Anomaly detection refers to the process used to detect images/data instances which are significantly different from most of the samples. Applications of anomaly detection include fraud detection in financial transactions, fault detection in manufacturing, intrusion detection in a computer network, monitoring sensor readings in an aircraft, spotting potential risk or medical problems in health data, and predictive maintenance. This thesis aims at exploring the state of the art and experimenting with different hypotheses to increase the speed and/or accuracy given the constraints of real world environment and datasets and gain more insight into requirements and limitations while applying it in manufacturing industries.