CNN–Aided Optical Fiber Distributed Acoustic Sensing for Early Detection of Red Palm Weevil: A Field Experiment †
Abstract
:1. Introduction
2. Experimental Setup
3. Classifying “Infested” and “Healthy” Acoustic Signals Using CNNs
4. Classifying Infested and Healthy Trees Using CNNs
5. Discussion
6. Conclusions
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Conflicts of Interest
References
- Food Chain Crisis. Available online: http://www.fao.org/food-chain-crisis/how-we-work/plant-protection/red-palm-weevil/en/ (accessed on 14 December 2020).
- Wahizatul, A.A.; Zazali, C.; Abdul, R.; Nurul’Izzah, A.G. A new invasive coconut pest in Malaysia: The red palm weevil (Curculionidae: Rhynchophorus ferrugineus). Planter 2013, 89, 97–110. [Google Scholar]
- Ferry, M.; Gomez, S. The red palm weevil in the Mediterranean area. Palms 2002, 46, 172–178. [Google Scholar]
- El-Mergawy, R.; Al-Ajlan, A. Red palm weevil, Rhynchophorus ferrugineus (Olivier): Economic importance, biology, biogeography and integrated pest management. J. Agric. Sci. Technol. A 2011, 1, 1–23. [Google Scholar]
- Mukhtar, M.; Rasool, K.G.; Parrella, M.P.; Sheikh, Q.I.; Pain, A.; Lopez-Llorca, L.V.; Aldryhim, Y.N.; Mankin, R.; Aldawood, A.S. New initiatives for management of red palm weevil threats to historical Arabian date palms. Fla. Entomol. 2011, 94, 733–736. [Google Scholar] [CrossRef]
- Llácer, E.; Miret, J.A.J. Efficacy of phosphine as a fumigant against Rhynchophorus ferrugineus (Coleoptera: Curculionidae) in palms. Span. J. Agric. Res. 2010, 3, 775–779. [Google Scholar] [CrossRef]
- Mankin, R. Towards user-friendly early detection acoustic devices and automated monitoring for red palm weevil management. In Proceedings of the Symposium Proceedings, Amsterdam, The Netherlands, 25–27 September 2019; pp. 142–147. [Google Scholar]
- Suma, P.; La Pergola, A.; Longo, S.; Soroker, V. The use of sniffing dogs for the detection of Rhynchophorus ferrugineus. Phytoparasitica 2014, 42, 269–274. [Google Scholar] [CrossRef]
- Rizzolo, A.; Bianchi, G.; Lucido, P.; Cangelosi, B.; Pozzi, L.; Villa, G.; Clematis, F.; Pasini, C.; Curir, P. Electronic nose for the early detection of red palm weevil (Rhynchophorus ferrugineous Olivier) infestation in palms: Preliminary results. In Proceedings of the II International Symposium on Horticulture in Europe 1099, Angers, France, 1–5 July 2012; pp. 347–355. [Google Scholar]
- Haff, R.; Slaughter, D. Real-time X-ray inspection of wheat for infestation by the granary weevil, Sitophilus granarius (L.). Trans. ASAE 2004, 47, 531. [Google Scholar] [CrossRef]
- Golomb, O.; Alchanatis, V.; Cohen, Y.; Levin, N.; Cohen, Y.; Soroker, V. Detection of red palm weevil infected trees using thermal imaging. In Precision Agriculture’15; Wageningen Academic Publishers: Wageningen, The Netherlands, 2015; pp. 322–337. [Google Scholar]
- Ashry, I.; Mao, Y.; Al-Fehaid, Y.; Al-Shawaf, A.; Al-Bagshi, M.; Al-Brahim, S.; Ng, T.K.; Ooi, B.S. Early detection of red palm weevil using distributed optical sensor. Sci. Rep. 2020, 10, 3155. [Google Scholar] [CrossRef]
- Rach, M.M.; Gomis, H.M.; Granado, O.L.; Malumbres, M.P.; Campoy, A.M.; Martín, J.J.S. On the design of a bioacoustic sensor for the early detection of the red palm weevil. Sensors 2013, 13, 1706–1729. [Google Scholar] [CrossRef]
