Blind Spot Detection Using Inverse Perspective Mapping

The patent replaces the mechanical/optical sensor-based speed detection system with an image processing system that uses computer vision techniques. The system captures images of the blind spot area, detects objects in consecutive frames, and calculates their movement by comparing positions across frames. This substitution enables more accurate differentiation between stagnating vehicles and stationary objects through visual motion analysis rather than relying solely on speed threshold comparisons.

The patent applies preliminary action by analyzing object motion characteristics before triggering a warning signal. The system detects objects in the blind spot, tracks their movement across multiple image frames, and evaluates their velocity and acceleration patterns. Only objects exhibiting motion characteristics consistent with a stagnating vehicle (moving at similar speed to the host vehicle) trigger warnings. This preliminary motion analysis filters out stationary objects like guardrails before they can generate false alarms.

The patent applies universality by using a single camera system that performs multiple functions: it captures images of the blind spot area, detects objects within the blind spot, tracks object movement across frames, calculates object speed and direction, and determines whether objects are stagnating vehicles or stationary objects. This multi-functional approach replaces what would traditionally require multiple specialized sensors, thereby maintaining high detection accuracy while reducing system complexity.