Saliency detection model based on wavelet transform and independent component analysis
Dongyue Chen, Danpei Zhao, Xiaosheng Yu, Zongwen Chen · 2012
Since people started researching the attention mechanism, numerous models have been put forward to simulate the attention selection. However, there were some shortcomings in all these proposed computational models, such as high complexity, low accuracy, heavy reliance on the choice of parameters and so on. A computational model based on wavelet transform and independent component analysis (ICA) is proposed in this paper to address these issues. In the proposed model, the visual saliency of a pixel is defined as the self-information of the local features. The model applies the multi-channel structure based on the YCbCrcolor space. The feature vector is obtained using the wavelet decomposition, and the joint probability density of feature vectors is evaluated with the ICA method. The experimental results show that our method outperforms most existing algorithms on the accuracy, complexity and efficiency.