TY - GEN
T1 - A Novel VLSI Architecture for Barrel Distortion Correction
AU - Ni, Chi Ting
AU - Chang, Chen An
AU - Liao, Jia Wei
AU - Chen, Pei Yin
N1 - Publisher Copyright:
© 2020 IEEE.
Copyright:
Copyright 2021 Elsevier B.V., All rights reserved.
PY - 2020/12
Y1 - 2020/12
N2 - Fisheye camera has been used in many field such as unmanned aerial vehicle, car reversing displays, endoscope, ect. Because the focal length of fisheye camera is too short, it will cause the barrel distortion. So far, most of the barrel distortion correction VLSI architectures are based on polynomial model or division model, and the parameters of polynomial model and division model are usually obtained from the camera parameters, which means that we can only perform barrel distortion correction on images taken by the same camera. So we have implemented the VLSI architecture of this algorithm [1] which can find out the parameters of polynomial model and division model by images. This algorithm [1] is only applied to the software, but the VLSI architecture has not been implemented yet. Thus, we are the first one who implement the algorithm into the VLSI architecture. Our design contains 102868 gate counts and works with a clock period of 10 ns and operates at a clock rate of 100 MHz by using TSMC's 90 nm cell library.
AB - Fisheye camera has been used in many field such as unmanned aerial vehicle, car reversing displays, endoscope, ect. Because the focal length of fisheye camera is too short, it will cause the barrel distortion. So far, most of the barrel distortion correction VLSI architectures are based on polynomial model or division model, and the parameters of polynomial model and division model are usually obtained from the camera parameters, which means that we can only perform barrel distortion correction on images taken by the same camera. So we have implemented the VLSI architecture of this algorithm [1] which can find out the parameters of polynomial model and division model by images. This algorithm [1] is only applied to the software, but the VLSI architecture has not been implemented yet. Thus, we are the first one who implement the algorithm into the VLSI architecture. Our design contains 102868 gate counts and works with a clock period of 10 ns and operates at a clock rate of 100 MHz by using TSMC's 90 nm cell library.
UR - http://www.scopus.com/inward/record.url?scp=85102183812&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85102183812&partnerID=8YFLogxK
U2 - 10.1109/ICS51289.2020.00050
DO - 10.1109/ICS51289.2020.00050
M3 - Conference contribution
AN - SCOPUS:85102183812
T3 - Proceedings - 2020 International Computer Symposium, ICS 2020
SP - 209
EP - 213
BT - Proceedings - 2020 International Computer Symposium, ICS 2020
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2020 International Computer Symposium, ICS 2020
Y2 - 17 December 2020 through 19 December 2020
ER -