Abstract
This paper presents an optimized implementation of a neural network for fall detection using a Silicon Retina stereo vision sensor. A Silicon Retina sensor is a bio-inspired optical sensor with special characteristics as it does not capture images, but only detects variations of intensity in a scene. The data processing unit consists of an event-based stereo matcher processed on a field programmable gate array (FPGA), and a neural network that is processed on a digital signal processor (DSP). The initial network used double-precision floating point arithmetic; the optimized version uses fixed-point arithmetic as it should be processed on a low performance embedded system. We focus on the performance optimization techniques for the DSP that have a major impact on the run-time performance of the neural network. In summary, we achieved a speedup of 48 for multiplication, 39.5 for additions, and 194 for the transfer functions and, thus, realized an embedded real-time fall detection system.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Giannakouris, K.: Regional population projections EUROPOP 2008: Most EU regions face older population profile in 2030. European Union (2010)
Bauer, R., Steiner, M.: Injuries in the European Union - Statistics Summary 2005-2007. European Union (2009)
Sivilotti, M.: Wiring consideration in analog vlsi systems with application to field programmable networks. Phd-thesis, California Institute of Technology (1991)
Fukushima, K., Yamaguchi, Y., Yasuda, M., Nagata, S.: An Electronic Model of the Retina. Proceedings of the IEEE 58 (1970)
Mead, C., Mahowald, M.K.: A silicon model of early visual processing. Neural Networks Journal 1 (1988)
Lichtsteiner, P., Posch, C., Delbruck, T.: A 128×128 120dB 30mW Asynchronous Vision Sensor that Responds to Relative Intensity Change. In: Proceedings of the IEEE International Solid-State Circuits Conference, SanFrancisco, USA (2006)
Lichtsteiner, P., Posch, C., Delbruck, T.: A 128×128 120 dB 15 μs Latency Asynchronous Temporal Constrast Vision Sensor. IEEE J. Solid-State Circuits 43 (2008)
Posch, C., Matolin, D., Wohlgenannt, R.: A QVGA 143 dB Dynamic Range Frame-Free PWM Image Sensor With Lossless Pixel-Level Video Compression and Time-Domain CDS. IEEE J. Solid-State Circuits 46 (2011)
Schraml, S., Belbachir, A., Milosevic, N., Schön, P.: Dynamic stereo vision system for real-time tracking. In: Proceedings of 2010 IEEE International Symposium on Circuits and Systems, ISCAS (2010)
Fu, Z., Delbruck, T., Lichtsteiner, P., Culurciello, E.: An Address-Event Fall Detector for Assisted Living Applications. IEEE Trans. Biomed. Circuits Syst. 2 (2008)
Wu, G.: Distinguishing fall activities from normal activities by velocity characteristics. J. Biomech. 11 (2000)
Anderson, D., Keller, J.M., Skubic, M., Chen, X., He, Z.: Recognizing Falls from Silhouettes. In: 28th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (2006)
Li, Q., Zhou, G., Stankovic, J.: Accurate, Fast Fall Detection Using Posture and Context Information. In: Proceedings of the 6th ACM Conference on Embedded Network Sensor Systems (2008)
Nyan, M.N., Tay, F.E.H., Mah, M.Z.E.: Application of motion analysis system in pre-impact fall detection. J. Biomech. 10 (2008)
Juang, C.F., Chang, C.M.: Human Body Posture Classification by a Neural Fuzzy Network and Home Care System Application. IEEE Trans. Syst., Man, Cybern. A, Syst., Humans 6 (2007)
Analog Devices: ADSP-BF537 Blackfin Processor Hardware Reference. Analog Devices (2005)
IEEE Std 754-2008: IEEE Standard for Floating-Point Arithmetic. IEEE Standard (2008)
Analog Devices: Fast Floating-Point Arithmetic Emulation on Blackfin Processors. Engineer-to-Engineer Note EE-185 (2007)
Humenberger, M., Schraml, S., Sulzbachner, C., Srp, A., Vajda, F.: Embedded Fall Detection with a neuronal Network and Bio-Inspired Stereo Vision. In: 2012 IEEE International Conference on Computer Vision and Pattern Recognition Workshops, CVPR Workshops (2012)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2012 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Sulzbachner, C., Humenberger, M., Srp, Á., Vajda, F. (2012). Optimization of a Neural Network for Computer Vision Based Fall Detection with Fixed-Point Arithmetic. In: Huang, T., Zeng, Z., Li, C., Leung, C.S. (eds) Neural Information Processing. ICONIP 2012. Lecture Notes in Computer Science, vol 7666. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-34478-7_3
Download citation
DOI: https://doi.org/10.1007/978-3-642-34478-7_3
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-34477-0
Online ISBN: 978-3-642-34478-7
eBook Packages: Computer ScienceComputer Science (R0)