What is Anomaly Detection and How it is Used in Computer Vision?
Hey there, in this article three things are covered, first what is anomaly detection from a vision perspective? what are the techniques used for anomaly detection? and where it is being used?
What is the Problem?
“An outlier (a.k.a anomaly) is an observation in a data set which appears to be inconsistent with the remainder of that set of data” — Johnson 1992
“An outlier is an observation which deviates so much from the other observations as to arouse suspicions that a different mechanism generated it” — Hawkins 1980
Anomalies are events that differ from the norm, occur infrequently, and don’t fit into the rest of the “pattern”. Anomalies include, for example:
- large stock market drops and gains as a result of global events
- faulty products in a factory or on a conveyor belt
- contaminated laboratory samples