PROJECT TITLE :

Scale-Invariant Amplitude Spectrum Modulation for Visual Saliency Detection

ABSTRACT:

Saliency detection is one among the key issues in simulating visual attention choice. Most attention models adopt the competitive structure to simulate the human visual system. Though these models offer exceptional results and convincing biological plausibility, they're still confronted with many difficulties in practical applications because of their extreme time cost and parameter sensitivity. Recently, a replacement saliency detection approach based on Fourier rework, as represented by spectral residual (SR) and part Fourier transform (PFT), has been attracting much attention for its excellent accuracy and computational speed. All these models will be unified into one framework referred to as amplitude spectrum modulation (ASM). The aim of this paper is to explore the intrinsic mechanism of ASM and develop an advanced ASM model. When analyzing SR and PFT, we offer a mathematical description for the fundamental plan and also the inherent limitations of the prevailing ASM models. A brand new saliency detective model, based mostly on the scale-invariant ASM, scene and context-primarily based modulation, and competitive structure, is also proposed breaking through the limitations of the ancient ASM models. Simulation results suggest that the proposed model is additional accurate in predicting human eye fixation and is a lot of strong against completely different types of stimulus compared with competing models.