图像清晰度评价函数的研究  

Study on image definition evaluation function

作　　者：严锦雯 贾星伟 隋国荣[1] 涂建坤[2] YAN Jinwen;JIA Xingwei;SUI Guorong;TU Jiankun(School of Optical-Electrical and Computer Engineering,University of Shanghai for Science and Technology,Shanghai 200093,China;Shanghai Electric Cable Research Institute,Shanghai 200093,China)

机构地区：[1]上海理工大学光电信息与计算机工程学院,上海200093 [2]上海电缆研究所,上海200093

出　　处：《光学仪器》2019年第4期54-58,共5页Optical Instruments

摘　　要：为了实现更为高效、准确的光纤端面自动对焦,对现有图像的清晰度评价函数进行了研究,并在此基础上,提出了一种新的计算简便、抗噪性好的清晰度评价函数。采用光纤几何参数测试仪获得光纤端面清晰成像面前后等间隔的24帧光纤端面图像,利用不同的图像清晰度评价函数对这24帧图像进行计算,得到相应的调焦曲线并分析其单峰性、单调性和抗噪性。结果显示,所提的新评价函数具有更好的表现,对自动对焦技术具有一定的参考价值。In order to realize more efficient and accurate automatic focusing of optical fiber end face,the definition evaluation function of existing images is studied.On the basis of this,a new definition evaluation function with simple calculation and good anti-noise is proposed.An optical fiber geometry measurement system is used to obtain 24 frames of equally spaced optical fiber endface image.Different image definition evaluation functions are used to calculate the 24 frames of images.The corresponding focusing curves are obtained and their monomorphism,monotonicity and anti-noise are analyzed.The results show that the proposed evaluation function has better performance and has certain reference value for automatic focusing technology.

关 键 词：清晰度评价 抗噪性 单峰性 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]