- Siriwardena, K.; Fernando, L.; Nanayakkara, N.; Perera, K.; Kumara, A.; Nanayakkara, T. Portable acoustic device for detection of coconut palms infested by Rynchophorus ferrugineus (Coleoptera: Curculionidae). Crop. Prot. 2010, 29, 25–29. [Google Scholar] [CrossRef]
- Wang, B.; Mao, Y.; Ashry, I.; Al-Fehaid, Y.; Al-Shawaf, A.; Ng, T.K.; Yu, C.; Ooi, B.S. Towards Detecting Red Palm Weevil Using Machine Learning and Fiber Optic Distributed Acoustic Sensing. Sensors 2021, 21, 1592. [Google Scholar] [CrossRef] [PubMed]
- Ashry, I.; Mao, Y.; Wang, B.; Sait, M.; Guo, Y.; Al-Shawaf, A.; NG, T.K.; Ooi, B.S. CNN-based detection of red palm weevil using optical-fiber-distributed acoustic sensing. In Proceedings of the Photonic Instrumentation Engineering IX, San Francisco, CA, USA, 22–27 January 2022; SPIE: Bellingham, WA, USA, 2022; Volume 12008, p. 120080U-1. [Google Scholar]
- Bao, X.; Zhou, D.P.; Baker, C.; Chen, L. Recent development in the distributed fiber optic acoustic and ultrasonic detection. J. Light. Technol. 2016, 35, 3256–3267. [Google Scholar] [CrossRef]
- Taylor, H.F.; Lee, C.E. Apparatus and Method for Fiber Optic Intrusion Sensing. U.S. Patent 5,194,847, 16 March 1993. [Google Scholar]
- Juškaitis, R.; Mamedov, A.; Potapov, V.; Shatalin, S. Distributed interferometric fiber sensor system. Opt. Lett. 1992, 17, 1623–1625. [Google Scholar] [CrossRef]
- Lu, Y.; Zhu, T.; Chen, L.; Bao, X. Distributed vibration sensor based on coherent detection of phase-OTDR. J. Light. Technol. 2010, 28, 3243–3249. [Google Scholar]
- Ashry, I.; Mao, Y.; Alias, M.S.; Ng, T.K.; Hveding, F.; Arsalan, M.; Ooi, B.S. Normalized differential method for improving the signal-to-noise ratio of a distributed acoustic sensor. Appl. Opt. 2019, 58, 4933–4938. [Google Scholar] [CrossRef]
- Schmidhuber, J. Deep learning in neural networks: An overview. Neural Networks 2015, 61, 85–117. [Google Scholar] [CrossRef] [PubMed]
- Chen, H.; Miao, F.; Chen, Y.; Xiong, Y.; Chen, T. A hyperspectral image classification method using multifeature vectors and optimized KELM. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. 2021, 14, 2781–2795. [Google Scholar] [CrossRef]
- Yao, R.; Guo, C.; Deng, W.; Zhao, H. A novel mathematical morphology spectrum entropy based on scale-adaptive techniques. ISA Trans. 2022, 126, 691–702. [Google Scholar] [CrossRef]
- Wu, H.; Chen, J.; Liu, X.; Xiao, Y.; Wang, M.; Zheng, Y.; Rao, Y. One-Dimensional CNN-Based Intelligent Recognition of Vibrations in Pipeline Monitoring With DAS. J. Light. Technol. 2019, 37, 4359–4366. [Google Scholar] [CrossRef]
- Mao, Y.; Ashry, I.; Alias, M.S.; Ng, T.K.; Hveding, F.; Arsalan, M.; Ooi, B.S. Investigating the performance of a few-mode fiber for distributed acoustic sensing. IEEE Photonics J. 2019, 11, 1–10. [Google Scholar] [CrossRef]
- Pinhas, J.; Soroker, V.; Hetzroni, A.; Mizrach, A.; Teicher, M.; Goldberger, J. Automatic acoustic detection of the red palm weevil. Comput. Electron. Agric. 2008, 63, 131–139. [Google Scholar] [CrossRef]
- Potamitis, I.; Ganchev, T.; Kontodimas, D. On automatic bioacoustic detection of pests: The cases of Rhynchophorus ferrugineus and Sitophilus oryzae. J. Econ. Entomol. 2009, 102, 1681–1690. [Google Scholar] [CrossRef]
- Gutiérrez, A.; Ruiz, V.; Moltó, E.; Tapia, G.; del Mar Téllez, M. Development of a bioacoustic sensor for the early detection of Red Palm Weevil (Rhynchophorus ferrugineus Olivier). Crop. Prot. 2010, 29, 671–676. [Google Scholar] [CrossRef]
