Title :
Millimeter Wave Image Restoration Using a Modified T-S Fuzzy Neural Network
Author :
Shang, Li ; Su, Pingang ; Huai, Wenjun ; Sun, Zhanli
Author_Institution :
Dept. of Electron. Inf. Eng., Suzhou Vocational Univ., Suzhou, China
Abstract :
In order to reduce the much unknown noise and improve the resolution of images acquired by millimeter wave imaging system, and combined the advantages of fuzzy theory and neural network, a new MMW image restoration method using a modified Takagi-Sugeno (T-S) fuzzy neural network model is proposed in this paper. The modified T-S fuzzy neural network has the excellent ability of adaptive learning, nonlinear expression and pattern classification. Utilizing this T-S fuzzy neural network model, it is needless to know the degraded model of a MMW image in advance and the noise in MMW images can be detected efficiently. Using the single noise ratio (SNR) to measure the quality of restoration images, simulation results show that the T-S fuzzy neural network can restore the MMW image satisfactorily.
Keywords :
fuzzy neural nets; image classification; image restoration; learning (artificial intelligence); millimetre wave imaging; MMW image restoration method; SNR; adaptive learning; millimeter wave image restoration; millimeter wave imaging system; modified Takagi-Sugeno fuzzy neural network model; nonlinear expression; pattern classification; quality measurement; single noise ratio; unknown noise reduction; Educational institutions; Filtering; Fuzzy neural networks; Image restoration; Imaging; Neural networks; Noise; Image restoration; T-S model; degraded image; fuzzy neural network; millimeter wave;
Conference_Titel :
Computer Science & Service System (CSSS), 2012 International Conference on
Conference_Location :
Nanjing
Print_ISBN :
978-1-4673-0721-5
DOI :
10.1109/CSSS.2012.61
