关键词: 贝叶斯分类, 灰度直方图, 关联模型, 纹理特征, 加权FCM算法, 图像自动标注

Abstract:

In order to cross the semantic gap between high-level semantic concepts and low-level visual features,  a new method about automatic image annotation was proposed. First,  the method of gray histogram was applied to segment images and texture features were  extracted from image regions. Second,  the greater weights could be  distributed to relevant features compared with less relevant features for FCM algorithm,  and better results of clustering could be achieved. Finally,  the correlation model between keywords and clustering regions was  established in accordance with the labeled images in the training sets by the approved Bayesian classification. The similarity between the testing images and the training images were calculated,  and the maximal conditional probability to annotate the new image regions was achieved. The experiments on a standard Corel dataset compared with other methods showed that the approved labeling approach performed more accurately and effectively than traditional labeling methods.

Key words: gray histogram, the correlation model, texture features, automatic image annotation, weighted FCM algorithm, Bayesian classification