基于非下采样Contourlet变换的显微图像清晰度评价算法  

Micro-image definition evaluation algorithm based on non-subsampled contourlet transform

作　　者：周厚奎[1,2] 葛品森[1,2] 冯海林[1,2] 

机构地区：[1]浙江农林大学信息工程学院,浙江杭州311300 [2]浙江农林大学浙江省林业智能监测与信息技术研究重点实验室,浙江杭州311300

出　　处：《合肥工业大学学报（自然科学版）》2014年第10期1204-1209,共6页Journal of Hefei University of Technology：Natural Science

基　　金：国家自然科学基金资助项目(61272313);浙江省科技厅公益性资助项目(2012C21015);浙江省教育厅资助项目(Y201224063);国家级大学生创新创业训练计划资助项目(201210341019);浙江农林大学创新训练计划资助项目(201210017;201210018)

摘　　要：针对显微图像噪声多、光照不均匀的特点,文章提出了一种基于NSCT的显微图像清晰度评价算法。算法首先通过NSCT对图像进行分解,然后基于改进的拉普拉斯能量获取子带系数的能量,最终由低、高频子带系数能量的比值得到评价结果。实验结果表明,与拉普拉斯、方差及小波变换评价算法相比,该算法在保持高灵敏度的同时具有较好的抗噪性能,评价结果更为准确、稳定、可靠。According to the characteristics of image noise and uneven illumination in micro-image ,a new algorithm based on the non-subsampled contourlet transform（NSCT） is proposed for micro-image definition evaluation .Firstly ,the images are decomposed by NSCT .Then the energy of low-frequen-cy coefficients and high-frequency coefficients is obtained by sum-modified Laplacian method .Finally , the evaluation function is defined by the ratio of the energy of low-frequency coefficients and high-fre-quency coefficients .Ten sets of experimental results show that the new algorithm has better perform-ance of non-deflection , single peak value , sensitivity and anti-noise than variance , Laplacian and wavelet methods .The evaluation results obtained by the new algorithm are more accurate ,stable and reliable .

关 键 词：清晰度评价 非下采样Contourlet变换(NSCT) 显微图像 自动对焦 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]