Motion Deblurring for Pedestrian Crossing Detection in Advanced Driver Assistance System
A. Sumi, T. Santha · 2017
Nowadays Advanced Driver Assistance Systems (ADAS) provide a solution for safe driving and better interface between the human and machine. A Lot of research is undergone using computer vision to provide automation in this field. In this paper, detection of a pedestrian is designed with two main tasks one is preprocessing and the second one is detection. In preprocessing stage the captured video was affected by motion blur due to movement of both sensor and object. This was removed using the Wiener filter and evaluated by histogram equalization and classification. The proposed work improves the pedestrian detection rate by clearing the blur effect.