Suspicious Activity Detecting Camera
T Sivasakthi, S Brindha, Hariharasudhan S. M, Vishal V. S, M Priyadharsan · 2022 International Conference on Communication, Computing and Internet of Things (IC3IoT) · 2022
It is difficult to detect unusual activities in modern society with the help of existing surveillance systems. Therefore, a new method is proposed to improve the surveillance system. It can be used in several public places like banks, railway stations etc. It is done with the ML models which can be created with the help of Teachable Machine. To create and train an ML model, there are several flexible options such as TensorFlow, Google Teachable, Edge Impulse, Lobe etc. In this paper it is done with the help of Teachable Machine. In Google Teachable, select the Pose Net option for tracking the various body movements and actions. After feeding the required datasets, click on ‘Train Model’. A JavaScript code is obtained. Save the JavaScript code as. html and open it in any browser with JavaScript enabled. Click on the start button and perform the suspicious activity in front of the camera. The ML model will try to recognize it and if found unusual, will issue an alert. You can also add functions for automatically calling or messaging the police Then check whether activities such as beating, gun firing, gun possession etc., are detected by the ML model during video processing. The model gives the likelihood of a particular activity to occur in the range of 0.00 to 1.00. With the help of various functions alert messages can be sent.