Ultra fast cnn based Hardware Computing Platform Concepts for adas visual Sensors and Evolutionary Mobile Robots




Download 3,22 Mb.
Pdf ko'rish
bet14/81
Sana16.05.2024
Hajmi3,22 Mb.
#238917
1   ...   10   11   12   13   14   15   16   17   ...   81
Bog'liq
Alireza Fasih

1.3.2
 
Practical significance of the thesis
 
Stability and real time processing are key issues in embedded systems and smart sensors. 
The concept of this thesis has the high potential of being easily implementable on a chip as 
a smart visual sensor. Visual computing on CNN can fulfill the real-time requirements of 
ADAS concepts. Processing visual data at sensor level does reduce the latency and the 
memory bottleneck, and at the same time it does increase the speed of processing while 
increasing the system’s efficiency. The system does target the ADAS concepts in terms of
real time processing, size, flexibility in design and configurability. The encapsulation of 
different ADAS concepts on a chip reduces the overall system cost. Integrating a system on 
chip with other peripheral is easier than designing a complex system which is not uniform. 
This design is particularly useful for developers that want to build different realtime ADAS 
concepts on the same platform. 
1.4
 
List of publications in the frame of this thesis 
The following eighteen (18) publications have been made in the frame of the research 
work of this thesis; seven (7) of them did appear in international scientific journals 
whereby the remaining did appear in the proceedings of international conferences: 
1.
Fasih A., Schwarzlmüller C., Kyamakya K., Al Machot F, “Video Enhancement for ADAS Systems based on
FPG
A and CNN Platform,”
International Journal of Signal and Image Processing
, online: 
HyperSciences_Publisher, Vol 1, no. 3, pp. 169-176, May 2010. 
2.
M. R. Ghahrodi , A. Fasih, 

A Hybrid Method in Driver and Multisensor Data Fusion, Using a Fuzzy Logic 
Supervisor for Vehicle Intelligence
,”
International Conference on Sensor Technologies and Applications 
SENSORCOMM-IEEE Computer Society
, 2007, pp. 393-398. 


 
17 
3.
Fasih A., Chedjou J.C., Schwarzlmüller C., Kyamakya K, “New Computational Modelling for Solving Higher
Order 
ODE Based On FPGA,”
ISAST Transactions on Electronics and Signal Processing,
vol. 4, no. 1, pp. 58-
61, October 2010. 
4.
Fasih A., Schwarzlmüller C., Chedjou J.C., Kyamakya K, “Framework for FPGA Based Real
-time Machine 
Vision Direct Convolution Versus CNN,”
ISAST Transactions on Electronics and Signal Processing, 
vol. 4, no. 
1, pp. 1-5, October 2010. 
5.
Schwarzlmüller C., Fasih A., Latif M.A., Kyamakya K, “Adaptive Contrast Enhancement Involving CNN
-
based Processing for Foggy Weather,”
ISAST Transactions on Electronics and Signal Processing,
vol. 4, no. 1, 
pp. 24-35, October 2010. 
6.
Fasih A., Chedjou J.C., Kyamakya K, “Cellular Neural Network Trainer and Template Optimization for
Advanced Robot Locomotion, Based on Genetic Algorithm,” International
Journal of Intelligent Systems 
Technologies and Applications (IJISTA),
Geneva: Inderscience Enterprises Limited vol. 8, no. 1-4, pp. 36-45, 
2010. 
7.
Fasih A., Chedjou C. J., Kyamakya K. “Cellular Neural Networks
-Based Genetic Algorithm for Optimizing the 
Behavior of an 
Unstructured Robot,”
International Journal of Computational Intelligence Systems (IJCIS),
Paris: Atlantis Press, 2009, pp. 124-133. 
8.
Fasih A., Schwarzlmüller C., Chedjou J.C., Kyamakya K, “An Ultra
-fast and Adaptive Framework for FPGA-
Based Real-Time Machine Vision for Advanced Driver Assistance Systems: a CNN-Based Processing 
Architecture,”
Proceedings of Mallorca Workshop 2010 - Autonomous Systems.
Aachen: Shaker Verlag 
GmbH, vol. 6, pp. 28-34, October 2010. 
9.
Fasih A., Khan U., Chedjou J.C., Kyamakya K, “Efficient Control of the dynamic system ’Coordinated Urban
Traffic Lights System’ using a novel Reinforcement learning Concept,”
ISTET International Symposium on 
Theoretical Electrical Engineering.
vol. 24, pp.195-198, June 2009. 
10.
Schwarzlmüller C., Fasih A
., Kyamakya K, “Enhancement of Rainy Weather Degraded Images,”
Proceedings 
of Mallorca Workshop 2010 - Autonomous Systems, 
October 2010, pp.20-27. 
11.
Fasih A., Trong T. D., Chedjou J. C., Kyamakya K, “New Computational Modeling for Solving Higher Order
ODE b
ased on FPGA,”
Proceedings of INDS´09. The Second International Workshop on nonlinear Dynamics 
and Synchronization,
vol. 2, pp. 49-53, June 2009. 
12.
Fasih A., Kyamakya K. , Chedjou J. C., Umair A. K, “Benchmarking of the Traditional Genetic Algorithm
Method with a Novell Approach and a further novel Scheme, the 2-Point Crossover (F-
Crossover)”
ISTET 
International Symposium on Theoretical Electrical Engineerin,
June 2009, pp. 195-198. 
13.
Schwarzlmüller C., Al Machot F., Fasih A., Kyamakya K, “A Novel Support V
ector Machine Classification 
Approach Involving CNN for Raindrop Detection,”
ISAST Transactions on Computers and Intelligent 
Systems,
vol. 2, pp. 52-65, November 2010. 
14.
Tuan D. T., Chedjou J. C., Fasih A., Kyamakya K, “A Novel Method for Computing the Minim
um Spanning 
Tree and Solution Based on Cellular Neural Networks Implemented on Digital Platforms,”
The Second 
International Workshop on nonlinear Dynamcis and Synchronization, INDS'09.
Vol. 2, pp 23-29, June 2009. 
15.
Khan U., Fasih A. Kyamakya K., Chedjou J.C
. “Genetic Algorithm Based Template Optimization for a Vision
System Used for Obstacle Detection,”
ISTET International Symposium on Theoretical Electrical Engineering

June 2009, pp. 164-168. 
16.
Schwarzlmüller C., Fasih A., Kyamakya K. “Partial Differential E
quation based Image Inpainting by using 
Cellular Neural Networks,” It will appear in ISAST Transactions on Computers and Intelligent Systems
2011.
17.
Fasih A., Chedjou C. J., Kyamakya K., “Cellular Neural Network Trainer and Template Optimization for
Advance
d Robot Locomotion, Based on Genetic Algorithm,”
15th International Conference on Mechatronics 
and Machine Vision in Practice,
December 2008, pp. 317-322. 
18.
Fasih A., Sasanca P., Kyamakya K,

CNN based High Performance Computing Platform for Real Time Image 
Processing on GPU,” It will appear in 3rd International Workshop on nonlinear Dynamics and
Synchronization 2011 

INDS’11.


 
18 
Chapter 2 

Download 3,22 Mb.
1   ...   10   11   12   13   14   15   16   17   ...   81




Download 3,22 Mb.
Pdf ko'rish

Bosh sahifa
Aloqalar

    Bosh sahifa



Ultra fast cnn based Hardware Computing Platform Concepts for adas visual Sensors and Evolutionary Mobile Robots

Download 3,22 Mb.
Pdf ko'rish