Title :
Image recognition via discrete Hopfield neural network
Author :
Lin, Qian ; Cai, Peng ; Zhang, Feng
Author_Institution :
Sch. of Electron. Inf. & Electr. Eng., Shanghai Jiao Tong Univ., Shanghai, China
Abstract :
The contemporary digital society is filled with the image world. Image digitization not only standardizes the image storage but also benefits the image recognition based on the computational methods. By the nature of Hopfield neural network, associative memory based on this mechanism can be effectively adopted for the image recognition application. This paper proposes the design and the prototype implementation of image recognition through discrete Hopfield neural network. Experiments demonstrate that our approach can recover the dirty image such as fragment or noisy picture by a high percentage of possibility.
Keywords :
Hopfield neural nets; image recognition; associative memory; computational method; discrete hopfield neural network; image digitization; image recognition; image storage; Educational institutions; Pixel; DHNN; Hopfield; associative memory; image recognition;
Conference_Titel :
Advances in Energy Engineering (ICAEE), 2010 International Conference on
Conference_Location :
Beijing
Print_ISBN :
978-1-4244-7831-6
DOI :
10.1109/ICAEE.2010.5557547