The Influence of Symmetry of Synaptic Connections in MLP on the Values of Object Recognition Metrics
Keywords:
passive mine detection, magnetic field sensors, symmetric MLP (SMLP), magnetic anomalies, Adam, RMSprop, SGDAbstract
The paper considers the problem of recognizing objects (anti-tank mines, anti-personnel mines, M14, booby traps) located in soils with different structures (dry sandy, dry calcareous, dry humus, moist sandy, humid and humus, moist calcareous) using MLP with symmetric synaptic connections between neurons of different layers (SMLP) based on data from FLC100 magnetic field sensors with a sensitivity of 10-10 – 10-4 Tesla. To train the SMLP network, a dataset of fluxgate magnetometer measurements in six soil types was used, divided into training (80%) and test (20%) subsets. The input data is encoded as a three-dimensional vector (voltage, height of the sensor above the ground with a mine, one-hot encoding of 6 soil types). The main difference between SMLP and MLP is the imposition of symmetric constraints on the weight matrix of the second hidden layer, which almost halves the number of synaptic connections between neurons in the hidden layers without significantly degrading recognition quality. The proposed symmetrization mechanism mitigates over-parameterization and overfitting in small-scale magnetometric datasets by introducing structural inductive bias and implicit regularization through symmetry constraints in parameter space. In contrast to convolutional architectures developed for high-dimensional spatial representations, the proposed approach is tailored to structured low-dimensional magnetometric measurements. The MLP and SMLP models were trained for 100 epochs with Adam, RMSprop, and SGD optimizers with a learning rate reduction using ReduceLROnPlateau callback. Experimental results show that SMLP achieves an average accuracy of 99.20% and AUC = 0.9996, which is only 0.09% lower than the traditional MLP (99.29%, AUC = 0.9997), but reduces the training time by 20–30% and behaves more stably under different optimizers. Thus, the SMLP model is productive for use on embedded devices with limited resources.
References
Landmine Monitor Report, The Monitor, 2024. [Online]. Available at: https://the-monitor.org/reports/landmine-monitor-2024.
M. Šimić, D. Ambruš, and V. Bilas, “Landmine identification from pulse induction metal detector data using machine learning,” IEEE Sensors Letters, vol. 7, pp. 1–4, 2023, https://doi.org/10.1109/LSENS.2023.3307091.
H. Liu, C. Zhao, J. Zhu, J. Ge, H. Dong, Z. Liu, and N. Mrad, “Active detection of small UXO-like targets through measuring electromagnetic responses with a magneto-inductive sensor array,” IEEE Sensors Journal, vol. 21, pp. 23558–23567, 2021, https://doi.org/10.1109/JSEN.2021.3107842.
M. Munschy, D. Boulanger, P. Ulrich, and M. Bouiflane, “Magnetic mapping for the detection and characterization of UXO: Use of multi-sensor fluxgate 3-axis magnetometers and methods of interpretation,” Journal of Applied Geophysics, vol. 61, pp. 168–183, 2006, https://doi.org/10.1016/j.jappgeo.2006.06.004.
A. Barnawi, K. Kumar, N. Kumar, B. Alzahrani, and A. Almansour, “A deep learning approach for landmines detection based on airborne magnetometry imaging and edge computing,” Computer Modeling in Engineering & Sciences, vol. 139, pp. 2117–2137, 2023, https://doi.org/10.32604/cmes.2023.044184.
D. Jafuno, A. Mian, G. Ginolhac, and N. Stelzenmuller, “Classification of buried objects from ground penetrating radar images using second order deep learning models,” IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 18, pp. 3185–3197, 2025, https://doi.org/10.1109/JSTARS.2024.3524424.
C. Yılmaz, Y. Sönmez, H. T. Kahraman, S. Soyler, and U. Güvenç, “Developing of decision support system for land mine classification by meta-heuristic classifier,” Proceedings of the International Symposium on Innovations in Intelligent Systems and Applications (INISTA), Sinaia, Romania, 2016, pp. 1–5, https://doi.org/10.1109/INISTA.2016.7571867.
J. Amarel, C. Rudolf, A. Iliopoulos, J. Michopoulos, and L. N. Smith, “Symmetry constrained neural networks for detection and localization of damage in metal plates,” arXiv preprint, 2024, https://doi.org/10.1063/5.0242345.
S. N. M. Kanafiah et al., “Metal shape classification of buried object using multilayer perceptron neural network in GPR data,” IOP Conf. Series: Materials Science and Engineering, vol. 705, Art. no. 012028, 2019, https://doi.org/10.1088/1757-899X/705/1/012028.
Y. Wang, X. Ma, Z. Chen, Y. Luo, J. Yi, and J. Bailey, “Symmetric cross entropy for robust learning with noisy labels,” Proceedings of the IEEE/CVF Int. Conf. Computer Vision (ICCV), 2019, pp. 322–330, https://doi.org/10.1109/ICCV.2019.00041.
