شناسایی پایدار فعالیت فیزیکی انسان بر اساس سنسورهای گوشی هوشمند
الموضوعات :
مهدی یزدیان دهکردی
1
(دانشگاه یزد)
زهرا عابدی
2
(دانشگاه یزد)
نسیم خانی
3
(دانشگاه یزد)
الکلمات المفتاحية: ژیروسکوپشتاب‌سنج شناسایی فعالیت فیزیکی انسان کیفیت سنسورگوشی هوشمندنویز سنسور,
ملخص المقالة :
در سال‌‌های اخیر تشخیص فعالیت فیزیکی انسان از روی داده‌های گرفته‌شده توسط سنسورهای ژیروسکوب و شتاب‌سنج در گوشی هوشمند، مورد توجه پژوهشگران قرار گرفته است. در این مقاله با به کارگیری روش تحلیل مؤلفه‌های اساسی، ویژگی‌هایی با بعد پایین و مناسب استخراج شده و کارایی چند طبقه‌بندی‌کننده مختلف شامل ماشین بردار پشتیبان، رگرسیون منطقی، ادابوست و شبکه عصبی کانولوشن برای طبقه‌بندی فعالیت‌ها بررسی و یک سیستم کارا برای این منظور پیشنهاد شده است. نتایج به دست آمده نشان می‌دهد که سیستم پیشنهادی توانسته است دقت تشخیص را نسبت به کارهای اخیر بهبود دهد. یکی از چالش‌هایی که لازم است در خصوص سیستم‌های تشخیص فعالیت مورد توجه قرار گیرد، میزان پایداری این سیستم‌ها نسبت به مدل‌های مختلف از گوشی‌های هوشمند است. با توجه به این که کیفیت سنسورها و نویز مرتبط با آنها از یک مدل گوشی به مدل دیگر متفاوت است، بنابراین بررسی میزان پایداری الگوریتم شناسایی فعالیت در نویزهای مختلف حایز اهمیت خواهد بود. در این مقاله کارایی و میزان پایداری طبقهبندیکنندهها در سطوح مختلف نویز نیز بررسی شده است. نتایج به دست آمده نشان می‌دهد که ماشین بردار پشتیبان با میانگین دقت 34/96% پایداری بهتری نسبت به نویز در مقایسه با سایر طبقهبندیکنندهها داشته است.
[1] D. Anguita, A. Ghio, L. Oneto, X. Parra, and J. L. Reyes-Ortiz, "A public domain dataset for human activity recognition using smartphones," in Proc. 21st European Symp. on Artificial Neural Networks, Computational Intelligence and Machine Learning, pp. 437-442, Bruges, Belgium, 24-26 Apr. 2013.
[2] S. Sun, Z. Kuang, L. Sheng, W. Ouyang, and W. Zhang, "Optical flow guided feature: a fast and robust motion representation for video action recognition," in Proc. of the IEEE Computer Society Conf. on Computer Vision and Pattern Recognition, pp. 1390-1399, Salt Lake City, UT, USA, 18-22 Jun. 2018.
[3] M. Safaei, P. Balouchian, and H. Foroosh, "TICNN: a hierarchical deep learning framework for still image action recognition using temporal image prediction," in Proc. Inte. Conf. on Image Processing, ICIP’18, pp. 3463-3467, Athens, Greece, 7-10 Oct. 2018.
[4] B. Kwon, J. Kim, K. Lee, Y. K. Lee, S. Park, and S. Lee, "Implementation of a virtual training simulator based on 360° multi-view human action recognition," IEEE Access, vol. 5, no. 1, pp. 12496-12511, Jul. 2017.
[5] J. Cao, W. Li, C. Ma, and Z. Tao, "Optimizing multi-sensor deployment via ensemble pruning for wearable activity recognition," Inf. Fusion, vol. 41, no. 1, pp. 68-79, May 2018.
[6] A. Doewes, S. E. Swasono, and B. Harjito, "Feature selection on human activity recognition dataset using minimum redundancy maximum relevance," in Proc. IEEE Int. Conf. on Consumer Electronics-Taiwan, pp. 171-172, Taipei, Taiwan, 12-14 Jun. 2017.
[7] H. Martin, A. M. Bernardos, J. Iglesias, and J. R. Casar, "Activity logging using lightweight classification techniques in mobile devices," Pers. Ubiquitous Comput., vol. 17, no. 4, pp. 675-695, Apr. 2013.
[8] M. Shoaib, H. Scholten, and P. J. M. Havinga, "Towards physical activity recognition using smartphone sensors," in Proc. IEEE 10th Int. Conf. on Ubiquitous Intelligence and Computing, and IEEE 10th Int. Conf. on Autonomic and Trusted Computing, pp. 80-87, Vietri sul Mere, Italy, 18-21 Dec. 2013..
