光学仪器, 2016, 38 (5): 407, 网络出版: 2016-12-23  

基于局部窗口与极值的显微图像细节增强

Microscopic image detail enhancement based on local window and extremum
作者单位
1 宁波永新光学股份有限公司, 浙江 宁波 315040
2 浙江大学 现代光学仪器国家重点实验室, 浙江 杭州 310027
3 杭州电子科技大学 电子信息学院, 浙江 杭州 310018
摘要
光学显微成像中,光学物镜、电子成像等对显微图像质量的影响较大,容易形成退化,导致获取的显微图像细节不够清晰。结合数码显微成像的具体需求,提出一种后处理细节增强处理方法。分析了数码显微成像系统的退化过程,强调了光学退化等带来的模糊细节的效应。一方面,利用尺寸变化的双窗口,可以框定凸显不同尺寸的细节信息,实现局部信息的分析;另一方面,利用局部窗口内极大值与极小值分析,并与原图信息比对以获取细节图像。两者综合,最终实现细节的增强。测试实验表明,该方法能够很好地适用于数码显微成像系统,运行速度快,增强效果好。以视觉清晰度指标、灰度平均梯度与拉普拉斯算子和指标作衡量,增强后评价指标提升的平均百分比分别为20.9%、71.2%与81.8%。
Abstract
In the optical microscopic imaging,optical lens and electronic imaging have great influence on the microscopic image quality,related to the degradation of image detail.Combining with the demand of digital microscopic imaging,we proposed a post-processing method for detail enhancement.The degradation of digital microscopic imaging system was analyzed,especially the blurring effect caused by optical degradation.On one hand,we used variable sizes of local windows to extract details.On the other hand,the maximum and minimum within local window were analyzed,and the detailed image was obtained.Finally,the details were enhanced.The experimental results demonstrated that this method was suitable for digital microscopic imaging system,and the method ran fast and gave good enhancement results.By using visual definition,gray mean gradient and Laplacian sum as the metrics in the proposed method the results were improved by 20.9%,71.2% and 81.8%,respectively.

赵巨峰, 毛磊, 刘承, 冯华君, 高秀敏. 基于局部窗口与极值的显微图像细节增强[J]. 光学仪器, 2016, 38(5): 407. ZHAO Jufeng, MAO Lei, LIU Cheng, FENG Huajun, GAO Xiumin. Microscopic image detail enhancement based on local window and extremum[J]. Optical Instruments, 2016, 38(5): 407.

关于本站 Cookie 的使用提示

中国光学期刊网使用基于 cookie 的技术来更好地为您提供各项服务,点击此处了解我们的隐私策略。 如您需继续使用本网站,请您授权我们使用本地 cookie 来保存部分信息。
全站搜索
您最值得信赖的光电行业旗舰网络服务平台!