Open AccessIdentification of a Standardized Automobile Bio-metric Security System based on Accuracy and Response time: Applicable for the Indian Automobile Market
Author(s) -
Subhranil Ganguly,
Kirubanand VB
Publication year - 2019
Publication title -
international journal of innovative technology and exploring engineering
Language(s) - English
Resource type - Journals
ISSN - 2278-3075
DOI - 10.35940/ijitee.f3573.078919
Subject(s) - computer science , metric (unit) , identification (biology) , ibm , computer hardware , embedded system , real time computing , artificial intelligence , engineering , operations management , botany , materials science , biology , nanotechnology
This paper aims at the identification of a standard security system for commercial and personal vehicles installed on a remote-controlled unlocking device that promises a high accuracy without compromising on the response time. The proposed technology combines three bio-metric security systems on the basis of their performance and response times to make vehicles more secured. The paper compares the efficiencies of different bio-metric security systems based upon their mean accuracy and response times. Primary data have been collected using existing devices and technology, a detailed statistical comparison is done using computational tools like IBM SPSS 2.0 and Microsoft Excel using statistical concepts like ANOVA, Square of means, Descriptive statistics, Central tendencies, etc. The hardware required for this proposed security system is already available at reasonable cost and can be implemented in the field of automobile and a standard security system can be identified for use across all variants of vehicles universally for all the manufacturers. The performance of the bio-metric devices was measured using a 16-megapixel Sony camera IMX371 Exmor RS sensor with a pixel size of 1.0 micro-meter, mounted on a OnePlus 5T mobile phone for face recognition and a fingerprint sensor with a claimed unlock speed of 0.2 seconds mounted on the same device. Mantra MI S100 single iris scanner was used with a high-resolution sensor (CMOS) and captures images with a JPEG2000 compression format.