system security. This paper will discuss the concept off biometrics and biometric system, multimodal biometric fusion techniques, crypto-biometrics and an algorithm for session key generation for secure communication of data.
Conference Paper: Harnessing healthcare data security in cloud[Show abstract] [Hide abstract]
ABSTRACT: Healthcare sector has a large volume of information like health records of patients that needs to be collected and stored. These Patient Health Records when equipped digitally become Electronic Health Records (EHRs). This large amount of data collected as EHRs has to be stored, processed and updated frequently. Also the data of the patients has to be available whenever and wherever necessary. Cloud provides the necessary infrastructure now-a-days at low cost and better quality. Cloud computing when used in Healthcare sector lowers the cost of storing, processing and updating with improved quality. But the security of data in the cloud is of concern today. The electronic health records consist of images of patients which are of high security concern. The EHRs include the patient scan images, x-rays etc. which are the patient's private data. Providing security for a large volume of data with efficiency is required. In this paper we present a novel mechanism by which the EHR images are secured efficiently and the privacy of the patient's data is preserved. Since the most confidential data of the patients are in the form of images, more care must be taken to secure the images. This is done by reducing the images into pixels and then encrypting the pixels using Paillier Cryptosystem. After the encryption it is stored in the cloud. The original data can be decrypted from the cloud with the private key. The key can be made available to the physicians and other specialists as required by the hospital.
- [Show abstract] [Hide abstract]
ABSTRACT: Unimodal biometric systems have to contend with a variety of problems such as noisy data, intra-class variations, restricted degrees of freedom, non-universality, spoof attacks, and unacceptable error rates. Some of these limitations can be addressed b y deploying multimodal biometric systems that integrate the evidence presented by multiple sources of information. This paper discusses a multibiometric based authentication system with GUI interface. The proposal includes an extraction algorithm to extract features from given finger print and a palm print feature extraction algorithm to extract palm print features. Then an integration of these two algorithms to perform a multibiometric authentication was done. By processing the test image of a person, the identity of the person is displayed with his/her own image. By the proposed fingerprint and palm print algorithm, it is found that it has less computation time and occupies less memory space compared to the existing algorithms. The Fingerprint and palm print matching results for the proposed methods are validated and the system integrity for all the cases were evaluated.
Chennai, Tamil Nadu, India
- Department of Electronics and Communication Engineering
Madras Institute of TechnologyChennai, Tamil Nadu, India