关键词: 机器视觉, 显著特征提取, 稀疏表达, 频段分割, 特征组合

Abstract: According to the human visual attention mechanism, this article introduces the salient features of image extracted in spatial domain using local clutter density contrast and global color distribution estimation, and extracted by the Efficient Band Divided Method in frequency domain, then refers to the principle that activity in cells responds to stimuli,such features combined by the theory of feature combination. Local clutter density contrast imitates the law of universal gravitation, computing the visual attention on Sparse matrix model.Proposing the Efficient Band Divided Method unites information of amplitude for salient feature detection in frequency domain,later utilizes all features with weight. It is proved that the algorithm can effectively extractive regions through some experiments.Visual attention model applied in virtual and reality interaction is proposed to detect the effective region using the principle of saliency from real scene and compare with other algorithms.

Key words: Machine vision, Saliency detection, Sparse representation, Efficient Band Divided Method, Feature combination