Applications of images anomalies detection using deep learning in department store

Deep learning is a branch of machine learning which itself is a branch of Artificial Intelligence. The use of deep learning to solve domain specific problems is on the rise. Deep learning has been successfully used to assist sales prediction in retail, disease detection in medicine, road infrastructure monitoring by checking cracks on the road, accidents prone zones, detect anomalous activities in the realm of cyber security etc. At present, user and machine generated data is available abundantly and the challenges for enterprises is to infer new information from the available data to increase profit for the enterprise, produce a reliable system and increase customer satisfaction. Deep learning has been successfully used in classification of data with high precision. However, there are bottlenecks when it comes to anomalies in data because building models to detect anomalies is more difficult than classification problems. This thesis aims to study image anomalies detection and their applications department store using design science research methods. This thesis presents a basic prototype application to demonstrate anomalies in product areas in department stores.