Title :
Biologically Motivated Salient Regions Detection Approach
Author :
Zhang, Qiaorong ; Xiao, Huimin
Author_Institution :
Coll. of Inf., Henan Univ. of Finance & Econ., Zhengzhou
Abstract :
A biologically motivated salient regions detection approach is proposed to identify the salient regions within complex natural images. Multi-scale image features such as intensity, color and orientation are extracted to get some feature maps. The log spectra of the feature maps are analyzed in frequency spectrum domain and the spectra residual are extracted. Then the corresponding feature saliency maps are constructed in spatial domain and theses feature saliency maps are combined to an integrated saliency map. The salient regions are detected in terms of the integrated saliency map. Computer experimental results show that the proposed method successfully detects salient regions in many natural images and artificial images.
Keywords :
biology computing; feature extraction; image colour analysis; object detection; biologically motivated salient regions detection; complex natural images; feature maps extraction; frequency spectrum domain; integrated saliency map; multiscale image features; spectra residual; Biology computing; Computational complexity; Data mining; Educational institutions; Frequency domain analysis; Humans; Image analysis; Image processing; Information technology; Visual system; image processing; saliency map; salient regions detection; visual attention;
Conference_Titel :
Intelligent Information Technology Application, 2008. IITA '08. Second International Symposium on
Conference_Location :
Shanghai
Print_ISBN :
978-0-7695-3497-8
DOI :
10.1109/IITA.2008.372