电子与信息学报
電子與信息學報
전자여신식학보
JOURNAL OF ELECTRONICS & INFORMATION TECHNOLOGY
2013年
4期
812-819
,共8页
朱波*%汶德胜%王飞%李华%宋宗玺
硃波*%汶德勝%王飛%李華%宋宗璽
주파*%문덕성%왕비%리화%송종새
图像处理%Bayer格式%去马赛克%字典学习%高斯混合模型
圖像處理%Bayer格式%去馬賽剋%字典學習%高斯混閤模型
도상처리%Bayer격식%거마새극%자전학습%고사혼합모형
Image processing%Bayer pattern%Demosaicking%Dictionary learning%Gaussian Mixture Model (GMM)
利用单片探测器获取彩色图像,插值算法的优劣对结果起着决定性的作用.为了改善恢复效果,该文设计了一种基于字典学习的非线性 Bayer 格式图像彩色插值算法.根据图像梯度的变化,首先,在上下左右方向利用局部方向插值方法(LDI)对Bayer格式图像进行合并计算,用高斯混合模型(GMM)分类法训练字典,运用主分量分析(PCA)方法提取训练结果中的主要分量为学习提供样本,通过学习,得到R , B通道缺失的G$分量.然后,应用G$分量,插值得到另外两种缺失分量μR和μB ,从而得到彩色图像.选取McMaster图像集作为字典,分别用算法对标准图像和使用DALSA公司彩色CMOS探测器开发的相机实际拍摄的图像进行插值恢复,较其它几种算法,视觉上伪彩色最少,峰值信噪比最优.整体性能优于现有的很多其它插值算法.
利用單片探測器穫取綵色圖像,插值算法的優劣對結果起著決定性的作用.為瞭改善恢複效果,該文設計瞭一種基于字典學習的非線性 Bayer 格式圖像綵色插值算法.根據圖像梯度的變化,首先,在上下左右方嚮利用跼部方嚮插值方法(LDI)對Bayer格式圖像進行閤併計算,用高斯混閤模型(GMM)分類法訓練字典,運用主分量分析(PCA)方法提取訓練結果中的主要分量為學習提供樣本,通過學習,得到R , B通道缺失的G$分量.然後,應用G$分量,插值得到另外兩種缺失分量μR和μB ,從而得到綵色圖像.選取McMaster圖像集作為字典,分彆用算法對標準圖像和使用DALSA公司綵色CMOS探測器開髮的相機實際拍攝的圖像進行插值恢複,較其它幾種算法,視覺上偽綵色最少,峰值信譟比最優.整體性能優于現有的很多其它插值算法.
이용단편탐측기획취채색도상,삽치산법적우렬대결과기착결정성적작용.위료개선회복효과,해문설계료일충기우자전학습적비선성 Bayer 격식도상채색삽치산법.근거도상제도적변화,수선,재상하좌우방향이용국부방향삽치방법(LDI)대Bayer격식도상진행합병계산,용고사혼합모형(GMM)분류법훈련자전,운용주분량분석(PCA)방법제취훈련결과중적주요분량위학습제공양본,통과학습,득도R , B통도결실적G$분량.연후,응용G$분량,삽치득도령외량충결실분량μR화μB ,종이득도채색도상.선취McMaster도상집작위자전,분별용산법대표준도상화사용DALSA공사채색CMOS탐측기개발적상궤실제박섭적도상진행삽치회복,교기타궤충산법,시각상위채색최소,봉치신조비최우.정체성능우우현유적흔다기타삽치산법.
@@@@Demosaicking is important for the quality of digital images in resource-constrained single chip devices. This paper presents an improved dictionary learning-based color demosaicking algorithm. Firstly, an initial interpolation is applied to the R , B channel by Local Directional Interpolation (LDI) and fused by analysis the joint distribution of the gradient. Gaussian Mixture Model (GMM)-based clustering is used to classify dictionary image into different classes. The Principal Component Analysis (PCA) is performed on these classes to choose the principal components for the dictionary construction. And then, dictionary learning is applied to obtain the interpolated G$ and the lost ?R and ?B are interpolated by the help of the reconstructed G$ , accordingly. Since R$,G$ and B$of the given pixels are better represented, the whole image can be reconstructed accurately. Taking McMaster color image dataset as dictionary, standard image and image from DALSA CMOS camera are used for effect evaluation of the demosaicking algorithm. Experimental results prove that the proposed algorithm outperforms some state-of-the-art demosaicking methods both in PSNR measure and visual quality.