本文選題：夏克-哈特曼波前傳感器 + 高斯光斑�。� 參考：《物理學(xué)報(bào)》2017年09期

【摘要】：針對夏克-哈特曼波前傳感器探測系統(tǒng)中噪聲隨時間及空間變化頻率較快的特點(diǎn),為了準(zhǔn)確估計(jì)系統(tǒng)的最優(yōu)閾值,根據(jù)高斯光斑與噪聲的分布特性,提出一種以滑動窗口內(nèi)像素均值及圖像信號的局部梯度作為參數(shù),構(gòu)造關(guān)于噪聲權(quán)重函數(shù)的方法,由此獲得子孔徑閾值的最優(yōu)估計(jì)值,并詳細(xì)分析了算法的基本原理和實(shí)現(xiàn)過程.以典型處理方法獲取的閾值與理論最優(yōu)閾值的誤差作為評價標(biāo)準(zhǔn),仿真和實(shí)驗(yàn)結(jié)果表明本文提出的閾值估計(jì)方法在不同信噪比、不同光斑大小的條件下,均能取得優(yōu)于典型閾值處理方法獲得的結(jié)果,且與理論最優(yōu)閾值的誤差小于10%.
[Abstract]:In order to estimate the optimal threshold of the system, according to the distribution characteristics of Gao Si spot and noise, in order to estimate the optimal threshold of the system accurately, aiming at the characteristics of fast frequency variation of noise with time and space in the detection system of Shack-Hartmann wavefront sensor, In this paper, a method of constructing the noise weight function using the pixel mean in the sliding window and the local gradient of the image signal as parameters is proposed, and the optimal estimation of the subaperture threshold is obtained. The basic principle and implementation process of the algorithm are analyzed in detail. The error between the threshold obtained by the typical processing method and the theoretical optimal threshold is taken as the evaluation criterion. The simulation and experimental results show that the proposed threshold estimation method is based on different SNR and different spot sizes. The results obtained are better than those obtained by the typical threshold processing method, and the error from the theoretical optimal threshold is less than 10%.
【作者單位】： 中國科學(xué)院自適應(yīng)光學(xué)重點(diǎn)實(shí)驗(yàn)室;中國科學(xué)院光電技術(shù)研究所;中國科學(xué)院大學(xué);
【分類號】：TP391.41;TP212

【相似文獻(xiàn)】

相關(guān)期刊論文 前5條

1 于洪春;鄧意成;鄭喜鳳;;面積約束下的最優(yōu)閾值法分割LED像素點(diǎn)陣[J];液晶與顯示;2012年05期

2 孫艷歌;邵罕;;基于改進(jìn)遺傳算法的最優(yōu)閾值圖像分割算法[J];信息系統(tǒng)工程;2010年06期

3 舒紅平,蔣建民;基于灰度最優(yōu)閾值的圖像分割方法及應(yīng)用[J];重慶工商大學(xué)學(xué)報(bào)(自然科學(xué)版);2003年04期

4 張建明;楊忠;李巍;;改進(jìn)KNN-SVM的性別識別[J];計(jì)算機(jī)工程與應(yīng)用;2009年04期

5 ;[J];;年期

，

本文編號：1928673