Electricity conservation techniques have gained more importance in recent years. Many smart techniques are invented to save electricity with the help of assisted devices like sensors. Though it saves electricity, it adds an additional sensor cost to the system. This work aims to develop a system that manages the electric power supply, only when it is actually needed i.e., the system enables the power supply when a human is present in the location and disables it otherwise. The system avoids any additional costs by using the closed circuit television, which is installed in most of the places for security reasons. Human detection is done by a Modified-single shot detection with a specific hyperparameter tuning method. Further the model is pruned to reduce the computational cost of the framework which in turn reduces the processing speed of the network drastically. The model yields the output to the Arduino micro-controller to enable the power supply in and around the location only when a human is detected and disables it when the human exits. The model is evaluated on CHOKEPOINT dataset and real-time video surveillance footage. Experimental results have shown an average accuracy of 85.82% with 2.1 seconds of processing time per frame.
We explore the capability of localizing node failures in communication networks from binary states (normal/failed) of end-to-end paths. Certain a set of nodes of importance, individually localizing failures inside this set necessitates that dissimilar noticeable path states connect with dissimilar node malfunction events. However, this circumstance is easier said than done to test on huge networks due to the requirement to itemise all promising node failures. Our first donation is a set of satisfactory/compulsory conditions for classifying a restricted numeral of failures within an uninformed node set that can be experienced in polynomial time. In adding up to network topology and positions of monitors, our circumstances also include restrictions forced by the penetrating mechanism used. We are here considering three probing mechanisms basically which differ according as to whether dimension paths are: (i) arbitrarily controllable; (ii) controllable but cycle-free; or (iii) uncontrollable (which are dogged by the evasion routing protocol). Our second donation is to calculate the potential of malfunction localization from beginning to end: 1) the utmost number of failures (wherever in the network) such that malfunctions inside a given node set can be exceptionally localized and 2) the major node set inside which failures can be exclusively localized underneath a given vault on the total amount of failures. Here both the methods in 1) and 2) can be transformed into the functions of a per-node property, which can be computed resourcefully based on the above satisfactory/compulsory conditions. We reveal how process 1) and 2) projected for enumerating malfunction localization capability can be used to calculate the collision of various parameters which includes topology, number of monitors, and probing mechanisms.
The demand for electrical energy in developing countries is apparently increasing thereby creating a large gap between the availability of the electrical resource and its growing demand. Globally reputed energy economists have recognized that 25% of reduction in energy consumption can be achieved by adopting efficient energy conservation techniques This paper presents one of the simplest ways of conservation techniques that enables the electric power supply only when it is actually needed. It is an automatic system that functions with the existing CCTV surveillance camera to enable/disable the electric power supply, only in the location where human is present / absent respectively. The proposed approach is demonstrated without the use of sensors, based on Regional Convolutional Neural Network (R-CNN). A new R-CNN model is constructed for CHOKEPOINT dataset and the optimization is done using Nadam technique. The results are then fed into Arduino micro controller to control the electric supply based on the presence/absence of human in the particular region.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.