Salient Region Detection by Tuning Spectrum Based on Lateral Inhibition

Detection of visually salient image regions is useful for applications like object segmentation, adaptive compression and object recognition. In this paper, we introduce a method for salient region detection using spectrum tuning based on lateral inhibition mechanism. The new method tunes the Fourier spectral adaptively using nonlinear function to inhibit the redundant image features and enhance the salient image features, and therefore the salient region could be detected effectively. We test this method on both real natural images and artificial pictures such as psychological patterns. Experimental results indicate that our method is more efficient than other methods in terms of detection rate, boundary integrity and contrast of salient region.