Fraktal Eğimden Arındırılmış Dalgalanma Analizi ile Yüzey Dokularının Sınıflandırılması

Öz

Dokunsal algılama, robotlara ve protezlere nesne tanıma, hassas manipülasyon, doğal etkileşim gibi yetenekler kazandırır. Dokunsal geri bildirim, fiziksel temas yoluyla bireylere dış çevreleri hakkında sürekli olarak hayati bilgiler sağlayarak önemli bir rol oynar. Bu nedenle, insan dostu biyomimetik elektronik ve esnek cihazlardaki hızlı gelişmeler, robotların özellikle tekstil gibi malzemeler için yerel geometri ve doku gibi malzeme özelliklerini ayırt etmelerini sağlar. Bu makalede, dokunsal sinyallere dayalı yüzey doku sınıflandırması için yeni bir yöntem önerilmiştir. Önerilen yöntemde, öncelikle 3 eksenli ivmeölçer (X, Y, Z) dokunsal sinyalleri ve mikrofon sinyalleri, örtüşmeyen kayan pencere yaklaşımı ile veri çoğaltma işlemine tabi tutulmuştur. Durağan olmayan zaman serilerinin kuvvet yasasının uzun vadeli korelasyonlarını araştırmak için etkili bir yöntem olan Fraktal Eğimden Arındırılmış Dalgalanma Analizi (F-EADA) ile öznitelik çıkarımı yapılmıştır. Son aşamada, yaygın olarak tercih edilen makine öğrenme algoritması olan Destek Vektör Makinası (DVM) ile ivmeölçer ve mikrofon sinyallerinden elde edilen öznitelikler ayrı ayrı ve birleştirilerek dokuların sınıflandırılması gerçekleştirilmiştir. Deneysel sonuçlar, pencere uzunluğu 1saniye olarak seçildiğinde ivmeölçer verilerinde %82,91, mikrofon verilerinde %98,33 ve her iki sensöre ait verilerin birlikte kullanımında ise %99,16 oranında sınıflandırma doğruluğu elde edilmiştir. Literatürdeki çalışmalar ile kıyaslandığında mikrofon verilerinde %12,08 ve ivmeölçer-mikrofon verilerinin birleştirilmesinde %0,56 daha yüksek sınıflandırma performans elde edilmiştir.

Anahtar Kelimeler

• Dokunsal sensör
• F-EADA
• Yüzey dokuları tanıma
• DVM.

Surface Textures Classification with Fractal Detrended Fluctuation Analysis

Öz

Tactile perception provides robots and prosthetics with capabilities such as object recognition, precise manipulation, and natural interaction. Tactile feedback plays an important role by continuously providing individuals with vital information about their external environment through physical contact. Therefore, rapid developments in human-friendly biomimetic electronics and flexible devices enable robots to distinguish material properties such as local geometry and texture, especially for materials such as textiles. In this paper, a new method for surface texture classification based on tactile signals is proposed. In the proposed method, firstly, 3-axis accelerometer (X, Y, Z) tactile signals and microphone signals are subjected to data augmentation with a non-overlapping sliding window approach. Feature extraction is performed with Fractal Detrended Fluctuation Analysis (FDFA), which is an effective method for investigating long-term correlations of power law of non-stationary time series. In the last stage, the textures were classified by using the Support Vector Machine (SVM), a widely preferred machine learning algorithm, using features obtained from accelerometer and microphone signals separately and combined. Experimental results show that when the window length is selected as 1 second, 82.91% classification accuracy is achieved for accelerometer data, 98.33% for microphone data, and 99.16% for the combined use of data from both sensors. Compared to studies in literature, 12.08% higher classification performance is achieved for microphone data and 0.56% higher classification performance is achieved when accelerometer-microphone data are combined.

