Authors : Rabia Riaz, Sanam Shahla Rizvi, Ayesha Mushtaq, Sana Shokat and Se Jin Kwon
Abstract: Burglar detection security systems have become a necessity in this age because of the increasing break-in cases in urban cities thus making these systems essential for residential as well as office usage. This study investigated how to model an intrusion detection system based on deep learning. Two deep learning approaches named generic Deep Neural Networks (DNN) and Convolution Neural Networks (CNN) are used for the training of the dataset. The experimental results showed that CNN approach is more suitable for burglar detection as it gives high accuracy with a superior performance as compared to the generic DNN approach. CNN provides a new research method with the improved accuracy of human intrusion detection. Experimental results found that CNN is compatible to solve classification problems and significantly faster and precise as compared to traditional object detection methods.
Rabia Riaz, Sanam Shahla Rizvi, Ayesha Mushtaq, Sana Shokat and Se Jin Kwon, 2019. Burglar Detection using Deep Learning Techniques. Journal of Engineering and Applied Sciences, 14: 2672-2686.