北京生物醫(yī)學(xué)工程

基于互信息與邊緣梯度相結(jié)合的多模醫(yī)學(xué)圖像配準(zhǔn)方法

Multimodality Medical Image Registration Based on Mutual Information Combined with Edge Gradient

作者：齊玲燕王俊

單位：南京郵電大學(xué)，圖像處理與圖像通信江蘇省重點(diǎn)實(shí)驗(yàn)室（南京210003）

關(guān)鍵詞：模極大值；邊緣檢測(cè)；互信息；梯度相似性；醫(yī)學(xué)圖像配準(zhǔn)

分類號(hào)：

出版年·卷·期（頁(yè)碼）：2011·30·4（359-362）

摘要：

針對(duì)傳統(tǒng)互信息配準(zhǔn)方法未利用圖像空間信息的缺點(diǎn),本文研究了圖像邊緣信息的梯度相似性。首先采用小波模極大值邊緣檢測(cè)提取出圖像邊緣，提出將邊緣圖像的梯度相似性系數(shù)與傳統(tǒng)的互信息相乘作為圖像配準(zhǔn)的目標(biāo)函數(shù)。然后通過(guò)使用Powell優(yōu)化算法對(duì)目標(biāo)函數(shù)進(jìn)行尋優(yōu),得出配準(zhǔn)變換參數(shù)。最后在互信息的基礎(chǔ)上引入圖像邊緣梯度信息,突出了全局最優(yōu)解。實(shí)驗(yàn)結(jié)果表明，該方法可以得到精確、有效的配準(zhǔn)結(jié)果。

To solve the drawback of registration based on typical mutual information neglecting the spatial information of image，this article studies the edge gradient similarity of image.Firstly the edge of image was extracted by calculating the wavelet transform modular maximum，and the gradient similarity coefficients of edge image was calculated and used to multiply by the mutual information to form the final registration metric.Then the registration transformation parameters were obtained by using Powell algorithm for optimizing the objective function.And finally the global optimal solution was captured by combining mutual information with edge gradient information of the images.Experimental results showed that the algorithm had high precision and effectiveness.

參考文獻(xiàn)：

［1］Penny GP，Weese J,Little JA,et al.A comparison of similarity measures for use in 2D-3D medial image registration ［J］.IEEE Transactions on Medical Imaging，1998，17 (4):586-595.
［2］諶安軍,陳煒,毛士藝.一種基于邊緣的圖像配準(zhǔn)方法［J］.電子與信息學(xué)報(bào),2004,24(5):679-684.
［3］劉青芳,李月娥.基于改進(jìn)互信息結(jié)合邊緣互方差的醫(yī)學(xué)圖像配準(zhǔn)方法研究［J］.測(cè)試技術(shù)學(xué)報(bào),2009,23(6):535-539.
［4］Lowe DG.Distinctive Image Features from Scale-Invariant Keypoints ［J］.International Journal of Computer Vision，2004，60(2):91-110.
［5］Wells III WM,Viola P,Atsumi H，et al.Multi-modal volume registration by maximization of mutual information ［J］.Medical Image Analysis,1996,1 (1):35-51.
［6］Maes F，Collignon A，Vandermeulen D,et al.Multimodality image registration by maximization of mutual information ［J］.IEEE Transactions on Medical Imaging,1997，16 (2):187-198.
［7］Josien P，Max A，Maintz JB, et al.Image registration by maximization of combined mutual information and gradient information ［J］.IEEE Transactions on Medical Imaging，2000,19 (8):809-815.
［8］Shams R,Sadeghip, Kennedy R,et al.Gradient intensity:A new mutual information-based registration method ［C］.IEEE Conference on Computer Vision and Pattern Recognition，Minneapolis，MN,2007:1-8.
［9］Mallat S.Singularity detection and processing with wavelets［J］.IEEE Transactions on Information Theory，1992，38(2):617-643.
［10］Liao JM,Li GL.Multimodality medical images registration based on expanded powell algorithm［J］.Journal of Chinese Modem Imaging，2007,4 (1):41-45.

服務(wù)與反饋：

【文章下載】【加入收藏】

提示：您還未登錄，請(qǐng)登錄！點(diǎn)此登錄

51黑料吃瓜在线观看,51黑料官网|51黑料捷克街头搭讪_51黑料入口最新视频