Anahtar Kelimeler

• Tactile sensor
• FDFA
• Surface textures recognition
• SVM

Kaynakça

1. Alves de Oliveira, Thiago, Ana-Maria Cretu, and Emil Petriu. 2017. “Multimodal Bio-Inspired Tactile Sensing Module for Surface Characterization.” Sensors 17(6): 1187. doi:10.3390/s17061187.
2. BenYahmed, Yahyia, Azuraliza Abu Bakar, Abdul RazakHamdan, Almahdi Ahmed, and Sharifah Mastura Syed Abdullah. 2015. “Adaptive Sliding Window Algorithm for Weather Data Segmentation.” Journal of Theoretical and Applied Information Technology 80(2): 322–33.
3. Cao, Guanqun, Jiaqi Jiang, Danushka Bollegala, Min Li, and Shan Luo. 2024. “Multimodal Zero-Shot Learning for Tactile Texture Recognition.” Robotics and Autonomous Systems 176: 104688. doi:10.1016/j.robot.2024.104688.
4. Castiglioni, Paolo, and Andrea Faini. 2019. “A Fast DFA Algorithm for Multifractal Multiscale Analysis of Physiological Time Series.” Frontiers in Physiology 10. doi:10.3389/fphys.2019.00115.
5. Chi, Cheng, Xuguang Sun, Ning Xue, Tong Li, and Chang Liu. 2018. “Recent Progress in Technologies for Tactile Sensors.” Sensors 18(4): 948. doi:10.3390/s18040948.
6. Dallaire, Patrick, Philippe Giguère, Daniel Émond, and Brahim Chaib-draa. 2014. “Autonomous Tactile Perception: A Combined Improved Sensing and Bayesian Nonparametric Approach.” Robotics and Autonomous Systems 62(4): 422–35. doi:10.1016/j.robot.2013.11.011.
7. Hu, Diane, Liefeng Bo, and Xiaofeng Ren. 2011. “Toward Robust Material Recognition for Everyday Objects.” In Procedings of the British Machine Vision Conference 2011, British Machine Vision Association, 48.1-48.11. doi:10.5244/C.25.48.
8. Kılıç, Cemil, Ömer Faruk Alçin, and Muzaffer Aslan. 2024. “Dokunsal Sensör Sinyalleri Ile Yüzey Dokularının Sınıflandırılması.” Computer Science. doi:10.53070/bbd.1596239.

