自適應(yīng)去噪濾波器組合的訓(xùn)練與設(shè)計(jì)方法
A Hybrid Filter Training and Design Method for Adaptive Noise Cancellation
-
摘要: 該文通過(guò)建立噪聲信道模型提出一種對(duì)噪聲自適應(yīng)的去噪濾波器組合進(jìn)行訓(xùn)練和設(shè)計(jì)的方法。通過(guò)對(duì)訓(xùn)練圖像及噪聲信道輸出圖像的有監(jiān)督訓(xùn)練，建立自適應(yīng)于噪聲的濾波器組合模型，并可在隨后的應(yīng)用中采用該模型對(duì)通過(guò)噪聲信道的圖像進(jìn)行盲濾波達(dá)到濾除圖像噪聲和保留圖像細(xì)節(jié)的目的。為驗(yàn)證設(shè)計(jì)方法的有效性，采用高斯調(diào)制的加權(quán)中值濾波器組對(duì)信道中常見(jiàn)的椒鹽噪聲和均勻分布脈沖噪聲進(jìn)行濾除，效果較現(xiàn)有方法有明顯優(yōu)勢(shì)。該設(shè)計(jì)方法可以推廣應(yīng)用到其它具有適應(yīng)性的去噪濾波器之上，使它們對(duì)不同類型、不同強(qiáng)度的信道噪聲進(jìn)行更柔性化的處理。Abstract: A hybrid filter training and design method for adaptive image noise cancellation with establishing noise channel model is presented in this paper. Trainings are performed with input and output test images for channels of different types and/or different intensities to establish channel-adaptive hybrid filter models. In practice the image transferred through the specific channel is filtered blindly by corresponding model to maintain detail and eliminate noise simultaneously. For certificating the efficiency of this design method, Gaussian weighted median filters are adopted to remove well known channel noise, i.e., pepper and salt noise and uniform distributed impulse noise, in this framework. The results outperform some prior methods markedly. This design method can be generalized to other filters with adaptability to treat different channel noise flexibly.
-
Justusson B J, Huang T S. Two Dimensional Digital Signal Processing II. Berlin: Springer, 1991: 161-196.[2]Eng H L, Ma K K. Noise adaptive soft-switching median filter[J].IEEE Trans. Image Processing.2001, 10(2):242-251[3]Lee C S, Kuo Y H, Yu P T. Weighted fuzzy mean filters for image processing[J].Fuzzy Sets and Systems.1997, 89:157-180[4]Rahman S M, Hasan M K. Wavelet-domain iterative center weighted median filter for image denoising[J].Signal Processing.2003, 83:1001-1012[5]Muneyasu M, Nishi N, Hinamoto T. A new adaptive center weighted median filter using counter propagation networks[J].Journal of the Franklin Institute.2000, 337:631-639[6]Lim J S. Two-Dimensional Signal and Image Processing. Englewood Cliffs, NJ: Prentice Hall Press, 1990, chapter 9.[7]Arakawa K. Median filter based on fuzzy rules and its application to image restoration[J].Fuzzy Sets and Systems.1996, 77:3-13 -
計(jì)量
- 文章訪問(wèn)數(shù): 2249
- HTML全文瀏覽量: 157
- PDF下載量: 744
- 被引次數(shù): 0
下載:
