Obstacle Detection in Indoor Environment for Visually Impaired Using Mobile Camera

Abstract

Abstract. Obstacle detection can improve the mobility as well as the safety of visually impaired people. In this paper, we present a system using mobile camera for visually impaired people. The proposed algorithm works in indoor environment and it uses a very simp le technique of using few p re-stored floor images. In indoor environ ment all unique floor types are considered and a single image is stored for each unique floor type. These floor images are considered as reference images. The algorithm acquires an input image frame and then a region of interest is selected and is scanned for obstacle using pre-stored floor images. The algorithm compares the present frame and the next frame and co mpute mean square error of the two frames. If mean square error is less than a threshold value α then it means that there is no obstacle in the next frame. If mean square error is greater than α then there are t wo possibilit ies; either there is an obstacle or the floor type is changed. In order to check if the floor is changed, the algorithm co mputes mean square error of next frame and all stored floor types. If minimu m of mean square error is less than a threshold value α then flour is changed otherwise there exist an obstacle. The proposed algorithm works in real-time and 96% accuracy has been achieved.

Smartphone-Based Cognitive Assistance of Blind People in Room Recognition and Awareness

“…Rahman et al [32] use the already stored exclusive ground images for reference in the indoor environment. e image taken is compared with prestored pictures via an algorithm, which calculates the Mean Square Error (MSE) and then compares it with a threshold value α.…”