VISION BASED SURVEILLANCE SYSTEM FOR THEFT DETECTION IN SUPERMARKETS
International Research Journal of Modernization in Engineering Technology and Science · 2023
Shoplifting, break-in and robbery are burdensome and challenging for human abilities to maintain.This system precisely aims to provide the solution by detecting these mentioned activities with comparatively less involvement of beings.To markup on these issues, surveillance systems for theft detection in supermarkets are introduced.It can be utilized in two methods, watching in real-time and the latter one is looking for evidence of what happened in the past.In this paper, we are presenting a system which is capable of alerting the owner about the activities in his\her absence.Also, static inputs which were recorded in the past can be uploaded.Frames are extracted from the video which makes the machine understand in a simplified manner.For object detection and assisting in an image, we have used InceptionV3.The model calculates the probability by considering the inputs while comparing them with the trained data.The main idea of the system is that they are trained using data sets about the classes and will be able to differentiate among them and alert the user in time, with an accuracy of 99.86%.