基于图像分类的图像美学评价研究  

Research on Image Aesthetic Evaluation Based on Image Classification

作　　者：高寒[1] 唐降龙[1] 刘家锋[1] 金野[1] 

机构地区：[1]哈尔滨工业大学计算机科学与技术学院,哈尔滨150001

出　　处：《智能计算机与应用》2013年第4期39-41,46,共4页Intelligent Computer and Applications

摘　　要：提出图像类别及图像内容在审美中的重要性,并由实验验证审美特征对于不同的图像类别体现出不同的适应性,因此确定图像类别对图像审美的影响。通过分析现有的图像审美特征的着重点,建立专用于图像审美的图像预分类方法,并确定对各个图像分类有效的审美特征,通过已标签图像的数据训练,计算各审美特征的权值,最终建立起内部先分类后审美的图像审美系统。This article suggests the importance of image classification and image content in aesthetics and the experimental verification that aesthetic features reflect different adaptability on different image categories, therefore, determines the impact that image class makes on the image aesthetic. Through the analysis of the existing focus on image aesthetic characteristics, specialized image classification method of image aesthetics can be established. After that, through training labelled image data and calculating the value of the aesthetic characteristics ,finally image aesthetic structure that internal classification is ahead of image aesthetics can be established.

关 键 词：图像审美 审美特征 敏感特征集合 图像内容 分类 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]