Path Hole Detection to Assist the Visually Impaired People in Navigation

The number of visually impaired people is increasing with the rapid growth of population. The visually impaired people face much difficulties in their daily living owing to losing their vision. Path hole is a major hindrance to their walking. So path hole detection has become a prominent issue to aid the visually impaired people. We proposed a solution by detecting path hole on the road surfaces using Convolution Neural Network. The proposed system is able to classify the road surfaces with path hole and non-path hole. We have used two benchmarked dataset named as KITTI ROAD and Pothole detection. We have trained Convolution Neural Network with training-testing partition. The performance of the system is measured in terms of accuracy, precision, recall and error rate. The overall accuracy and error obtained by the system are 97.12% and 0.065 respectively in testing phase for 20 iterations. Additionally, precision and recall obtained by the system are 96.68% and 95.77% in testing phase.