J. Vyhnanek, M. Janoseka, and P. Ripka, “AMR gradiometer for mine detection and sensing,” Procedia Engineering, vol. 25, pp. 362–366, 2012, https://doi.org/10.1016/j.proeng.2011.12.089.
Z. Wen, S. Han, C. Gao, Y. Chen, L. Guo, and Y. Zhang, “A deep learning method for recognizing types of unexploded ordnance based on magnetic detection,” IEEE Transactions on Geoscience and Remote Sensing, vol. 62, pp. 1–13, 2024, Art. no. 5918313, https://doi.org/10.1109/TGRS.2024.3405478.
S. A. Stankevich and I. Y. Saprykin, “Optical and magnetometric data integration for landmine detection with UAV,” WSEAS Transactions on Environment and Development, vol. 20, pp. 1059–1066, 2024, https://doi.org/10.37394/232015.2024.20.96.
I. Poliachenko, V. Kozak, V. Bakhmutov, S. Cherkes, and T. Bilyi, “Exploratory study of potential applications of UAV magnetic surveys for unexploded ordnance detection in coastline zone,” Geofizicheskiy Zhurnal, vol. 47, no. 1, 2025, https://doi.org/10.24028/gj.v47i1.316760.
X. Zhou, Z. Chen, H. Chen, S. Wang, and Z. O. Kubeka, “Edge detection of source body from magnetic anomaly based on ResNet,” Remote Sensing, vol. 16, Art. no. 4139, 2024, https://doi.org/10.3390/rs16224139.
M. F. Işık, Ç. Suiçmez, and C. Yılmaz, “Low-cost mine detector design using magnetic anomaly method,” Politeknik Dergisi, vol. 25, no. 4, pp. 1729–1740, 2022. https://doi.org/10.2339/politeknik.1080410.
H. Huang and I. J. Won, “Characterization of UXO-like targets using broadband electromagnetic induction sensors,” IEEE Transactions on Geoscience and Remote Sensing, vol. 41, no. 3, pp. 652–663, 2003, https://doi.org/10.1109/TGRS.2003.809936.
H. Tanaka and D. Kunin, “Noether’s learning dynamics: Role of symmetry breaking in neural networks,” Proceedings of the 40th International Confereence on Machine Learning (ICML), 2023.
X. S. Hu, S. Zagoruyko, and N. Komodakis, “Exploring weight symmetry in deep neural networks,” Computer Vision and Image Understanding, vol. 187, 2019, https://doi.org/10.1016/j.cviu.2019.07.006.
G. Zhou and S. Reichle, “UAV-based multi-sensor data fusion processing,” International Journal of Image and Data Fusion, vol. 1, no. 3, pp. 283–291, 2010, https://doi.org/10.1080/19479832.2010.497343.
B. Rusyn, O. Lutsyk, R. Kosarevych et al., “Rethinking deep CNN training: A novel approach for quality-aware dataset optimization,” IEEE Access, vol. 12, pp. 137427–137438, 2024, https://doi.org/10.1109/ACCESS.2024.3414651.
B. Rusyn, O. Lutsyk, R. Kosarevych et al., “Features extraction from spectral remote sensing images based on multi-threshold binarization,” Scientific Reports, vol. 13, no. 1, Art. no. 19655, 2023, https://doi.org/10.1038/s41598-023-46785-7.
K. D. G. Maduranga, V. Zadorozhnyy, and Q. Ye, “Symmetry-structured convolutional neural networks,” Neural Computing and Applications, vol. 35, pp. 4421–4434, 2022, https://doi.org/10.1007/s00521-022-08168-3.
K. D. G. Maduranga, V. Zadorozhnyy, and Q. Ye, “Symmetry-structured convolutional neural networks,” Neural Computing and Applications, vol. 35, pp. 4421–4434, 2023, https://doi.org/10.1007/s00521-022-08168-3.
R. Peleshchak, V. Lytvyn, O. Mediakov, and I. Peleshchak, “Morphology of convolutional neural network with diagonalized pooling,” in Communications in Computer and Information Science, 2023, pp. 161–172, https://doi.org/10.1007/978-3-031-27034-5_11.
X. S. Hu, S. Zagoruyko, and N. Komodakis, “Exploring weight symmetry in deep neural networks,” arXiv preprint, 2018, doi: 10.48550/arXiv.1812.11027.
R. Agrawal and J. Pradeep, “Land mines detection,” Kaggle Dataset, 2020. [Online]. Available at: https://www.kaggle.com/datasets/ritwikb3/land-mines-detection.
Downloads
Published
How to Cite
Issue
Section
License
International Journal of Computing is an open access journal. Authors who publish with this journal agree to the following terms:• Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
• Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
• Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work.
