Unit:
Κατεύθυνση Ηλεκτρονικός Αυτοματισμός (H/A)Library of the School of Science
Author:
Santorinaios Alexandros
Supervisors info:
Διονύσιος Ι. Ρεΐσης, Καθηγητής Τμήμα Φυσικής ΕΚΠΑ,
Νυσταζάκης Έκτορας, Καθηγητής Τμήμα Φυσικής ΕΚΠΑ,
Τζανακάκη Άννα, Αναπληρώτρια Καθηγήτρια Τμήμα Φυσικής ΕΚΠΑ
Original Title:
Ανάπτυξη εφαρμογών για επεξεργασία εικόνας σε FPGA.
Translated title:
Image processing applications on FPGAs
Summary:
The following thesis examines edge detection, which is a fundamental technique in image processing within the field of computer vision.
To be more precise, this thesis will examine the operation of the Sobel algorithm and it’s implementation in Matlab and on the FPGA software development platform, Vivado.
The implementation process will be divided into two stages.
In the first stage, a bit-accurate model or bit-true implementation will be constructed in MATLAB software. This model will serve as a simple implementation of the algorithm, as well as a method for verifying the results obtained from data processing on the FPGA. Additionally, preprocessing procedures will be performed in MATLAB, such as converting the image to grayscale and transforming input data into a suitable format for use on the FPGA.
Next, the same algorithm will be implemented in the hardware description language (HDL) VHDL. The AXI4-Stream communication protocol will be used to connect the structural units (entities). Finally, the FPGA processed data will be compared with the MATLAB results to verify the correctness of the implementation.
Main subject category:
Science
Keywords:
FPGA,SOBEL,MATLAB,AXI4-STREAM,VHDL