- Tofailli, K. The early detection of red palm weevil: A new method. In Proceedings of the IV International Date Palm Conference 882, Abu Dhabi, United Arab Emirates, 15–17 March 2010; pp. 441–449. [Google Scholar]
- Hetzroni, A.; Soroker, V.; Cohen, Y. Toward practical acoustic red palm weevil detection. Comput. Electron. Agric. 2016, 124, 100–106. [Google Scholar] [CrossRef]
- Rasool, K.G.; Husain, M.; Salman, S.; Tufail, M.; Sukirno, S.; Mehmood, K.; Farooq, W.A.; Aldawood, A.S. Evaluation of some non-invasive approaches for the detection of red palm weevil infestation. Saudi J. Biol. Sci. 2020, 27, 401–406. [Google Scholar] [CrossRef]
- Koubaa, A.; Aldawood, A.; Saeed, B.; Hadid, A.; Ahmed, M.; Saad, A.; Alkhouja, H.; Ammar, A.; Alkanhal, M. Smart Palm: An IoT framework for red palm weevil early detection. Agronomy 2020, 10, 987. [Google Scholar] [CrossRef]
- Mohamed, A.; Hany, A.; Adly, I.; Atwa, A.; Ragai, H. AI for Acoustic Early Detection of the Red Palm Weevil. In Proceedings of the 2021 16th International Conference on Computer Engineering and Systems (ICCES), Cairo, Egypt, 15–16 December 2021; pp. 1–4. [Google Scholar]
- Kagan, D.; Alpert, G.F.; Fire, M. Automatic large scale detection of red palm weevil infestation using street view images. ISPRS J. Photogramm. Remote Sens. 2021, 182, 122–133. [Google Scholar] [CrossRef]
- Karar, M.E.; Abdel-Aty, A.H.; Algarni, F.; Hassan, M.F.; Abdou, M.; Reyad, O. Smart IoT-based system for detecting RPW larvae in date palms using mixed depthwise convolutional networks. Alex. Eng. J. 2022, 61, 5309–5319. [Google Scholar] [CrossRef]
Method | Processing Technique | Invasive or Not | Performance or Accuracy | Advantages (Disadvantages) |
---|---|---|---|---|
An acoustic sensor (commercial piezoelectric microphone), 2008 [27] | Speech recognition method, vector quantization (VQ), and Gaussian mixture modeling (GMM) | Not | 98% accuracy | Automatic detection using simple commercial hardware (A sound-isolated box is used) |
An acoustic sensor (Piezoelectric sensor), 2009 [28] | Feature extraction, GMM | Invasive | 99.1% accuracy | Automatic detection with well-designed algorithms (High computational complexity) |
An acoustic sensor (electronic device with acoustic probe), 2010 [29] | FFT, studying the sound intensity around 2250 Hz | Invasive | The infested sound intensity increases around 1 dB from −20 dB | Detection of a small number of larvae with a simple signal processing method (Low contrast between infested and non-infested sound) |
An acoustic device (acoustic probe and headphone set), 2010 [14] | Bandpass filtering, amplification | Invasive | 97% accuracy | Simple and portable hardware (Manual identification with four detection positions needed) |
A radiography system (X-ray technology), 2012 [30] | Visual detection based on X-ray photos | Not | Observable larvae on the photos | Simple and visual operation (Difficult for large-scale applications) |
An acoustic sensor (audio probe), 2013 [13] | Filtering and amplification, feature vector quantization | Invasive | 90% accuracy | Autonomous and continuous detection with explicit audio analysis algorithm (Extensive field experiments are needed in the future) |
Thermal imaging (infrared thermal camera), 2015 [11] | Thermal infrared images (TIR), leaf temperature maps, canopy representative temperature, crop water stress index (CWSI) | Not | Less than 75% accuracy | Large-scale and non-invasive detection (Susceptible to environmental conditions) |
