Human Behaviour Monitoring and Analysis for Efficiently Energy Management
Mentor: Marta Marrón Romera
Phone: (+34) 918856586
(+34) 630201843
University: University of Alcalá
Partner Host Institution:
Keywords: Human-Behavior-Analysis (HAR), Energy-Efficiency-Management, DeepLearning, Video-Surveillance

Human Behaviour Monitoring and Analysis for Efficiently Energy Management

The GOT-Energy-TALENT proposal hereby presented by GEINTRA Group involves tasks related to the monitoring and analysis of different indicators in the context of home and city surveillance in order to optimize the smart cities capabilities in terms of energy management and security. The works under this proposal are thus focused in two relevant aims in technology supporting current and future societies: security and energy.

This proposal in the GOT-Energy-Talent call is oriented to support the transition to a reliable, sustainable and competitive smart city, within the Energy Efficiency topic: being a no-regret option for Europe, energy consumption is needed to progressively decrease by 2020 to 2030.

The overall objective of the proposal focuses on understanding what is happening in a scene of interest, and its relation to security and energy efficiency analysis, including both activities and behavior of groups and individuals. The main target is to automatically process the different scenarios of interest and observe the typical activities that take place there in order to perform security and energy efficiency analysis.

Thus, main subtopics of the proposal aim to identify relevant features of human activity in a given sensed area (either indoor or outdoor). These features will be related to physical behaviour (number of people in the scene, their relative positions and movement patterns, their physical complexity, their dress conditions, etc.), and also semantic behaviour of individuals and groups (interaction among different people in the scene, activities being carried out, people and group behaviour, etc.). The output of the proposal works will be a compound physical-semantic feature structure that can be further related to security information (possible security threats, and anomalities in the scene under consideration), and also energy related issues (need for specific lighning, heating and cooling conditions in order to minimize energy waste, by adaptation to the physical conditions of the detected people, and to their behavioural patterns).

From a technological and algorithmic point of view, the proposal will heavily depends on the integration of varied distributed sensors, and machine learning strategies both for physical feature monitoring and for human behaviour pattern analysis, including a throrough study of the interaction between these patterns and those related to security issues and energy pattern behaviour, with specific emphasis on current deep learning approaches.

Departament: Electronics
Research Group: GEINTRA
More Information:
Relevants projects on the area: HEIMDAL: "Multisensory semantic detection of anomalous situations in environments without restrictions" Ref: TIN2016-75982-C2-1-R Ministry of Economy and Competitiveness 30 Dec 2016 - 29 Dec 2020
Relevants publications on the area: 1.- Carlos A. Luna, Javier Macias-Guarasa, Cristina Losada, Marta Marron, Manuel Mazo, Sara Luengo-Sanchez, Roberto Macho-Pedroso, "Headgear Accessories Classification Using an Overhead Depth Sensor", Sensors, 17:8, 2017, doi = {10.3390/s17081845}
2.- Alvaro Marcos-Ramiro, Daniel Pizarro, Marta Marron, Daniel Gatica-Perez, "Let Your Body Speak: Communicative Cue Extraction on Natural Interaction using RGBD Data", IEEE Transactions on Multimedia, 2015, doi = {10.1109/TMM.2015.2464152}