The vein-based biometric systems use vein patterns of the human body as a unique identity verification tool in various cloud- and Internet of Things (IoT)-based online and offline environments. It is very true that stealing of vein patterns cannot be possible, because it is the indivisible part of the human body. Currently, the vein patterns of a finger, face, palm, hand, palm dorsa, heart, and wrist of humans are being used for identification of a person in cloud- and Internet of Things (IoT)-based environments. In this paper, the vein patterns of palm dorsa are considered as biometric modal for authorizing a person to access cloud computing services. Nowadays, numerous researchers have developed their own algorithms to augment and extract vein images which can be useful for biometric authentication in cloud-based systems using MATLAB or Python or C languages for achieving their purposes. But, most of the methods are not optimized for hardware design purposes. The implementation of parallel processing capability in hardware is a major advantage to improve the algorithm speed performance. However, the design of the algorithm in terms of hardware is always a challenging task because of resource limitations, hardware complexity, and time utilization constraints.

    The aim of this paper is to develop hardware design of the improved vein enhancement and feature extraction algorithm in a cloud computing environment. The improved algorithm enhances the image quality, removes the noise, and finally detects the palm-dorsa vein pattern which can be used for identity verification if an end user is accessing available services of a private/public/protected cloud. The components of the proposed algorithm are designed using hardware description language for hardware realization. ModelSim-Altera (MSA) has been used as a hardware simulation platform for achieving the goal. First, the vein image is applied with the resample technique to remove the noise. Next, the segmentation technique consisting of difference of Gaussian and threshold is used to segment the veins. For hardware design of the resample technique, parallel pipeline hardware has been developed to improve processing time and high throughput. It is designed to perform bi-cubic computation in parallel pipeline hardware to accommodate fast processing and high throughput with fewer hardware resources which are the necessities of big data systems. Further, the temporary interpolation points (TIPs) are used to store the vein images in the form of data in external memory only once, and hence redundant memory access is avoided.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    An Efficient Palm-Dorsa-Based Approach for Vein Image Enhancement and Feature Extraction in Cloud Computing Environment


    Additional title:

    Unmanned Syst. Tech.


    Contributors:


    Publication date :

    2020-04-21


    Size :

    22 pages





    Type of media :

    Article/Chapter (Book)


    Type of material :

    Electronic Resource


    Language :

    English