[9] I. Suarez, A. Jahn, C. Anderson, and K. David, "Improved activity recognition by using enriched acceleration data," in Proc. of the ACM Int. Joint Conf. on Pervasive and Ubiquitous Computing, pp. 1011-1015, Osaka, Japan, 7-11 Sept. 2015.
[10] C. A. Ronao and S. B. Cho, "Human activity recognition with smartphone sensors using deep learning neural networks," Expert Syst. Appl., vol. 59, no. 1, pp. 235-244, Oct. 2016.
[11] D. N. Tran and D. D. Phan, "Human activities recognition in android smartphone using support vector machine," in Proc. IntConf. on Intelligent Systems, Modelling and Simulation, ISMS’17, pp. 64-68, Bangkok, Thailand, 25-27 Jan. 2016.
[12] J. Silva, M. Monteiro, and F. Sousa, "Human activity classification with inertial sensors," Studies in Health Technology and Informatics, vol. 200, no. 1, pp. 101-104, Jun. 2014.
[13] A. Kos, S. Tomazic, and A. Umek, "Evaluation of smartphone inertial sensor performance for cross-platform mobile applications," Sensors (Switzerland), vol. 16, no. 4, pp. 477-492, Apr. 2016.
[14] K. Nirmal, et al., "Noise modeling and analysis of an IMU-based attitude sensor: improvement of performance by filtering and sensor fusion," Advances in Optical and Mechanical Technologies for Telescopes and Instrumentation II, vol. 9912, no. 1, pp. 1-10, Jul. 2016.
[15] C. K. I. Williams, "Learning with kernels: support vector machines, regularization, optimization, and beyond," J. Am. Stat. Assoc., vol. 98, no. 462, pp. 489-489, Jun. 2003.
[16] M. Galar, A. Fernandez, E. Barrenechea, H. Bustince, and F. Herrera, "An overview of ensemble methods for binary classifiers in multi-class problems: experimental study on one-vs-one and one-vs-all schemes," Pattern Recognition, vol. 44, no. 8, pp. 1761-1776, Aug. 2011.
[17] C. J. Lin, R. C. Weng, and S. Sathiya Keerthi, "Trust region Newton method for large-scale logistic regression," J. of Machine Learning Research, vol. 9, no. 7, pp. 627-650, Jun. 2007.
[18] Y. Freund and R. E. Schapire, "Experiments with a new boosting algorithm," in Proc. Int Conf. on Machine Learning, vol. 96, pp. 148-156, Bari, Italy, 3-6 Jul. 1996.
[19] W. Iba Ai and P. Langley, "Induction of one-level decision trees," in Proc. of 9th Int. Conf. on Machine Learning, vol. 1, pp. 233-240, Aberdeen, Scotland, 1-3 Jul. 1992.
[20] S. U. Khan, N. Islam, Z. Jan, I. Ud Din, and J. J. P. C. Rodrigues, "A novel deep learning based framework for the detection and classification of breast cancer using transfer learning," Pattern Recognit. Letter, vol. 125, no. 1, pp. 1-6, Jul. 2019..
[21] X. Zhao, Y. Wu, G. Song, Z. Li, Y. Zhang, and Y. Fan, "A deep learning model integrating FCNNs and CRFs for brain tumor segmentation," Medical Image Analysis., vol. 43, no. 1, pp. 98-111, Jan. 2018.
[22] L. Wang, X. Qian, Y. Zhang, J. Shen, and X. Cao, "Enhancing sketch-based image retrieval by CNN semantic re-ranking," IEEE Trans. Cybern., vol. 50, no. 7, pp. 3330-3342, Jul. 2020.
[23] W. Liu, Z. Wang, X. Liu, N. Zeng, Y. Liu, and F. E. Alsaadi, "A survey of deep neural network architectures and their applications," Neurocomputing, vol. 234, no. 1, pp. 11-26, Apr. 2017.
[24] J. Schmidhuber, "Deep learning in neural networks: an overview," Neural Networks, vol. 61, no. 1, pp. 85-117, Jan. 2015.
[25] B. Romera-Paredes, M. S. H. Aung, and N. Bianchi-Berthouze, "A one-vs-one classifier ensemble with majority voting for activity recognition," in Proc. 21st European Symp on Artificial Neural Networks, Computational Intelligence and Machine Learning, pp. 443-448, Bruges, Belgium, 24-26 Apr. 2013.
[26] T. D. T. Nguyen, T. T. Huynh, and H. A. Pham, "An improved human activity recognition by using genetic algorithm to optimize feature vector," in Proc. 10th Int. Conf. on Knowledge and Systems Engineering, KSE’18, pp. 123-128, pp. 123–128, Ho Chi Minh City, Vietnam, 1-3 Nov. 2018.
