Current traffic light triggering is based on inductive loop sensors. Unfortunately, motorbikes (scooters, motorcycles, etc) have a difficult time triggering these sensors. In this paper, we propose an image processing algorithm to detect motorbikes at a traffic stop using a fixed camera. The algorithm tracks the trajectory of the objects in the footage by motion segmentation and connected component labeling. Classification can be created to categorize these objects as incoming traffic based on the objects trajectory. To handle different lighting conditions in the motion segmentation, we take a dual approach by selecting RGB or Opponent colorspace. RANSAC is utilized to help trajectory creation. Experimental tests using real video footage exhibit robust results under varying conditions.
Speed is a concern on video processing. Would it be possible to detect on still imagery every few frames? Perhaps using a haar features with a cascade of adaboost classifiers? They work well enough with faces. Could be good to take at look at this approach.