摘要
视觉词典容量是影响图像场景分类精度的重要因素之一,大容量的视觉词典因计算量较大影响了分类的效率,而小容量的视觉词典由于多义词问题的严重致使场景分类精度降低.针对该问题,提出一种基于概念格层次分析的视觉词典生成方法.首先生成关于训练图像视觉词包模型的初始视觉词典;然后在构造的概念格上利用概念格的概念层次性,通过动态地调整外延数阈值,获取粒度大小不同容量的描述图像各场景语义的约简视觉词典;最后对各类约简视觉单词构成向量进行异或,删除多义词,进而生成有效描述图像场景语义的视觉词典.实验结果表明,文中方法是有效的.
The visual dictionary size is an important factor that affects the performance of scene classification.The large capacity of visual dictionary can influence the classification efficiency due to the lager calculation,while the small capacity of visual dictionary can reduce the classification accuracy because of theinfluences of polysemy. To solve the problem, a generating method of visual dictionary based on the conceptlattice hierarchy is proposed in this paper. First, the initial visual dictionary of training images onbag-of-visterms model is generated. Then, with the use of concept lattice’s hierarchy analysis, the differentgranularities of reduced visual dictionaries are extracted from the concept lattice by setting different extensionthresholds. Finally, the polysemy is deleted by making XOR operations on all types of the reduced visualdictionaries, and a visual dictionary for better describing the image content is generated. Experimentalresults show that this method is effective.
出处
《计算机辅助设计与图形学学报》
EI
CSCD
北大核心
2015年第1期136-141,共6页
Journal of Computer-Aided Design & Computer Graphics
基金
国家自然科学基金(61373099).
关键词
视觉词包
视觉词典
多义词
概念格
层次分析
bag-of-visterms
visual dictionary
polysemy
concept lattice
hierarchy analysis