Abstract
Accurate soil type classification is essential for sustainable agriculture, environmental monitoring, and effective land management. Due to the inherent variability in soil types and the prevalent class imbalance in soil imaging datasets, reliable classification remains a complex challenge. This study combined four unique soil image datasets, yielding 11 distinct soil type classes. The MobileNet model was modified to enhance classification with essential adjustments: replacing ReLU with Leaky ReLU, incorporating an average pooling layer (APL), and integrating a Long Short-Term Memory (LSTM) Network. Since class imbalance was evident in the extracted features, the Synthetic Minority Oversampling Technique (SMOTE) was employed to balanced these features. The MobileNet-LSTM-SMOTE approach demonstrated strong performance, achieving a Kappa score of 98.6% and a 5-fold average Kappa score of 97.9%. Additionally, testing on an independent dataset confirmed the model’s robustness and generalizability. We also conducted a comprehensive performance comparison, including statistical analysis, against five established models—MobileNet, Xception, ResNet50V2, VGG-19, and DenseNet121. The results underscore the efficacy of proposed method in overcoming the challenges of class imbalance and feature variability, offering a reliable solution for soil type classification.
Similar content being viewed by others
Data availability
No datasets were generated or analysed during the current study.
Dataset-1: https://www.kaggle.com/datasets/jayaprakashpondy/soil-image-dataset
Dataset-2:https://www.kaggle.com/datasets/prasanshasatpathy/soil-types
Dataset-3: https://www.kaggle.com/datasets/matshidiso/soil-types
Dataset-4:https://www.kaggle.com/datasets/faisalkhaan/soil-image-classification
References
Abdulraheem MI, Zhang W, Li S, Moshayedi AJ, Farooque AA, Hu J (2023) Advancement of remote sensing for soil measurements and applications: a comprehensive review. Sustainability 15(21):15444. https://doi.org/10.3390/su152115444
Abeje BT, Salau AO, Gela BM, Mengistu AD (2024) Soil type identification model using a hybrid computer vision and machine learning approach. Multimedia Tools Appl 83(1):575–589. https://doi.org/10.1007/s11042-023-15692-4
Araújo SO, Peres RS, Ramalho JC, Lidon F, Barata J (2023) Machine learning applications in agriculture: current trends, challenges, and future perspectives. Agronomy 13(12):2976. https://doi.org/10.3390/agronomy13122976
Azizi A, Gilandeh YA, Mesri-Gundoshmian T, Saleh-Bigdeli AA, Moghaddam HA (2020) Classification of soil aggregates: a novel approach based on deep learning. Soil Tillage Res 199:104586. https://doi.org/10.1016/j.still.2020.104586
Barman U, Choudhury RD (2020) Soil texture classification using multi class support vector machine. Inform Process Agric 7(2):318–332. https://doi.org/10.1016/j.inpa.2019.08.001
Behrens T, Schmidt K, MacMillan RA, Viscarra Rossel RA (2018) Multi-scale digital soil mapping with deep learning. Sci Rep 8(1):15244. https://doi.org/10.1038/s41598-018-33516-6
Bhat SA, Hussain I, Huang N-F (2023) Soil suitability classification for crop selection in precision agriculture using GBRT-based hybrid DNN surrogate models. Ecol Inf 75:102109. https://doi.org/10.1016/j.ecoinf.2023.102109
Bhattacharya B, Solomatine DP (2006) Machine learning in soil classification. Neural Netw 19(2):186–195. https://doi.org/10.1016/j.neunet.2006.01.005
Cabras S, Nueda MEC (2024) Transfer learning in multiple hypothesis testing. Entropy 26:49. https://doi.org/10.3390/e26010049
Duarte IMR, Rodrigues CMG, Pinho AB (2018) Classification of soils. In: Bobrowsky PT, Marker M (eds) Encyclopedia of engineering geology. Springer International Publishing, pp 125–133. https://doi.org/10.1007/978-3-319-73568-9_52
Elaalem MM, Zurqani HA, Ben Mahmoud KR, Elhawej AR (2021) Soil classification and properties. In Zurqani HA (eds) The soils of Libya. World Soils Book Series. Springer, Cham. https://doi.org/10.1007/978-3-030-66368-1_4
Elreedy D, Atiya AF (2019) A comprehensive analysis of synthetic minority oversampling technique (smote) for handling class imbalance. Inf Sci 505:32–64. https://doi.org/10.1016/j.ins.2019.07.070
