基于局部语义上下文的场景分类方法

Scene classification based on local semantic context

针对现有方法难以解决复杂场景图像分类的问题,本文提出一种基于局部语义上下文的场景分类方法。该方法将整个图像分割为一系列超像素,从超像素提取局部特征表示图像的局部观察;在观察图像和场景类别标签之间引入表示超像素区域语义的随机变量,通过不同随机变量之间的依赖关系引入局部语义上下文信息,较好地描述了图像观察、图像内容与场景类别标签之间的语义关联度,最后定义判别图像场景类别的目标函数,采用优化方法推断图像的场景类别。在标准图像库进行的实验证明了该方法的有效性。

An approach to scene classification based on local semantic context is proposed to solve the problemof complex sceneclassification. The set of super-pixel of image is segmented, and the local feature extracted fromthe super-pixel is used as the localimage observation. The random variable is introduced to represent local semantic objects, and the spatial context of the image isdescribed by the relationship among the variables and their neighbourhood. A new mathematical formula is used to describe therelationship among image observations, image content and scene categorization, and the semanticmodeling and scene classificationof the whole image is obtained. The test on the standard image dataset has demonstrated the effectiveness of the proposed approachcomparable with other state-of-the-art approaches.