Subclass Evidence

This relationship gives evidence for the presence of an object of class $M$, given the presence of an object of subclass S. As above, the notion of subclass is that of a specialization, and an object may have several. Here, the implication is a necessary one, because an instance of a given subclass is necessarily an instance of the class. Hence, the plausibility of the object must not be less than that of its subclasses.

The constraints that specify the subclass relationship calculation are:

• The more plausible the subclass, the more plausible the object.
• The plausibility of the object is at least that of each subclass.
• The context of the subclasses is the context of the object.

These constraints lead to the following formal definition of the subclass evidence computation:

Given:
	a model instance of class $M$ in image context $C$
	a set {$S_i$} of subclasses of $M$
	a set {$p_i$} of plausibilities of model $S_i$ in context $C$
Then, the subclass relationship evidence value is:
	$evd_{subcl} = max_i(p_i)$

If no subclass evidence is available, this computation is not applied. The invocation network fragment for this evidence type is similar to those shown previously.