How do targets, nontargets, and scene context influence real-world object detection?

How do targets, nontargets, and scene context influence real-world object detection?