Evaluating Resources Cost of a Convolutional Neural Network Aiming an Embedded System
Resumo
The use of Machine Learning algorithms in image classification problems have yielded satisfactory results in recent years. Classification algorithms such as Support Vector Machines (SVMs) combined with robust feature extractors like Histogram of Oriented Gradients (HOG) have been used to achieve accuracy results over 95%. Very recently, with the researches applied in the deep learning fields, Convolutional Neural Networks (CNNs) have shown to work extremely well with data that has high dimensionality like images. This work focuses on evaluating the resources costs of deploying a CNN in an embedded platform to solve the people detection problem. An implementation of the CNN classification algorithm was developed, and tests were carried out both in a PC and in an embedded platform. Furthermore, a study on the amount of memory and time spent by a classic CNN was executed. The results point out that new networks must be designed to fit in low-resources embedded platforms.
Referências
Andersson, Bengt J.: On measurement of velocity by pitot tube. Ark. Mat., 3(5):391–394, January 1958.
BEA: Final Report On the accident on 1st June 2009 to the Airbus A330-203. Bureau d’Enquetes et d’Analyses, 2012.
Haasl, Sjoerd and Göran Stemme: Comprehensive Microsystems Chapter 2.07: Flow Sensors. Elsevier Science, 2007.
UTC: SmartProbe® Air Data Systems. UTC Aerospace Systems, 2015.
SAE: Guidelines and Methods for Conducting the Safety Assessment Process on Civil Airborne Systems and Equipment. SAE International, 1996.
SAE: Guidelines for Development of Civil Aircraft and Systems. SAE International, 2010.
Makinwa, Kofi A. A. and Johan H. Huijsing: A wind-sensor interface using thermal sigma delta modulation techniques. Sensors and Actuators A 92 (2001) 280-285, 2000.
Makinwa, Kofi A. A. and Johan H. Huijsing: A smart wind sensor using thermal sigma-delta modulation techniques. Sensors and Actuators A 97-98 (2002) 15-20, 2001.
S. Askari, M. Nourani and A. Namazi: Fault-tolerant a/d converter using analogue voting. IET Circuits, Devices & Systems, 2011.
Alves, Emerson M. A. and Frank S. Torres: Aumento de confiabilidade de sistemas embutidos usando redundância e algoritmos de decisões baseados em reconhecimento de padrões. XXI Congresso Brasileiro de Automática - CBA2016, 2016.
Oliveira, Livia Camargos Rodrigues de and Roberto Kawakami Harrop Galvao: A four-signal voting algorithm for aircraft redundant sensors. 21st Brazilian Congress of Mechanical Engineering - COBEM 2011, 2011.
