Convolutional Neural Network on Embedded Platform for People Presence Detection in Low Resolution Thermal Images
Authors:
- 1 May 2019
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 7610-7614
Abstract
Detection of human presence is a key feature in Human Computer Interaction. Solutions based on cameras are attractive, but require computer vision techniques to extract meaningful data, which can be expensive from a computational point of view. In this work, we present a new system that merges a low resolution thermal camera with advanced feature extraction techniques such as Convolutional Neural Networks. We demonstrate the possibility to adapt their execution to resource-constrained platform without significant loss of performance, by processing data on a 32-bit low power microcontroller, performing the classification on thermal video stream. It achieve 76.7% of accuracy in the microcontroller, requiring only 16.5 mW in continuous classification mode and using 6 kB of RAM.Keywords
This publication has 13 references indexed in Scilit:
- Synthetic Sensors2017
- Optimizing Memory Efficiency for Deep Convolutional Neural Networks on GPUs2016
- Edge Computing: Vision and ChallengesIEEE Internet of Things Journal, 2016
- An Event-Driven Ultra-Low-Power Smart Visual SensorIEEE Sensors Journal, 2016
- A Comparative Study of Vision Based Human Detection Techniques in People Counting ApplicationsProcedia Computer Science, 2015
- A comprehensive review of background subtraction algorithms evaluated with synthetic and real videosComputer Vision and Image Understanding, 2014
- People Counting and Human Detection in a Challenging SituationIEEE Transactions on Systems, Man, and Cybernetics - Part A: Systems and Humans, 2010
- Challenges of passive infrared indoor localization2008
- Robust Multi-Pedestrian Tracking in Thermal-Visible Surveillance Videos2006
- Pyroelectric devices and materialsReports on Progress in Physics, 1986