您的位置:首页  > 论文页面

基于Gabor分解和二维熵的视觉注意

发表时间:2014-02-28  浏览量:1855  下载量:508
全部作者: 吕琦,王斌,张立明
作者单位: 复旦大学信息科学与工程学院;复旦大学电磁波信息科学教育部重点实验室
摘 要: 基于现有模型的优缺点,从频率域的角度提出一种基于显著图的计算模型。模型主要结合了感受野的调节机制,从空间域的角度实现显著图的计算。通过2个主要步骤可以得到显著图:首先对输入图像在不同尺度和不同方向上进行Gabor小波分解,以得到特征图;接下来以二维熵作为度量融合与挑选这些特征图,并得到最终的显著图结果。实验证明:所提出的算法在视觉注视点的预测上要优于大多数现有算法,不论是对心理图像还是包含不同大小显著物体的自然图像。除此之外,Gabor的生物特性使得所提方法更加可靠,并且能更好地适应不同情况。
关 键 词: 计算神经网络;视觉注意力;扩展的经典感受野;Gabor分解;二维熵
Title: Visual attention based on Gabor decomposition and 2D entropy
Author: LÜ Qi, WANG Bin, ZHANG Liming
Organization: School of Information Science and Technology, Fudan University; Key Laboratory of Information Science of Electromagnetic Waves, Ministry of Education, Fudan University
Abstract: Based on both capabilities and defects of existing models, the paper proposes a computational saliency-oriented model from the perspective of frequency domain. This model employs the mechanism of receptive field and the saliency is computed in spatial domain. A saliency map can be generated by two main steps: firstly, Gabor wavelet decomposition of the input image at different levels and orientations is used to produce the feature components, and then these components are selected and fused in the sense of 2D entropy to generate the final saliency map. It is proved that the proposed algorithm outperforms most of state-of-the-art algorithms at human fixation prediction for both psychological patterns and natural images including salient objects with arbitrary sizes. Besides, biological plausibility of Gabor filter makes our approach more reliable and adaptive to various stimuli.
Key words: computation neural net; visual attention; extended classical receptive field; Gabor decomposition; 2D entropy
发表期数: 2014年2月第4期
引用格式: 吕琦,王斌,张立明. 基于Gabor分解和二维熵的视觉注意[J]. 中国科技论文在线精品论文,2014,7(4):306-313.
 
0 评论数 0
暂无评论
友情链接