====== Differences ====== This shows you the differences between two versions of the page.
Both sides previous revision Previous revision Next revision | Previous revision Next revision Both sides next revision | ||
projects [2013/10/15 03:04] pszwed [Fuzzy Semantic Petri Nets] |
projects [2013/10/20 15:15] pszwed [Fuzzy Semantic Petri Nets] |
||
---|---|---|---|
Line 5: | Line 5: | ||
Fuzzy Semantic Petri Nets (FSPN) are Petri nets coupled with an underlying fuzzy ontology. | Fuzzy Semantic Petri Nets (FSPN) are Petri nets coupled with an underlying fuzzy ontology. | ||
- | *The fuzzy ontology stores facts about a development of certain complex event (or process). They are expressed as fuzzy unary or binary relations between participating objects. The ontology ABox is updated with new observations. | + | *The fuzzy ontology stores facts about a development of a certain complex event (or a process). They are expressed as fuzzy unary or binary relations between participating objects. The ontology ABox is updated with new observations. |
- | *The Petri net defines a scenario of an event, its places correspond to the scenario steps | + | *The Petri net defines a scenario of an event (or a process, if you like), its places correspond to the scenario steps |
*Tokens in FSPN represent combinations of objects participating in the event. They are attributed with weights expressing likelihood of a scenario step occurrence | *Tokens in FSPN represent combinations of objects participating in the event. They are attributed with weights expressing likelihood of a scenario step occurrence | ||
*Guards in FSPN are conjunctions of fuzzy predicates referencing the coupled ontology | *Guards in FSPN are conjunctions of fuzzy predicates referencing the coupled ontology | ||
- | FSPN were applied to the problem of video event recognition. In the presented example (the event scenario and dataset was published at [[http://www.cvg.rdg.ac.uk/PETS2006/|PETS 2006]]) an abandonned laguagge is correctly detected. | + | FSPN were applied to the problem of video event recognition. In the presented example an abandonned laguagge is correctly detected. (The event scenario and the dataset were published at [[http://www.cvg.rdg.ac.uk/PETS2006/|PETS 2006 Workshop]]) . |
- | The clip shows the GUI of a software, that internally usess the FSPN approach to recognize video events. | + | Scenario: |
+ | -A man enters the scene and is still for some time | ||
+ | -Then he takes his rucksack off and puts it on the floor | ||
+ | -He stays close to the luggage for some time. | ||
+ | -He leaves the rucksack unattended (the object is classified //unattended// if the distance between it and its owner is greater than 3 m) | ||
+ | -Finally, the man disappears. | ||
+ | |||
+ | The clip below shows the GUI of a software, that internally usess the FSPN approach to recognize video events. | ||
<html> | <html> | ||
<iframe width="560" height="315" src="//www.youtube.com/embed/_hArl9i20TY" frameborder="0" allowfullscreen></iframe> | <iframe width="560" height="315" src="//www.youtube.com/embed/_hArl9i20TY" frameborder="0" allowfullscreen></iframe> | ||
</html> | </html> | ||
- |