Real-Time Surveillance for Critical Activity Detection in ICUs

Abstract

In today’s scenario, motion detection has gathered attention of researchers due to its promising applications in numerous areas. Such as video surveillance, patient monitoring, traffic management, security, video games, military armors, object classification, and sign language recognition. The review of this intelligent application demands gradation of technologies. The findings say that it is still in its early developmental stage and there is need to improve its robustness when applied to a complex and changing environment. Therefore, it is effective to improve the surveillance techniques. Upgradation can be obtained successfully using 3D camera, but it is expensive. The methodology mentioned in this paper follows the human eyes visualization concept by using pair of identical two-dimensional cameras to generate stereoscopic video. Growing number of cameras enables new signal processing applications. The amount of the data also increases which is to be processed to be supportive to design new modified algorithm to obtain accuracy. This can also track the movement using optimized Kalman filter. This real-time method is useful in monitoring and detecting every inch and second of information of interested areas. The technique mentioned in this paper is beneficial for online and offline applications. Project implemented using this paper trim down memory requirements for activity storage. It is efficient, sensitive, and absolutely useful for society welfare. The proposed method in this paper will activate a warning system, highlight the changes, and capture the live streaming video when minute movement of coma patient is detected, also it keeps track on mental stress of patients.