The week's pick
Reseach Article
Classifying Buried Objects by Combining CNN with Fourier Transform
| International Journal of Computer Applications |
| Foundation of Computer Science (FCS), NY, USA |
| Volume 185 - Number 34 |
| Year of Publication: 2023 |
| Authors: Ahmed M.D.E. Hassanein, Mostafa M.M. Aboalamayem |
10.5120/ijca2023923118
|
Ahmed M.D.E. Hassanein, Mostafa M.M. Aboalamayem . Classifying Buried Objects by Combining CNN with Fourier Transform. International Journal of Computer Applications. 185, 34 ( Sep 2023), 17-22. DOI=10.5120/ijca2023923118
Abstract
Identifying buried objects needs signal processing which requires high computational power. In this paper, the frequency domain and time domain signals are used in combination with Convolutional Neural Network to classify the buried objects represented by the signals. The Fourier transform is used to change from the frequency domain to the time domain as it gives a holistic view of the environment around the antennas. The proposed method doesn’t require to go into any complex signal processing steps which can save a lot of effort. A Bow Tie antenna is used to take measurements of five different setups. One setup is when the environment between the two antennas is filled with air, another is when it is filled with sand and another is when the sand contains buried rod at depth 3, 5 and 10cm. Each setup represents a class. The resulting measurements are in the frequency domain. Fourier transform is used to obtain those measurements in the time domain. The aim is to use a proposed CNN network to classify the frequency domain measurements to their corresponding classes. Another aim is to see the effect of using the frequency domain signals along with the time domain signals to classify them. The frequency domain achieved 87% accuracy in classifying the signals. While, the frequency domain and time domain signals achieved 93% accuracy.
References
- Nguyen, Van & Zwanenburg, Ewout & Limmer, Steffen & Luijben, Wessel & Bäck, Thomas & Olhofer, Markus. (2021). A Combination of Fourier Transform and Machine Learning for Fault Detection and Diagnosis of Induction Motors. 344-351. doi: 10.1109/DSA52907.2021.00053.
- Lafta, Raid & Zhang, Ji & Tao, Xiaohui & Li, Yan & Zhu, Xiaodong & Luo, Yonglong & Chen, Fulong. (2017). Coupling a Fast Fourier Transformation with a Machine Learning Ensemble Model to Support Recommendations for Heart Disease Patients in a Telehealth Environment. IEEE Access. PP. 1-1. 10.1109/ACCESS.2017.2706318.
- Yi, Kun; Zhang, Qi; Wang, Shoujin; He, Hui; Long, Guodong; Niu, Zhendong. Neural Time Series Analysis with Fourier Transform: A Survey. Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI). https://doi.org/10.48550/arXiv.2302.02173
- Mehrabkhani, Soheil. (2022). Fourier Transform Approach to Machine Learning III: Fourier Classification. https://doi.org/10.48550/arXiv.2001.06081
- Mehrabkhani, Soheil. (2019). Fourier Transform Approach to Machine Learning I: Fourier Regression. https://doi.org/10.48550/arXiv.1904.00368
- Anaya-Isaza, Andrés; Zequera-Diaz, Martha. (2022) Fourier transform-based data augmentation in deep learning for diabetic foot thermograph classification. Biocybernetics and Biomedical Engineering, Volume 42, Issue 2, Pages 437-452, ISSN 0208-5216, https://doi.org/10.1016/j.bbe.2022.03.001.
- Yaghoobian, O. and Amindavar, H. (2020) Detection of Underground Buried Objects Using Chirplet Transform. 28th Iranian Conference on Electrical Engineering (ICEE), Tabriz, Iran, pp. 1-5, doi: 10.1109/ICEE50131.2020.9260633.
- Pham, Minh-Tan & Lefèvre, Sébastien. (2018). Buried Object Detection from B-Scan Ground Penetrating Radar Data Using Faster-RCNN. 6804-6807. 10.1109/IGARSS.2018.8517683.
- Yurt, R., Torpi, H., Kizilay, A. et al. Buried object characterization by data-driven surrogates and regression-enabled hyperbolic signature extraction. Sci Rep 13, 5717 (2023). https://doi.org/10.1038/s41598-023-32925-6
- Al-Nuaimy, W.; Huang, Y.; Nakhkash, M.; Fang, M.T.C.; Nguyen, V.T.; Eriksen, A. (2000) Automatic detection of buried utilities and solid objects with GPR using neural networks and pattern recognition, Journal of Applied Geophysics, Volume 43, Issues 2–4, Pages 157-165, ISSN 0926-9851, https://doi.org/10.1016/S0926-9851(99)00055-5.
- Huyen, Nguyen Thi; Ha, Duong Duc; Hiep, Pham Thanh. (2020) Buried objects detection in heterogeneous environment using UWB systems combined with curve fitting method, ICT Express, Volume 6, Issue 4, Pages 348-352, ISSN 2405-9595, https://doi.org/10.1016/j.icte.2020.06.006.
- Aboalamayem, M. M. M. (2023). Ground penetrating radar antenna, Graduation Thesis.
- Kingma, D. P.; Ba., Jimmy (2015) Adam: A Method for Stochastic Optimization, Conference paper at the 3rd International Conference for Learning Representations, San Diego, 2015. https://doi.org/10.48550/arXiv.1412.6980
- Yarlagadda, R. K. R. (2010). Fourier Transform Analysis, Analog and Digital Signals and Systems, pp.109-153. DOI:10.1007/978-1-4419-0034-0 4.
- Glorot, Xavier & Bengio, Y. (2010). Understanding the difficulty of training deep feedforward neural networks. Journal of Machine Learning Research - Proceedings Track. 9. 249-256.
- Gomes, Dipta & Saif, Saifuddin. (2021). Robust Underwater Fish Detection Using an Enhanced Convolutional Neural Network. International Journal of Image Graphics and Signal Processing. 13. 44-54. 10.5815/ijigsp.2021.03.04.
- Maxwell, Aaron & Warner, Timothy & Guillén, Luis Andrés. (2021). Accuracy Assessment in Convolutional Neural Network-Based Deep Learning Remote Sensing Studies—Part 1: Literature Review. Remote Sensing. 13. 2450. 10.3390/rs13132450.
Index Terms
Keywords