Novel Real-Time Automobile Detection Algorithm for Blind Spot Area

Abstract

Many facets of automobile technology are moving to information technology (IT)-based convergence systems. Sensors and equipment used to prevent accidents that occur due to drive negligence are under development. An automobile detection algorithm for blind spot areas while driving is proposed. Detection of candidate vehicles is based on vehicle features, and then a vehicle is subject to verification of the detected candidate vehicle area. Using a Haar-like feature for vehicle detection, an adaptive vehicle verification process based on brightness values of vehicle areas is used. The hierarchical scheme of the proposed algorithm is efficient and accurate, based on detection and verification results in the blind spot area.