Fan Z, Herrick JE, Saltzman R, Matteis C, Yudina A, Nocella N, Crawford E, Parker R, Van Zee J (2017) Measurement of soil color: a comparison between smartphone camera and the munsell color charts. Soil Sci Soc Am J 81(5):1139–1146. https://doi.org/10.2136/sssaj2017.01.0009
Haq YU, Shahbaz M, Asif S, Ouahada K, Hamam H (2023) Identification of soil types and salinity using modis terra data and machine learning techniques in multiple regions of Pakistan. Sensors 23(19):8121. https://doi.org/10.3390/s23198121
He H, Bai Y, Garcia EA, Li S (2008) ADASYN: Adaptive synthetic sampling approach for imbalanced learning. In: 2008 IEEE International Joint Conference on Neural Networks (IEEE World Congress on Computational Intelligence), pp 1322–1328
Howard AG, Zhu M, Chen B, Kalenichenko D, Wang W, Weyand T, Andreetto M, Adam H (2017) Mobilenets: efficient convolutional neural networks for mobile vision applications. https://doi.org/10.48550/ARXIV.1704.04861
Kabała C, Charzyński P, Chodorowski J, Drewnik M, Glina B, Greinert A, Waroszewski J (2019) Polish soil classification: principles, classification scheme and correlations. Soil Sci Ann 70(2)
Kullu A, Mohanty A, Prasad B, Rajasmita R, Dehury PP, Pani P, Paital B (2021) Soil type as one of the Major contributing factors for top ten Agri-Producing States of India. J ISSN 2766:2276
Laghrissi F, Douzi S, Douzi K, Hssina B (2021) Intrusion detection systems using long short-term memory (lstm). J Big Data 8(1):65. https://doi.org/10.1186/s40537-021-00448-4
Lanjewar MG, Gurav OL (2022) Convolutional Neural Networks based classifications of soil images. Multimedia Tools Appl 81(7):10313–10336. https://doi.org/10.1007/s11042-022-12200-y
Lanjewar MG, Morajkar P, Payaswini P (2023) Modified transfer learning frameworks to identify potato leaf diseases. Multimed Tools Appl 83:50401–50423. https://doi.org/10.1007/s11042-023-17610-0
Lanjewar MG, Panchbhai KG, Patle LB (2024a) Fusion of transfer learning models with LSTM for detection of breast cancer using ultrasound images. Comput Biol Med 169:107914. https://doi.org/10.1016/j.compbiomed.2023.107914
Lanjewar MG, Panchbhai KG, Patle LB (2024b) Sugar detection in adulterated honey using hyper-spectral imaging with stacking generalization method. Food Chem 450:139322. https://doi.org/10.1016/j.foodchem.2024.139322
Lu Y, Perez D, Dao M, Kwan C, Li J (2018) Deep learning with synthetic hyperspectral images for improved soil detection in multispectral imagery. In: 2018 9th IEEE Annual Ubiquitous Computing, Electronics & Mobile Communication Conference (UEMCON), pp 666–672. https://doi.org/10.1109/UEMCON.2018.8796838
Maniatopoulos A, Mitianoudis N (2021) Learnable leaky relu (Lelelu): an alternative accuracy-optimized activation function. Information 12(12):513. https://doi.org/10.3390/info12120513
Nahm FS (2022) Receiver operating characteristic curve: overview and practical use for clinicians. Korean J Anesthesiology 75(1):25–36. https://doi.org/10.4097/kja.21209
Ozdemir C (2024) Neural Comput Applic 36:6257–6270. https://doi.org/10.1007/s00521-024-09484-6. Adapting transfer learning models to dataset through pruning and Avg-TopK pooling
Ozdemir C, Dogan Y (2024) Advancing brain tumor classification through MTAP model: an innovative approach in medical diagnostics. Med Biol Eng Comput 62:2165–2176. https://doi.org/10.1007/s11517-024-03064-5
Ozdemir C, Dogan Y, Kaya Y (2024) RGB-Angle-Wheel: a new data augmentation method for deep learning models. Knowl Based Syst 291:111615. https://doi.org/10.1016/j.knosys.2024.111615
Özdemir C (2023) Avg-topk: a new pooling method for convolutional neural networks. Expert Syst Appl 223:119892. https://doi.org/10.1016/j.eswa.2023.119892
Pan Y, Pi D, Khan IA, Khan ZU, Chen J, Meng H (2021) DenseNetFuse: a study of deep unsupervised DenseNet to infrared and visual image fusion. J Ambient Intell Humaniz Comput 12(11):10339–10351. https://doi.org/10.1007/s12652-020-02820-3
Panchbhai KG, Lanjewar MG (2024) Enhancement of tea leaf diseases identification using modified SOTA models. Neural Comput Appl. https://doi.org/10.1007/s00521-024-10758-2
Panchbhai KG, Lanjewar MG, Malik VV, Charanarur P (2024) Small size CNN (CAS-CNN), and modified MobileNetV2 (CAS-MODMOBNET) to identify cashew nut and fruit diseases. Multimed Tools Appl. https://doi.org/10.1007/s11042-024-19042-w