An acoustic sensor (piezoelectric microphone), 2016 [31] | Likelihood indication by observer, speech recognition algorithm same as that in Ref. [27] | Not | 75% accuracy by humans, 80% accuracy by machine | Manual and automated detection are compared (Susceptible to wind) |
Some optical devices (digital camera, thermal camera, TreeRadarUnit (Radar 2000, Radar 900), resistograph, magnetic DNA biosensor, and near-infrared spectroscopy (NIRS)), 2020 [32] | Visual analysis, the analysis of variance (ANOVA) PROC GLM procedure, response of the leaf spectral absorbance | Not | Accuracy: visual approach 87%, Radar 2000 77%, Radar 900 73%, resistograph 73%, thermal camera 61%, digital camera 52%, and magnetic DNA 63% | All used methods are non-invasive with a detailed comparison (Accuracy needs to be further improved) |
An IoT system (commercial accelerometer sensor), 2020 [33] | FFT, the estimation of power spectral density (PSD), peaks average difference (PAD) analysis | Invasive | Observable signature of the infestation | Simple hardware with a connection to network (Low sensitivity and contrast) |
An acoustic sensor (USB microphone), 2021 [34] | Feature extraction using Mel Frequency Cepstrum Coefficient (MFCC), discrete Fourier transform (DFT), artificial neural network (ANN), Alexnet-convolutional neural networks (CNN) | Not | 99.2% accuracy | Simple hardware and concise algorithm (A plastic tube is used to imitate the real tree) |
A large-scale imaging detection method (aerial and street view), 2021 [35] | CNN, faster R-CNN ResNet-50 FPN, XResNet, | Not | Aerial and street images can be mapped to actual palm trees | Automatic large-scale detection (Limited number of infested palm tree images available online) |
An IoT system (acoustic detection of the public TreeVibes dataset), 2021 [36] | Modified mixed depthwise CNN (MixConvNet) | Invasive | 95.90% accuracy | Integration in a smartphone application with advanced algorithm (Only verified on the public TreeVibes dataset) |
An optical fiber distributed acoustic sensor (ours) | CNN | Not | Around 97.0% accuracy | Provides 24/7 monitoring on large-scale farms (Low performance at high wind speeds) |
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |
© 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Ashry, I.; Wang, B.; Mao, Y.; Sait, M.; Guo, Y.; Al-Fehaid, Y.; Al-Shawaf, A.; Ng, T.K.; Ooi, B.S. CNN–Aided Optical Fiber Distributed Acoustic Sensing for Early Detection of Red Palm Weevil: A Field Experiment. Sensors 2022, 22, 6491. https://doi.org/10.3390/s22176491
Ashry I, Wang B, Mao Y, Sait M, Guo Y, Al-Fehaid Y, Al-Shawaf A, Ng TK, Ooi BS. CNN–Aided Optical Fiber Distributed Acoustic Sensing for Early Detection of Red Palm Weevil: A Field Experiment. Sensors. 2022; 22(17):6491. https://doi.org/10.3390/s22176491
Chicago/Turabian StyleAshry, Islam, Biwei Wang, Yuan Mao, Mohammed Sait, Yujian Guo, Yousef Al-Fehaid, Abdulmoneim Al-Shawaf, Tien Khee Ng, and Boon S. Ooi. 2022. "CNN–Aided Optical Fiber Distributed Acoustic Sensing for Early Detection of Red Palm Weevil: A Field Experiment" Sensors 22, no. 17: 6491. https://doi.org/10.3390/s22176491
APA StyleAshry, I., Wang, B., Mao, Y., Sait, M., Guo, Y., Al-Fehaid, Y., Al-Shawaf, A., Ng, T. K., & Ooi, B. S. (2022). CNN–Aided Optical Fiber Distributed Acoustic Sensing for Early Detection of Red Palm Weevil: A Field Experiment. Sensors, 22(17), 6491. https://doi.org/10.3390/s22176491