In the course, make iterated processing to get good result 并在此過程中,對濾波結(jié)果進(jìn)行迭代,以獲得優(yōu)化的結(jié)果。
3, on the base of the traditional spatial filtering, the author present, a new despeckle algorithm, that make use of iterated processing and correlated neighbourhood model, iterated filtering method of the sar image combining the correlated neighbourhood model with maximum a posteriori filter . first, a series of templates refecting direction information are established and every template is present for a kind of neighbour structure . then on the basis of sar images statistical property, the maximum a posteriori estimate of the real intensity under observation image values is got by bayes formulatio-n 3、針對傳統(tǒng)空間濾波器的不足,引入迭代處理和相關(guān)鄰域模型的概念,提出了基于相關(guān)鄰域模型的最大后驗(yàn)迭代濾波。該算法引用一系列反映局部邊界特征的鄰域模型,以描述圖像的細(xì)節(jié)。引入強(qiáng)度的先驗(yàn)概率分布模型,利用bayes方法,對各個(gè)結(jié)構(gòu)進(jìn)行實(shí)際強(qiáng)度的最大后驗(yàn)估計(jì)。