• Kiran Yesugade
• Vaishnavi Bankar
• Priya Raut
• Payal Salunkhe
• Shreya Sonawane

Due to the increased crime rate around the world , various organizations have started deploying surveillance systems at their locations with the help of CCTV cameras. Video surveillance plays a vital role in today’s world. This system helps in detecting suspicious activities without human intervention. In this paper , we aim to automatically track people and detect unusual or suspicious activities from surveillance video and alert the shop owners. Hence, the main motive of this system is to take real-time videos from CCTV as an input and pass it to the CNN model created with the help of transfer learning and detect ‘Shoplifting’, ‘Robbery’ or ’Break-In’ in the store and notify it to the owners as soon as it occurs. Monitoring of activities is performed through consecutive frames which are extracted from the video. In general , we introduced a framework that processes raw data received from a camera which is installed at a particular location. Firstly, the proposed framework helps in detecting objects and tracking activities and then the activities are classified and it results in generating alerts to the authorized person.

surveillance , suspicious activity, human intervention, monitoring, authorized person.

"DEEP LEARNING APPROACH FOR SUSPICIOUS ACTIVITY DETECTION", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.10, Issue 2, page no.f730-f733, February-2023, Available :http://www.jetir.org/papers/JETIR2302581.pdf

"DEEP LEARNING APPROACH FOR SUSPICIOUS ACTIVITY DETECTION", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.10, Issue 2, page no. ppf730-f733, February-2023, Available at : http://www.jetir.org/papers/JETIR2302581.pdf