Analysis of Machine Learning Algorithms’ Performance in Multitemporal Image Classification for Agricultural Management (Case Study: Qazvin Plain)

Document Type : Research/Original/Regular Article

Authors

1 Assistant Professor, Water Science and Engineering Department, Agriculture and Natural Resources Faculty, Imam Khomeini International University, Qazvin, Iran

2 Ph.D in Irrigation and Drainage, Water Science and Engineering Department, Agriculture and Natural Resources Faculty, Imam Khomeini International University, Qazvin, Iran

Abstract

Abstract

Introduction

Due to the world's population growth and the severe problems caused by climate change, there is an immediate demand for sustainable farming methods and effective natural resource management. Agriculture forms the foundation of economic stability and food security, so methods that lead to maximum resource utilization are needed. However, they cause minimal environmental disturbance. This goal requires the ability to carry out fast and accurate crop mapping since it is used in strategizing agricultural activities, assessing yield, and ensuring sustainability.

Monitoring and mapping crop types by conventional field-based approaches have been the norm for a long time. They are primarily laborious, time-consuming, and expensive, especially on large scales. Remote sensing approaches, fueled by new-generation satellite imagery and machine-learning capabilities, offer a viable alternative to enabling large-scale monitoring and detailed classification of land cover and crop types. The most promising tools in this respect are the Sentinel-1 and Sentinel-2 satellites, with high spatial, temporal, and spectral resolution data, ideally suited for agricultural monitoring.

Sentinel-1 provides radar imagery, which is especially useful for monitoring vegetation structure even in cloudy conditions, while Sentinel-2 provides high-resolution optical images with multiple spectral bands suited for vegetation study. These datasets can be combined to capture complementary information on the physical and spectral characteristics of the land surface.

There are different kinds of ML algorithms. Here, the performance of three most common algorithms are compared: RF, SVM, and XGBoost. The relative strengths of each of the algorithms in classification provide insights that are critical to their suitability in diverse agricultural scenarios.



Materials and Methods

This study was conducted in the Qazvin Irrigation Network, an agriculturally significant area of about 80,000 hectares in Iran. The study area includes different types of land cover, such as wheat, alfalfa, fallow land, urban, and bare land, which are the most cultivated crops in the area. This heterogeneity makes classification difficult, especially in semi-arid regions where crops and other land cover classes have similar spectral signatures. The data from the Sentinel-1 and Sentinel-2 satellites were used for such challenges. The former acquires reliable radar data that captures surface roughness and soil moisture even under cloudy conditions. Meanwhile, the latter delivers high-resolution optical imagery with many spectral bands, which is excellent for studying vegetation health and structure. Several necessary data-preprocessing steps were carried out to ensure that accurate classifications were developed. Atmospheric and sensor noise reduction was accomplished via radiometric and geometric correction, respectively, for radar and optical imagery. Derived spectral indices such as NDVI, SAVI, and LAI aided the detection of vegetation characteristics under study by enhancing separability at spectral levels. Furthermore, temporal fusion was performed by combining images taken at different times to account for the phenological changes in vegetation over the growing seasons. These preprocessing steps allowed for a robust dataset representing spatial and temporal land cover changes.

Finally, three machine-learning algorithms were implemented for classifying the preprocessed satellite images: RF, SVM, and XGBoost. The reasons for choosing RF, an ensemble-based approach, include its robustness to noise and its ability to handle complicated datasets. On the other hand, SVM was adopted because it optimizes classification boundaries through its kernel-based feature. XGBoost is a highly accurate advanced gradient boosting technique that can realize large-scale computing with low expenses. The dataset was then divided into a training and testing set in a 70/30 ratio to prevent overfitting and ensure the model's reliability. Classification accuracy was assessed based on overall accuracy and the kappa coefficient, while the Jeffries-Matusita (JM) test quantified spectral separability between land cover classes.



Results and Discussion

The results demonstrated that integrating optical and radar data significantly improved classification accuracy. Among the three algorithms, RF outperformed the others, achieving an overall accuracy of 93.98% and a kappa coefficient of 0.996. These results highlight RF's ability to handle spectrally overlapping classes and complex datasets effectively.

The XGBoost algorithm also performed well, achieving an overall accuracy of 93.94%. However, its performance was slightly hindered by its inability to distinguish between classes with similar spectral characteristics, such as wheat and alfalfa. While providing reasonable results, SVM achieved a lower overall accuracy of 83.79%, mainly due to its sensitivity to spectral overlap.

The JM test revealed that certain classes, such as wheat and alfalfa, exhibited low spectral separability. This limitation underscores the importance of integrating radar data and spectral indices to enhance differentiation. The study also highlighted the potential of temporal data fusion to capture phenological changes, further improving classification performance.



Conclusion

This study indicated the potential of integrating multi-source remote sensing data and machine learning algorithms for crop classification in semi-arid regions. The RF algorithm proved the most accurate and robust method, showing its adaptability to the heterogeneous and complicated nature of the datasets. XGBoost and SVM are also very promising, but their performance could be improved further with additional parameter optimization.

Future research should investigate the application of more advanced techniques, such as CNNs and deep learning frameworks, to improve classification accuracy further. A deeper understanding of the dynamics of crops and land use changes can be achieved by including multi-temporal and multi-spectral datasets.

The results of such a study can have substantial implications for sustainable agriculture and resource management. In this context, remote sensing and machine learning technologies offer means to address critical challenges related to food security and environmental conservation in the most climate-vulnerable regions.

