Abstract
Purpose
Soil classification is important in the field of geotechnical engineering. Soil types are usually defined by combinations of soil properties, which are interrelated and interactive. This means that distinguishing between soil types is laborious and uncertain. Instead, images are routinary and widespread. Thus, this study presents a framework using convolutional neural networks (CNN) to determine soil types from soil images. Besides, the image acquisition distance factors are incorporated and evaluated in the framework.
Methods
The properties of color and texture were collocated to define eight types of soil. To collect images effectively, an image acquisition method was designed. Then, the images of eight types of soil were collected at eight acquisition distances (10 to 80 cm). Two types of models including single-range scale model and the multirange scale model were trained and evaluated as per the framework.
Results
The accuracy of the single-range scale models was between 19 and 98%. In addition, the multirange scale models achieved an accuracy range between 51 and 98%. Moreover, the mean uncertainty of the former was between 0.11 and 0.16, and the latter was between 0.05 and 0.12.
Conclusion
The models can effectively infer soil types from images and improve robustness through multi-distance training as per the proposed framework. The property of color has a higher priority of the classification than the texture. Moreover, image-based soil classification is extremely sensitive to distance factors, and the perceptual distance of 70 cm was shown to be the better one among the eight selected distances.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Data availability
The data that support the findings of this study are available from the corresponding author (Zhan Shu), upon reasonable request.
References
AbdulAzeem Y, Bahgat WM, Badawy M (2021) A CNN based framework for classification of Alzheimer’s disease. Neural Comput Appl 33(16):10415–10428. https://doi.org/10.1007/S00521-021-05799-W/TABLES/10
Abraham S, Huynh C, Vu H (2019) Classification of soils into hydrologic groups using machine learning. Data 5(1):2. https://doi.org/10.3390/data5010002
Ajdadi FR, Gilandeh YA, Mollazade K, Hasanzadeh RPR (2016) Application of machine vision for classification of soil aggregate size. Soil Tillage Res 162:8–17. https://doi.org/10.1016/j.still.2016.04.012
Arnay R, Hernández-Aceituno J, Mallol C (2021) Soil micromorphological image classification using deep learning: the porosity parameter. Appl Soft Comput 102:107093. https://doi.org/10.1016/j.asoc.2021.107093
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. Inf Process Agric 7(2):318–332. https://doi.org/10.1016/j.inpa.2019.08.001
Chen X, Spencer BF, LI J, Guan Z, Pang Y (2023) Optimization of distribution patterns of link beams in a double-column tall pier bent subjected to earthquake excitations. Earthq Eng Struct Dyn 52(3):641–659. https://doi.org/10.1002/eqe.3778
Chung SO, Cho KH, Kong JW, Sudduth KA, Jung KY (2010) Soil texture classification algorithm using RGB characteristics of soil images. IFAC Proc 43(26):34–38. https://doi.org/10.3182/20101206-3-JP-3009.00005
Croffie MET, Williams PN, Fenton O, Fenelon A, Daly K (2021) Rubidium measured by XRF as a predictor of soil particle size in limestone and siliceous parent materials. J Soils Sediments 22:818–830. https://doi.org/10.1007/S11368-021-03109-4
de Oliveira Morais PA, de Souza DM, de Melo Carvalho MT, Madari BE, de Oliveira AE (2019) Predicting soil texture using image analysis. Microchem J 146:455–463. https://doi.org/10.1016/j.microc.2019.01.009
Dornik A, Drăguţ L, Urdea P (2018) Classification of soil types using geographic object-based image analysis and random forests. Pedosphere 28(6):913–925. https://doi.org/10.1016/S1002-0160(17)60377-1
