Gemini
Orbital (a.k.a., CP2106: Independent Software Development Project) is the School of Computing’s Year 1 self-directed summer module.
For this project, my partner and I embarked on a rather unique problem threatening our society today. With an increase in the amount of violence island-wide with the use of weapons like knives, katanas, guns; and obviously fist-fighting cases, we felt a call to increase the safety in Singapore - namely, NUS. Albeit the tight security measures here, we decided that it was time for an improvement. With promising advances in Computer Vision in recent years, we seized this opportunity to work on a project that could potentially assist NUS Campus Security and make our school a better place for our fellow students, cultivating a safe and secure space for learning.
As such, we worked on existing algorithms aimed at improving surveillance by identifying weaponry and violent behaviour through CCTV systems using emerging technologies in Computer Vision. Using state-of-the-art convolutional neural networks to classify and display defined classes (e.g. guns, knives etc.) trained using our custom labeled datasets, our project aims to decrease the necessary workload and manpower required by the security team as they would not need to constantly monitor all surveillance footage simultaneously. #project #computervision #machinelearning
Project README: https://lnkd.in/dsJrgv5y
Project Log: https://lnkd.in/djMBTnFG
Project Video: https://lnkd.in/dpT-Hwa8