Smart Home Surveillance System and Intruder Detection Using Local Binary Pattern Histogram
N. Archana, R. Menaka, R. Jothiraj, S. Kalidass · 2022
In today’s society, there has been an increase in the number of suspicious and offensive behaviours. Many public areas, such as shopping malls, banks, and hotels, have CCTV cameras installed to safeguard the safety of their customers and visitors. CCTV cameras are now installed in many homes, particularly in those with single individuals and the elderly. Because human observation of these security cameras 24 hours a day, seven days a week is nearly impossible. It necessitates a workforce and their undivided attention in order to determine whether the collected activities are unusual or suspicious. As a result, this flaw necessitates the necessity to automate this procedure with high precision, as well as the ability to determine which frames and areas of the recording contain suspicious activity, allowing for faster judgment. As a result, we are using deep learning models for Intelligent Threat Recognition System to save time and labour waste. Its purpose is to detect signals of enmity in real time, allowing abnormalities to be distinguished from typical patterns. This method reports the technologies available in Open Computer Vision libraries and the methodologies to implement them using python. To provide the user friendly interface tkinter GUI interface is used. It uses haar cascade to detect the face and Local Binary pattern Algorithm to recognize the face. It also send a alert mail if it finds any uncommon activities.