基于视觉注意机制的感兴趣区提取方法

以精确获取图像中对象感兴趣区域为目标,提出一种基于视觉注意机制和K均值聚类相结合的感兴趣区提取方法。图像经过视觉特征提取、高斯金字塔多尺度变换后,依据多特征图合并策略生成显著图。采用K均值聚类方法分割图像的候选区域,并结合显著图提取图像感兴趣区。实验结果表明,运用该方法提取的感兴趣区更接近人类的视觉注意过程,并具有一定的抗噪能力。

Regions of Interest Extraction Based on Visual Attention Mechanism

In view of exactly acquiring objects of natural images,the way of extracting regions of interest based on vision attention mechanism and k-means clustering was presented.After multi-scale Gaussian pyramids transform,multi-feature maps were combined into a saliency map.The natural image was segmented image regions with the k-means clustering algorithm.Combining with the saliency map,regions of interest was extracted.The experimental results show that the proposed method is closer to the process of human visual attention and demonstrate its effectiveness and robustness.