Pham BT, Nguyen MD, Nguyen-Thoi T, Ho LS, Koopialipoor M, Quoc NK, Van Le H (2021) A novel approach for classification of soils based on laboratory tests using Adaboost, Tree and ANN modeling. Transp Geotechnics 27:100508
Politi M (2022) Paired t-test to evaluate machine learning classifiers using python. In: Medium. https://towardsdatascience.com/paired-t-test-to-evaluate-machine-learning-classifiers-1f395a6c93fa. Accessed 27 Oct 2024
Rahman SAZ, Mitra C, K., Mohidul Islam SM (2018) Soil classification using machine learning methods and crop suggestion based on soil series. In: 2018 21st International Conference of Computer and Information Technology (ICCIT), pp 1–4. https://doi.org/10.1109/ICCITECHN.2018.8631943
Raschka S (2022, April 25). Creating confidence intervals for machine learning classifiers. Sebastian Raschka, PhD. https://sebastianraschka.com/blog/2022/confidence-intervals-for-ml.html
Raschka S (2023) paired_ttest_5x2cv: 5x2cv paired t test for classifier comparisons - mlxtend. https://rasbt.github.io/mlxtend/user_guide/evaluate/paired_ttest_5x2cv/. Accessed 27 Oct 2024
Rodrigues PMS, Schaefer CEGR, Silva JDO, Júnior F, Santos WGD, R. M., Neri AV (2016) The influence of soil on vegetation structure and plant diversity in different tropical savannic and forest habitats. J Plant Ecol rtw135. https://doi.org/10.1093/jpe/rtw135
Roopashree S, Anitha J, Challa S, Mahesh TR, Venkatesan VK, Guluwadi S (2024) Mapping of soil suitability for medicinal plants using machine learning methods. Sci Rep 14(1):3741
Sarker IH (2021) Deep learning: a comprehensive overview on techniques, taxonomy, applications and research directions. SN Comput Sci 2(6):420. https://doi.org/10.1007/s42979-021-00815-1
Soil types. (n.d.-a). Retrieved March 20, 2024 from https://www.kaggle.com/datasets/prasanshasatpathy/soil-types
Soil types. (n.d.-b). Retrieved March 20, 2024 from https://www.kaggle.com/datasets/matshidiso/soil-types
Soil image dataset. (n.d.). Retrieved March 20, 2024 from https://www.kaggle.com/datasets/jayaprakashpondy/soil-image-dataset
Soil Image Classification. https://www.kaggle.com/datasets/faisalkhaan/soil-image-classification. Accessed 27 Oct 2024
Tahat M, Alananbeh MM, Othman KA, Y., Leskovar I, D (2020) Soil health and sustainable agriculture. Sustainability 12(12):4859. https://doi.org/10.3390/su12124859
Uddin M, Hassan MR (2022) A novel feature based algorithm for soil type classification. Complex Intell Syst 8(4):3377–3393. https://doi.org/10.1007/s40747-022-00682-0
Vijayakumar V, Balakrishnan N (2021) RETRACTED ARTICLE: Artificial intelligence-based agriculture automated monitoring systems using WSN. J Ambient Intell Humaniz Comput 12(7):8009–8016. https://doi.org/10.1007/s12652-020-02530-w
Wong T-T, Yeh P-Y (2020) Reliable accuracy estimates from k -fold cross validation. IEEE Trans Knowl Data Eng 32:1586–1594. https://doi.org/10.1109/TKDE.2019.2912815
Zhao Z, Chow TL, Rees HW, Yang Q, Xing Z, Meng F-R (2009) Predict soil texture distributions using an artificial neural network model. Comput Electron Agric 65(1):36–48. https://doi.org/10.1016/j.compag.2008.07.008
Funding
Not applicable.
Author information
Authors and Affiliations
Contributions
Madhusudan Lanjewar: Methodology, Software, Validation, Formal analysis, and Investigation. Kamini Panchbhai: Conceptualization, Writing Original Draft, and Visualization. Aditi Venkatesh Naik: Conceptualization and Methodology.
Corresponding author
Ethics declarations
Ethical and informed consent
Not applicable.
Competing interests
The authors declare no competing interests.
Conflict of interest
The authors declared that they have no conflicts of interest to this work. We declare that we do not have any commercial or associative interest that represents a conflict of interest in connection with the work submitted.
Additional information
Communicated by Hassan Babaie.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Panchbhai, K.G., Lanjewar, M.G. & Naik, A.V. Modified MobileNet with leaky ReLU and LSTM with balancing technique to classify the soil types. Earth Sci Inform 18, 77 (2025). https://doi.org/10.1007/s12145-024-01521-1
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s12145-024-01521-1