Iris recognition Using Fast Walsh Hadamard Transform Based feature Space
Journal Title: International Journal of Science Engineering and Advance Technology - Year 2016, Vol 4, Issue 9
Abstract
The significance of Iris detection and recognition in area of bioinformatics and pattern recognition has been increased from last few decades. Looking at the importance of Iris detection and recognition, we propose a robust, stable and reliable computational model. Features are extracted from iris images using two different approaches such as Hilbert transform and Fast wavelet Hadamard Transform (FWHT).Random forest is used as a classification algorithm. 5-folds cross validation test is applied to evaluate the performance of K-nearest neighbor. Among three feature spaces, FWHT feature space has achieved promising results. The success rate of K-nearest neighbor on FWHT feature space is 94.4%. After examining the results, we have observed that our model might be useful and helpful for iris detection in future work.
Authors and Affiliations
Shahid Akbar1| Department of Computer Science, Abdul Wali Khan University, Mardan, KPK, PakistanŮ shahidakbarcscs@gmail.com, Maqsood Hayat| Department of Computer Science, Abdul Wali Khan University, Mardan, KPK, Pakistan, Mohammad Sohail| Department of Physics, Abdul Wali Khan University, Mardan, KPK, PakistanŮ sohail.dagiwal@gmail.com, Haroon Khan| Department of Computer Science, Abdul Wali Khan University, Mardan, KPK, Pakistan, haroon_rajjar@yahoo.com
Scheduling Cached-Content in Wireless Networks With Elastic and Inelastic Traffic
With the exponential growth of wireless content access, the need for content placement and scheduling at wireless base stations increased rapidly. We study a system under which users are divided into clusters based o...
We propose a shared data based algorithm that analytically chooses the optimal feature for classification. This common information based component choice calculation can manage straightly and nonlinearly subordinate...
A New Multilevel Inverter Topology of D-STATCOM to Compensate Reactive and Nonlinear Loads
This work proposes an enhanced hybrid distribution static compensator (D-STATCOM) topology to address some practical issues like power loss, filter size, performance etc. A LCL filter has been utilized at the front e...
Designing Of Bidirectional Dc-Dc Converter For High Power Application With Current Ripple Reduction Technique
This paper presents the design of bidirectional dc/dc converter with three bridges. In this configuration, only a single bridge is used for bidirectional control and two bridges are used for electrical isolation and f...
Sensor networks are utilized as a part of various application spaces, for example, digital physical framework frameworks, natural checking, control lattices, and so on. Information are created at an extensive number...