Smart Campus Surveillance: Detecting Rule Breakers with Deep Learning

Advancement in Technology have created huge impact on all sectors like Education, Business and Industrial sector. Deep learning algorithms have made these technology to another level in terms of speed, accuracy and efficiency of the data. Deep learning algorithms can be implemented in educational sector to improves its advancement in technology. This project implements Deep Learning for Smart Campus Surveillance which detects rule breakers effectively. Wearing identity card is one of the major rule in campus but avoiding them, it will be considered as rule breaker. There are few dataset available for detecting identity card in campus. For this problem, Dataset of detecting identity cards have been created with 6000 images. Advanced Deep learning algorithms like YOLO (You Only Look Once) for object-detection and Convolutional neural networks(CNN) for facial recognition. Yolov7 detects the identity card with two classes id card wearing and identity card not wearing datasets. It produces 70.6% mAP, 91.3% precision, and78.5
^%
recall.