Esfandiarpour-Boroujeni I, Shahini Shamsabadi M, Shirani H, Mosleh Z, Bagheri Bodaghabadi M, Salehi MH (2020) Comparison of error and uncertainty of decision tree and learning vector quantization models for predicting soil classes in areas with low altitude variations. Catena 191:104581. https://doi.org/10.1016/j.catena.2020.104581
Gawlikowski J, Tassi CRN, Ali M, Lee J, Humt M, Feng J, Kruspe A, Triebel R, Jung P, Roscher R, Shahzad M, Yang W, Bamler R, Zhu XX (2022) A survey of uncertainty in deep neural networks. arXiv. https://arxiv.org/abs/2107.03342
Hahn C, Gloaguen R (2008) Estimation of soil types by non linear analysis of remote sensing data. Nonlin Processes Geophys 15(1):115–126. https://doi.org/10.5194/npg-15-115-2008
Han J, Moraga C, Sinne S (1996) Optimization of feedforward neural networks. Eng Appl Artif Intell 9(2):109–119. https://doi.org/10.1016/0952-1976(95)00001-1
Han P, Dong D, Zhao X, Jiao L, Lang Y (2016) A smartphone-based soil color sensor: for soil type classification. Comput Electron Agric 123:232–241. https://doi.org/10.1016/j.compag.2016.02.024
He K, Zhang X, Ren S, Sun J (2016) Deep residual learning for image recognition. arXiv. https://arxiv.org/abs/1512.03385
Howard A, Sandler M, Chu G, Chen LC, Chen B, Tan M, Wang W, Zhu Y, Pang R, Vasudevan V, Le Q, Adam H (2019) Searching for MobileNetV3. arXiv. https://arxiv.org/abs/1905.02244
Inazumi S, Intui S, Jotisankasa A, Chaiprakaikeow S, Kojima K (2020) Artificial intelligence system for supporting soil classification. Results Eng 8:100188. https://doi.org/10.1016/j.rineng.2020.100188
Kim B, Yuvaraj N, Sri Preethaa KR, Arun Pandian R (2021) Surface crack detection using deep learning with shallow CNN architecture for enhanced computation. Neural Comput Appl 33(15):9289–9305. https://doi.org/10.1007/S00521-021-05690-8/TABLES/6
Kim Y, Yun TS (2021) How to classify sand types: a deep learning approach. Eng Geol 288:106142. https://doi.org/10.1016/j.enggeo.2021.106142
Kingma DP, Ba JL (2017) Adam: a method for stochastic optimization. arXiv. https://arxiv.org/abs/1412.6980
Koenderink JJ (1984) The structure of images. Biol Cybern 50(5):363–370. https://doi.org/10.1007/BF00336961
Krüger F (2016) Activity, context, and plan recognition with computational causal behavior models. Ph.D. Thesis, University of Rostock, Mecklenburg, Germany. https://pdfs.semanticscholar.org/bebf/183d2f57f79b5b3e85014a9e1d6392ad0e5c.pdf. Accessed 12 Mar 2022
Li N, Ni H, Sheng Y, Zhao Z (2019a) Gradient adapter for hard-threshold deep neural networks. Int J Innov Comput Inf Control 15(3):1023–1035. https://doi.org/10.24507/IJICIC.15.03.1023
Li Z, Chen F, He M, Long W, Ou J, Li M (2023) Experimental investigation on self-centering steel-timber hybrid beam-column connections. J Struct Eng 149(3):04022256. https://doi.org/10.1061/JSENDH.STENG-11570
Li Z, Tao D, Li M, Shu Z, Jing S, He M, Qi P (2019b) Prediction of damage accumulation effect of wood structural members under long-term service: a machine learning approach. Materials 12(8):1243. https://doi.org/10.3390/MA12081243
Lindeberg T (1994) Scale-space theory: a basic tool for analyzing structures at different scales. J Appl Statistics 21(1–2):225–270. https://doi.org/10.1080/757582976
Liu C, Wang W, Wang M, Lv F, Konan M (2017) An efficient instance selection algorithm to reconstruct training set for support vector machine. Knowl Based Syst 116:58–73. https://doi.org/10.1016/j.knosys.2016.10.031
Ma N, Zhang X, Zheng HT, Sun J (2018) ShuffleNet V2: practical guidelines for efficient CNN architecture design. arXiv. https://arxiv.org/abs/1807.11164
Ma R, Hu F, Liu J, Zhao S (2021) Evaluating the effect of soil internal forces on the stability of natural soil aggregates during vegetation restoration. J Soils Sediments 21(9):3034–3043. https://doi.org/10.1007/S11368-021-03011-Z
