Title :
A noise-agnostic self-adaptive image processing application based on evolvable hardware
Author :
Mora, Javier ; Gallego, Angel ; Otero, Andres ; Lopez, B. ; de la Torre, E. ; Riesgo, T.
Author_Institution :
Center of Ind. Electron. - CEI, Univ. Politec. de Madrid, Madrid, Spain
Abstract :
Evolvable Hardware (EH) is a technique that consists in the usage of reconfigurable hardware devices whose configuration is controlled by an Evolutionary Algorithm (EA). In this demo, a proof of concept is shown in which an EH-based filter is used to remove different types of noise from an image. This filter, together with all the needed control and a processor to run the EA, is fully implemented in an FPGA without needing any external control.
Keywords :
evolutionary computation; field programmable gate arrays; filtering theory; image denoising; EA; EH-based filter; FPGA; evolutionary algorithm; evolvable hardware; noise-agnostic self-adaptive image processing; Libraries; Training;
Conference_Titel :
Design and Architectures for Signal and Image Processing (DASIP), 2013 Conference on
Conference_Location :
Cagliari