局部相位量化在模糊图像识别中的研究进展  

Research Progress of Local Phase Quantization in Blurred Image Recognition

作　　者：刘靖丹 逯洋[1] 王淳 LIU Jingdan;LU Yang;WANG Chun(College of Mathematics and Computer,Jilin Normal University,Siping 136000,China)

机构地区：[1]吉林师范大学数学与计算机学院,吉林四平136000

出　　处：《景德镇学院学报》2023年第3期6-11,共6页Journal of JingDeZhen University

基　　金：吉林省发展和改革委员会创新项目(2021C038-7);吉林省创新创业人才基金(2023QN31);吉林省自然科学基金(YDZJ202301ZYTS157)。

摘　　要：现代图像处理技术的不断发展,为人们提供了大量丰富的图像信息。图像处理手段很多,如图像去噪、图像平滑、图像去模糊、图像填充、图像分割等。处理模糊图像问题最具代表性的算法是局部相位量化(Local Phase Quantization,LPQ)和基于局部二进制模式(Local Binary Pattern,LBP)的算子。与LBP特征算子相比,LPQ特征算子对图像模糊不敏感,其特征提取更加稳定。从介绍LPQ算子开始,首先对LPQ算子的发展现状和应用领域进行分类和总结;其次详细阐述和评论每种实验方法;最后思考和讨论LPQ算子的发展方向。The continuous development of modern image processing technology has provided a large amount of rich image information for people.There are many image processing methods,such as image denoising,image smoothing,image blurring,image filling,image segmentation,and so on.The most representative algorithms for blurred image processing are Local Phase Quantization(LPQ),and an operator based on Local Binary Pattern(LBP).Compared with LBP-featured operator,LPQ-featured operator is less sensitive to image blur and its feature extraction is more stable.Starting with the introduction of LPQ operator,This paper firstly classifies and summarizes the development status and application fields of LPQ operator.Secondly,each experimental method is elaborated and reviewed in detail.Finally,the development direction of LPQ operator is considered and discussed.

关 键 词：局部相位量化 模糊图像识别 人脸识别 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]