改进C-V分割算法在多光谱成像仪中的应用  

Application of improved C-V segmentation in multi-spectral imager

作　　者：张艳超[1,2] 赵建[1] 

机构地区：[1]中国科学院长春光学精密机械与物理研究所,吉林长春130033 [2]中国科学院大学,北京100049

出　　处：《中国光学》2015年第1期68-73,共6页Chinese Optics

基　　金：吉林省重点科技攻关项目(No.20140204058GX);长春市重大科技攻关项目(No.14KG011)

摘　　要：为了进一步提高C-V模型的分割速度、降低初始轮廓曲线位置对分割结果的耦合性,提高多光谱成像仪图像分割效率,本文提出一种改进的C-V模型。该模型通过将每次迭代得到的距离函数的最大值引入C-V模型的Dirac函数,对该函数进行自适应参数修正,以拓宽活动轮廓线的有效作用范围,进而大大降低分割算法的迭代次数。实验结果表明,与经典的C-V模型相比,改进的C-V模型在其终止条件下得到了较理想的分割效果,降低了初始曲线位置对最终分割结果的影响,且新模型的收敛速度在原有的基础上至少提高了7倍。改进的C-V模型在实时性及全局性方面都得到了明显改进,进一步提高了该算法在多光谱成像仪的图像分割方面的鲁棒性。In order to improve the image segmentation speed when using C-V model, reduce the segmentation coupling with initial contour position , and improve the image segmentation efficiency of multi-spectral imager , an improved C-V model is proposed in the paper .In this model , the Dirac function′parameter is corrected adaptively by introducing the maximum value of distance function in each iteration .In this way , the effective range of active contour is broadened , and the number of iterations is reduced .The experimental results show that the ideal segmentation effect is obtained by the improved C-V model with the iteration termination condition.Compared with the classic C-V model, the influence of initial contour position on segmentation is reduced.In addition, the convergence speed is improved by 7 times.The characteristics of real time and global nature both become better .Therefore , the robustness of multi-spectral imager segmentation is improved accordingly .

关 键 词：多光谱成像仪 图像分割 偏微分方程 几何活动轮廓模型 C-V模型 

分 类 号：TP394.1[自动化与计算机技术—计算机应用技术]