An obstacle segmentation and classification system for the visually impaired

Abstract

In the majority of cases blindness is caused by retinal degenerative conditions such as Age-Related Macular Degeneration (AMD) and Retinitis Pigmentosa (RP). However, blind individuals still retain central visual pathways and processing mechanisms. Recent advances in Time of Flight (TOF) imaging technology have presented new opportunities to develop improved sensory substitution systems for compensation of visual sensory loss. In sensory substitution information from an artificial receptor/sensor is coupled to the brain via a Human Machine Interface (HMI). The brain is able to utilise this information in place of what is usually transmitted from an intact sense organ such as the eye. The human brain has an adaptive capacity called plasticity. Research has shown that inputs from sensory substitution devices can reach numerous brain structures including those anatomically and physiologically related to the lost sensory modality. This is provided that the information from the HMI/artificial receptors is appropriate to restore function. The authors segmentation and classification system (VisionRE) aims at improving the shortcomings of the current state of the art by accurately segmenting and classifying 3D objects/obstacles in real-time and presenting the 3D information to the user in a more intuitive manner via a multimodal electro-tactile and audio HMI.