Performance analysis on visual attention using spiking and oscillatory neural model. (1st January 2013)

• Record Type: Journal Article
• Authors: Andrushia, A. Diana; Thangarajan, R.; Sebastian, Greeshma
• Is Part Of: International journal of computational vision and robotics. Volume 3:Number 4(2013)
• Page Start: 293
• Page End: 307
• Publication Date: 2013-01-01
• Subjects: attention -- spiking neurons -- nucleus basalis -- bottom-up and top-down method; Computer vision -- Periodicals; Robotics -- Periodicals; Artificial intelligence -- Periodicals; 006.3705
• Languages: English
• ISSNs: 1752-9131

Abstract

Visual attention is a process of sustained concentration on a specific stimulus. This concentration can be increased by activating the nucleus basalis in the basal forebrain using the spiking neuron model. Input stimulus is converted into spikes. Neurons are transmitting information in the form of pulse. By using this information spiking neuron model for the basal forebrain is simulated. Bottom-up and top-down method is used in lateral geniculate nucleus (LGN). The feedback connections are applied in the visual cortex for the enhancement of visual attention. To analyse the performance of spiking and oscillatory model segmentation and separation accuracy are obtained which shows the oscillatory model produce better accuracy for visual stimuli.