网络爬虫的图像美学质量数据库及其美学评价系统设计

The image aesthetic database collection using web crawler and aesthetic evaluation method

本文提出了一种基于网络爬虫建立图像美学数据库并评价图像美学的方法.首先使用网络爬虫获取数据库图像,然后提取图像的美学特征,进行美学评价.本方法提取的美学特征涉及了全局和区域特征两方面,因此在特征提取之前需要使用功率谱斜度的方法分离图像的主题、背景区域.实验结果表明,本方法可以较好地抓取并且分析网络图像,并且评价方法与人类的主观感知有很好的一致性,能准确地反映人类的美学鉴赏系统.

This paper proposes an image aesthetic database collection using web crawler and aesthetic evaluation method.The web crawler downloads the images to establish the database from the professional photo website,and some image aesthetic features are extracted to evaluate the performance of the database,which include two main parts,the global features and the region features.During extracting the features,the slope of the local magnitude spectrum is used to separate subject area and the background of the image.Experimental results show that our web crawler is a better way to capture and analyze internet images,and the proposed image aesthetic features are efficiently evaluate the image aesthetic,which is the same as human subjective perception,and show the aesthetic appreciation ability accurately.