Maleki S, Khormali F, Mohammadi J, Bogaert P, Bagheri Bodaghabadi M (2020) Effect of the accuracy of topographic data on improving digital soil mapping predictions with limited soil data: an application to the Iranian loess plateau. Catena 195: 104810. https://doi.org/10.1016/j.catena.2020.104810
Marino S, Beauseroy P, Smolarz A (2019) Weakly-supervised learning approach for potato defects segmentation. Eng Appl Artif Intell 85:337–346. https://doi.org/10.1016/j.engappai.2019.06.024
Ng W, Minasny B, McBratney A (2020) Convolutional neural network for soil microplastic contamination screening using infrared spectroscopy. Sci Total Environ 702(1):134723. https://doi.org/10.1016/j.scitotenv.2019.134723
Olukan TA, Chiou YC, Chiu CH, Lai CY, Santos S, Chiesa M (2020) Predicting the suitability of lateritic soil type for low cost sustainable housing with image recognition and machine learning techniques. J Build Eng 29:101175. https://doi.org/10.1016/j.jobe.2020.101175
Pan H, Liang J, Zhao Y, Li F (2020) Facing the 3rd national land survey (cultivated land quality): soil survey application for soil texture detection based on the high-definition field soil images by using perceptual hashing algorithm (pHash). J Soils Sediments 20(9):3427–3441. https://doi.org/10.1007/S11368-020-02657-5
Pegalajar MC, Ruiz LGB, Sánchez-Marañón M, Mansilla L (2020) A munsell colour-based approach for soil classification using fuzzy logic and artificial neural networks. Fuzzy Sets Syst 401:38–54. https://doi.org/10.1016/J.FSS.2019.11.002
Ping Y, Fang C, Shi F, Wu H, Yam MCH, Chen Y (2022) Experimental and numerical studies on SMA-viscoelastic hybrid self-centering braces. Smart Mater Struct 31(9):095048. https://doi.org/10.1088/1361-665X/ac8849
Shu Z, Ning B, Chen J, Li Z, He M, Luo J, Dong H (2023) Reinforced moment-resisting glulam bolted connection with coupled long steel rod with screwheads for modern timber frame structures. Earthq Eng Struct Dyn. https://doi.org/10.1002/eqe.3789
Shukla G, Garg RD, Srivastava HS, Garg PK (2018) An effective implementation and assessment of a random forest classifier as a soil spatial predictive model. Int J Remote Sens 39(8):2637–2669. https://doi.org/10.1080/01431161.2018.1430399
Simonyan K, Zisserman A (2014) Very deep convolutional networks for large-scale image recognition. arXiv. https://arxiv.org/abs/1409.1556
Singh P, Raj P, Namboodiri VP (2020) EDS pooling layer. Imag Vis Comput 98:103923. https://doi.org/10.1016/J.IMAVIS.2020.103923
Srivastava P, Shukla A, Bansal A (2021) A comprehensive review on soil classification using deep learning and computer vision techniques. Multimed Tools Appl 80:14887–14914. https://doi.org/10.1007/s11042-021-10544-5
Sudarsan B, Ji W, Adamchuk V, Biswas A (2018) Characterizing soil particle sizes using wavelet analysis of microscope images. Comput Electron Agric 148:217–225. https://doi.org/10.1016/j.compag.2018.03.019
Yang YH (2007) Consistency of cross validation for comparing regression procedures. Ann Statist 35(6):2450–2473. https://doi.org/10.1214/009053607000000514
Zeiler MD, Fergus R (2013) Visualizing and understanding convolutional networks. arXiv. https://arxiv.org/abs/1311.2901
Acknowledgements
This work was supported by the Natural Science Foundation of Shanghai (18ZR1414500).
Funding
Natural Science Foundation of Shanghai, 18ZR1414500, Dejiang Wang.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors declare no competing interests.
Additional information
Responsible editor: Jun Zhou
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
Wang, D., Si, Y., Shu, Z. et al. An image-based soil type classification method considering the impact of image acquisition distance factor. J Soils Sediments 23, 2216–2233 (2023). https://doi.org/10.1007/s11368-023-03474-2
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11368-023-03474-2