Abnormal Behaviour Recognition
Mahfuzul Haque

www.monash.edu.au
Literature
• Supervised Approach
• Unsupervised Approach
• Semi-supervised Approach

www.monash.edu.au
2
Supervised Approach
• Based on the assumption that there exist
well-defined and known a priori
behaviour classes (both normal and
abnormal)
• However, in reality, abnormal behaviour
is both rare and far from being well
defined, resulting in insufficient clearly
labelled data required for supervised
model building.
www.monash.edu.au
3
Unsupervised Approach

• Further categorized into two different
types:
– With explicit behaviour model
– With no behaviour explicit model

www.monash.edu.au
4
Unsupervised – No Explicit Model
• Either clustering on observed patterns and
labelling those forming small clusters as being
abnormal
• Or building a database of spatiotemporal
patches using only regular / normal behaviour
and detecting those patterns that cannot be
composed from the database as being
abnormal.
• Applicability on previously unseen (normal)
behaviour?

www.monash.edu.au
5
Semi-supervised Approach
• Two-stage training process
• Stage one: normal behaviour model is
learned using labelled normal patterns.
• Stage two: an abnormal behaviour model
is then learned unsupervised using
Bayesian adaptation.
• Still suffers from the laborious and
inconsistent manual data labelling
process
www.monash.edu.au
6
Abnormal Behaviour Detection in Literature
Abnormal Behaviour Detection

Supervised

Unsupervised

Manual Labelling,
Prior assumption of well
define behaviour classes

No Behaviour Model
Clustering of observed patterns,
Database of spatiotemporal patches

Semi-supervised
Normal behaviour model using
manual labelling,
Abnormal behaviour model
unsupervised using Bayesian
adaptation

Explicit Behaviour Model
More Recent Approach
Mixture of Dynamic Bayesian Networks (DBNs)

www.monash.edu.au
7

Kb behaviour-recognition

  • 1.
  • 2.
    Literature • Supervised Approach •Unsupervised Approach • Semi-supervised Approach www.monash.edu.au 2
  • 3.
    Supervised Approach • Basedon the assumption that there exist well-defined and known a priori behaviour classes (both normal and abnormal) • However, in reality, abnormal behaviour is both rare and far from being well defined, resulting in insufficient clearly labelled data required for supervised model building. www.monash.edu.au 3
  • 4.
    Unsupervised Approach • Furthercategorized into two different types: – With explicit behaviour model – With no behaviour explicit model www.monash.edu.au 4
  • 5.
    Unsupervised – NoExplicit Model • Either clustering on observed patterns and labelling those forming small clusters as being abnormal • Or building a database of spatiotemporal patches using only regular / normal behaviour and detecting those patterns that cannot be composed from the database as being abnormal. • Applicability on previously unseen (normal) behaviour? www.monash.edu.au 5
  • 6.
    Semi-supervised Approach • Two-stagetraining process • Stage one: normal behaviour model is learned using labelled normal patterns. • Stage two: an abnormal behaviour model is then learned unsupervised using Bayesian adaptation. • Still suffers from the laborious and inconsistent manual data labelling process www.monash.edu.au 6
  • 7.
    Abnormal Behaviour Detectionin Literature Abnormal Behaviour Detection Supervised Unsupervised Manual Labelling, Prior assumption of well define behaviour classes No Behaviour Model Clustering of observed patterns, Database of spatiotemporal patches Semi-supervised Normal behaviour model using manual labelling, Abnormal behaviour model unsupervised using Bayesian adaptation Explicit Behaviour Model More Recent Approach Mixture of Dynamic Bayesian Networks (DBNs) www.monash.edu.au 7