9. Kong, Yongkang, Guanyin Cheng, Mengqin Zhang, Yongting Zhao, Wujun Meng, Xin Tian, Bihao Sun, Fuping Yang, and Dapeng Wei. 2024. “Highly Efficient Recognition of Similar Objects Based on Ionic Robotic Tactile Sensors.” Science Bulletin 69(13): 2089–98. doi:10.1016/j.scib.2024.04.060.
10. Kursun, Olcay, and Ahmad Patooghy. 2020a. “An Embedded System for Collection and Real-Time Classification of a Tactile Dataset.” IEEE Access 8: 97462–73. doi:10.1109/ACCESS.2020.2996576.
11. Kursun, Olcay, and Ahmad Patooghy. 2020b. “Journal of Computer Science.” IEEE Access 8: 97462–73. doi:10.1109/ACCESS.2020.2996576.
12. Liu, Qi, Ming Ling, Yanxiang Zhu, Yibo Rui, and Rui Wang. 2024. “Preventing Short Violations in Clock Routing with an SVM Classifier before Powerplanning and Placement.” Microelectronics Journal 153. doi:10.1016/j.mejo.2024.106429.
13. Ma, Feihong, Yuliang Li, and Meng Chen. 2024. “Tactile Texture Recognition of Multi-Modal Bionic Finger Based on Multi-Modal CBAM-CNN Interpretable Method.” Displays 83: 102732. doi:10.1016/j.displa.2024.102732.
14. Necip, Mustapha, and Ana Maria Cretu. 2023. “Texture Classification Based on Sound and Vibro-Tactile Data †.” In Engineering Proceedings, Basel Switzerland: MDPI, 5. doi:10.3390/ecsa-10-16082.
15. Okamoto, Shogo, Hikaru Nagano, and Hsin Ni Ho. 2016. “Psychophysical Dimensions of Material Perception and Methods to Specify Textural Space.” In Pervasive Haptics: Science, Design, and Application, , 3–20. doi:10.1007/978-4-431-55772-2_1.
16. Peng, C.-K., S. V. Buldyrev, S. Havlin, M. Simons, H. E. Stanley, and A. L. Goldberger. 1994. “Mosaic Organization of DNA Nucleotides.” Physical Review E 49(2): 1685–89. doi:10.1103/PhysRevE.49.1685.
17. Peng, Yiyao, Ning Yang, Qian Xu, Yang Dai, and Zhiqiang Wang. 2021. “Recent Advances in Flexible Tactile Sensors for Intelligent Systems.” Sensors 21(16): 5392. doi:10.3390/s21165392.
18. Prescott, Tony, Ben Mitchinson, and Robyn Grant. 2011. “Vibrissal Behavior and Function.” Scholarpedia 6(10): 6642. doi:10.4249/scholarpedia.6642.
19. Qian, Xi-Yuan, Gao-Feng Gu, and Wei-Xing Zhou. 2011. “Modified Detrended Fluctuation Analysis Based on Empirical Mode Decomposition for the Characterization of Anti-Persistent Processes.” Physica A: Statistical Mechanics and its Applications 390(23–24): 4388–95. doi:10.1016/j.physa.2011.07.008.
20. Qin, Liguo, Luxin Hao, Xiaodong Huang, Rui Zhang, Shan Lu, Zheng Wang, Jianbo Liu, et al. 2024. “Fingerprint-Inspired Biomimetic Tactile Sensors for the Surface Texture Recognition.” Sensors and Actuators A: Physical 371: 115275. doi:10.1016/j.sna.2024.115275.
21. Rocha Lima, Bruno Monteiro, Thiago Eustaquio Alves de Oliveira, and Vinicius Prado da Fonseca. 2021. “Classification of Textures Using a Tactile-Enabled Finger in Dynamic Exploration Tasks.” In 2021 IEEE Sensors, IEEE, 1–4. doi:10.1109/SENSORS47087.2021.9639755.
22. Rocha Lima, Bruno Monteiro, Vinicius Prado da Fonseca, Thiago Eustaquio Alves de Oliveira, Qi Zhu, and Emil M. Petriu. 2020. “Dynamic Tactile Exploration for Texture Classification Using a Miniaturized Multi-Modal Tactile Sensor and Machine Learning.” SYSCON 2020 - 14th Annual IEEE International Systems Conference, Proceedings. doi:10.1109/SysCon47679.2020.9275871.
23. Shao, Shiliang, Ting Wang, Yun Su, Chen Yao, Chunhe Song, and Zhaojie Ju. 2021. “Multi-IMF Sample Entropy Features with Machine Learning for Surface Texture Recognition Based on Robot Tactile Perception.” International Journal of Humanoid Robotics 18(02): 2150005. doi:10.1142/S0219843621500055.
24. Strese, Matti, Clemens Schuwerk, Albert Iepure, and Eckehard Steinbach. 2017. “Multimodal Feature-Based Surface Material Classification.” IEEE Transactions on Haptics 10(2): 226–39. doi:10.1109/TOH.2016.2625787.
25. Vafaeipour, Majid, Omid Rahbari, Marc A. Rosen, Farivar Fazelpour, and Pooyandeh Ansarirad. 2014. “Application of Sliding Window Technique for Prediction of Wind Velocity Time Series.” International Journal of Energy and Environmental Engineering 5(2–3): 1–7. doi:10.1007/s40095-014-0105-5.
26. Vapnik, Vladimir N. 1999. “An Overview of Statistical Learning Theory.” IEEE Transactions on Neural Networks 10(5): 988–99. doi:10.1109/72.788640.
27. Zacharia, V., A. Bardakas, A. Anastasopoulos, M.A. Moustaka, E. Hourdakis, and C. Tsamis. 2024. “Design of a Flexible Tactile Sensor for Material and Texture Identification Utilizing Both Contact-Separation and Surface Sliding Modes for Real-Life Touch Simulation.” Nano Energy 127: 109702. doi:10.1016/j.nanoen.2024.109702.