Analysis of brain signal on visual noise affecting driver’s attention using deep learning

One of the main reasons for road accidents is driver’s brain activities during driving, some of which is heavily influenced by roadside visuals like posters, billboards and other objects around. Our objective is to prove that those posters, billboards, and roadside activities are some of the reasons for driver’s distraction that create high arousal on negative valence in certain part of driver’s brain and causes frequent road accidents. Our experiment shows that images of visually distasteful creates quite opposite reaction to that images of visually pleasing to human brain. Taking pictures from different locations of Dhaka city representing the actual roadside scenario that produces high stimulation in brain and can lead to human errors. In this research, brain data was collected using EEG device by displaying categorically positive, negative, and neutral pictures to the viewers. Collected brain data was analyzed using Deep Learning techniques. From the result, this research established strong correlations between different categories of images and the neural activities in the brain. The results are conclusive about perturbation on visually distracting elements along the roadside which may contribute to minor to major road accidents.