基于AKAZE特征匹配的花山岩画图像识别研究  

Research on Zuojiang Huashan Rock Art Image Recognition Based on AKAZE Feature Matching

作　　者：关雪敏 何皓怡[1] GUAN Xuemin;HE Haoyi(Guangxi Minzu Normal University,Chongzuo Guangxi 532200,China)

机构地区：[1]广西民族师范学院,广西崇左532200

出　　处：《信息与电脑》2023年第19期146-148,共3页Information & Computer

基　　金：崇左市科技计划项目(项目编号:崇科20220641);广西民族师范学院校级科研项目(项目编号:2021BS006)。

摘　　要：近年来,对花山岩画进行的研究不断增加,特别是对于该岩画图像的识别和分析方面引起了广泛关注。文章提出了一种基于加速的核化自适应区域提取(Accelerated-Kernelized-Adaptive-Zone-Extraction,AKAZE)特征匹配的识别方法,来识别和分析花山岩画的图像。通过在识别过程中利用AKAZE特征匹配算法,文章能够从花山岩画图像中提取出独有的特征,并将其与已知岩画特征进行匹配,从而准确识别花山岩画。实验证明,该方法在花山岩画图像的识别和分析上取得了较好的效果,为花山岩画的保护和研究提供了一种有效的手段。In recent years,research on Huashan rock paintings has been increasing,especially in the recognition and analysis of rock painting images,which has attracted widespread attention.This paper proposes a recognition method based on Accelerated-Kernelized-Adaptive-Zone-Extraction(AKAZE)feature matching,which can effectively recognize and analyze images of Huashan rock paintings.By utilizing the AKAZE feature matching algorithm during the recognition process,this article is able to extract unique features from Huashan rock art images and match them with known rock art features,thereby achieving accurate recognition of Huashan rock art.Experiments have shown that this method has achieved good results in the recognition and analysis of Huashan rock art images,providing an effective means for the protection and research of Huashan rock art.

关 键 词：花山岩画 加速的核化自适应区域提取(AKAZE) 特征匹配 图像识别 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]