(For USM Staff/Student Only)

EngLib USM > Ω School of Electrical & Electronic Engineering >

Large-scale integrated circuit (lsi) implementation of 2-dimensional discrete fourier transform (2d-dft) for phase only correlation(poc)/Teo Soon Tian

Large-scale integrated circuit (lsi) implementation of 2-dimensional discrete fourier transform (2d-dft) for phase only correlation(poc)_Teo Soon Tian_E3_2011_NI
Projek ini adalah berkenaan dengan litar bersepadu skala besar bagi perlaksanaan 2 Dimensi-Transformasi Fourier diskrit (2D-DFT) untuk fasa-sahaja korelasi(POC). POC merupakan satu teknik yang digunakan untuk membanding persamaan antara dua imej dengan menjalankan sintesis atas fasa imej yang diperolehi dari 2D-DFT. POC boleh digunakan dalam pengecaman biometrik seperti pengenalpastian urat jari. Sementara, 2D-DFT merupakan alat pemprosesan imej yang digunakan untuk menterjemahkan imej kepada komponen amplitud dan komponen fasa. Keseluruhan seni bina 2D-DFT untuk masukan 5×5 telah dikodkan dalam Verilog HDL. Simulasi dan sintesis atas seni bina ini telah dilaksanakan dengan menggunakan alat-alat Synopsys , Verilog Compiler Simulator(VCS) dan Design Vision, kesemuanya perlu bersepadu dengan ARM Artisan Silterra Digital 0.18µm. Setiap masukan adalah terhad dari 0 hingga 255 iaitu 25 bit masukan yang termasuk 1 bit tanda. Penempatan dan penghalaan telah dilaksanakan dengan menggunakan Cadence Encounter dan layout fizikal tanpa pokok jam yang mempunyai keluasan 4.168mm2 telah dihasilkan. Oleh itu, seni bina 2D-DFT untuk masukan 5×5 yang boleh menghasilkan 25 keluaran benar dan khayalan telah berjaya dilaksanakan dalam LSI dengan purata ketepatan 99.696%. ___________________________________________________________________________________ This project is about the Large Scale Integrated Circuit (LSI) implementation of 2 dimensional-Discrete Fourier Transform (2D-DFT) for Phase-Only Correlation (POC). POC is a technique to compare the similarity of two images by carrying out synthesize refer to the phase of images which obtained from 2D-DFT. POC is important because it can be used on many biometric recognition and image matching applications such as finger vein recognition. Meanwhile, 2D-DFT is an important image processing tool in POC that can transform the image into its frequency domain representation such as amplitude and phase components. The whole 2D-DFT architecture for a 5×5 input has been coded in VERILOG HDL and simulation and synthesis has been carried out by using Synopsys tools, Verilog Compiler Simulator(VCS) and Design Vision, respectively, with ARM Artisan Silterra Digital 0.18µm . Each input is limited from 0 to 255 , which is 25 bit input including 1 bit of sign bit. Placement and routing has been carried out by using Cadence Encounter and a physical layout of 4.168mm2 without clock tree produced. Hence, the 2D-DFT architecture for a 5 × 5 input that can produce 25 real and imaginary outputs with average accuracy of 99.696% has been successful implemented on LSI.
Contributor(s):
Teo Soon Tian - Author
Primary Item Type:
Final Year Project
Language:
English
Subject Keywords:
image ; Large Scale Integrated Circuit (LSI) ; biometric recognition
First presented to the public:
4/1/2011
Original Publication Date:
2/28/2020
Previously Published By:
Universiti Sains Malaysia
Place Of Publication:
School of Electrical & Electronic Engineering
Citation:
Extents:
Number of Pages - 150
License Grantor / Date Granted:
  / ( View License )
Date Deposited
2020-02-28 10:02:26.422
Date Last Updated
2020-02-28 10:32:46.43
Submitter:
Nor Hayati Ismail

All Versions

Thumbnail Name Version Created Date
Large-scale integrated circuit (lsi) implementation of 2-dimensional discrete fourier transform (2d-dft) for phase only correlation(poc)/Teo Soon Tian1 2020-02-28 10:02:26.422