Keywords

Main Subjects

References

منابع
اسدی، بهار، و شمس‌الدینی، علی (1403). تفکیک محصولات زراعی با استفاده از ترکیب تصاویر سنتینل-1 و 2 در استان اردبیل. سنجش از دور و GIS ایران، 16(3)، 25-46 doi:/10.48308/gisj.2023.103095
اسماعیل‌نژاد، رضا، و زینال‌زاده، کامران (1398). ارزیابی تغییرات کاربری اراضی با استفاده از سنجش از دور و سیستم اطلاعات جغرافیایی در زیرحوضه نازلوچای. مدیریت خاک و تولید پایدار، 9(4)، 159-172 doi:10.22069/ejsms.2020.16657.1892
پورحسن، ناهید، شاه حسینی، رضا، و سیدی، سید تیمور ( ۱۴۰۰). ارائه روش طبقه‌بندی مبتنی بر یادگیری عمیق در تفکیک انواع محصولات کشاورزی با استفاده از تصاویر ماهواره‌ای سری زمانی. علوم و فنون نقشه برداری، ۱۱(۱)، 142-129. dor: 20.1001.1.2322102.1400.11.1.10.8
خسروی، ایمان (1403). تهیۀ نقشۀ نوع محصول کشاورزی از سری زمانی تصاویر لندست-8 با استفاده از روش‌های یادگیری ماشین (مطالعۀ موردی: مرودشت استان فارس). جغرافیا و برنامه‌ریزی محیطی، 35(2)،66-45.
doi:10.22108/gep.2024.138615.1601
درویش‌هندی، محسن، و امیری تکلدانی، ابراهیم (1402). چالش‌های مدیریت آب کشاورزی در شبکه آبیاری دشت قزوین. تحقیقات آب و خاک ایران، 54(12)، 1962-1945
 doi:10.22059/ijswr.2023.365841.669581
صالحی شفا، نیما، بابازاده، حسین، آقایاری، فیاض، صارمی، علی، غفوری، محمدرضا، صفوی، مسعود، و پناهدار، علی (1403)، تدوین الگوی کشت بهینه به‌منظور مدیریت تغییرات سطح آب زیرزمینی دشت شهریار. مدل‌سازی و مدیریت آب و خاک، 3(2)، 235-217.
 doi:10.22098/mmws.2022.11792.1169
راستی، سعید، مهدوی فرد، مصطفی، شیخ قادری، هدایت، نصیری، ابوذر، و تکتاز، نازنین زهرا (1401). بهبود دقت طبقه‌بندی با ترکیب تصاویر چندفصلی سنتینل 1 و 2 به‌منظور تهیة نقشة کاربری اراضی در فضای ابری گوگل ارث انجین (مطالعة موردی: استان گیلان). جغرافیا و روابط انسانی، 5(3)، 373-357  . doi:10.22034/gahr.2022.336692.1696
رمضانی اعتدالی، هادی، و احمدی، مژگان (1403). بررسی ارتباط بین شاخص‌های خشکسالی با عملکرد ذرت با استفاده از روش جنگل تصادفی (مطالعة موردی: شبکه آبیاری دشت قزوین). نیوار، 126-127، 137-127. doi:10.30467/nivar.2024.467444.1299
ساعی جمال آباد، موسی، آبکار، علی اکبر، و مجردی، برات (1397). طبقه‌بندی گندم زمستانه با استفاده از آنالیز تصاویر بهینه چند زمانی مبتنی بر الگوریتم جنگل تصادفی. علوم و فنون نقشه برداری، 8 (2 )، 150-133. dor:20.1001.1.2322102.1397.8.2.9.8
قدسی، زینب، خیرخواه زرکش، میرمسعود، و قرمزچشمه، باقر (1399). مقایسة دقت روش‌های ماشین بردار پشتیبان و جنگل تصادفی در تهیة نقشة کاربری اراضی و محصولات زراعی، با استفاده از تصاویر چندزمانة سنتینل-2. سنجش از دور وGIS ایران، 12(4)، 92-73. doi:10.52547/gisj.12.4.73
 
 
References
Alejo, R., Garcia, V., Sotoca, J.M., Mollineda, R.A., & Sánchez, J.S. (2006). Improving the classification accuracy of RBF and MLP neural networks trained with imbalanced samples. IDEAL 2006. Lecture Notes in Computer Science, Vol 4224. Springer, Berlin, Heidelberg. doi:10.1007/11875581_56.
Aria, M., Cuccurullo, C., & Gnasso, A. (2021). A comparison among interpretative proposals for Random Forests. Machine Learning with Applications, 6, 100094. doi:10.1016/j.mlwa.2021.100094.
Asadi, B., & Shamsoddini, A. (2024). Crop mapping using a combination of Sentinel-1 and 2 images in Ardabil province. Remote Sensing and GIS, 16(3), 25-46. doi:10.48308/gisj.2023.103095[In Persian]
Asgari, S., & Hasanlou, M. (2023). A comparative study of machine learning classifier for crop type mapping using vegetation indices. ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, X-4/W1-2022, 79. doi: 10.5194/isprs-annals-X-4-W1-2022-79-2023
Ashourloo, D., Nematollahi, H., Huete, A., Aghighi, H., Azadbakht, M., Shahrabi, H.S., & Goodarzdashti, S. (2022). A new phenology-based method for mapping wheat and barley using time-series of Sentinel-2 images. Remote Sensing of Environment, 280, 113206. doi: 10.1016/j.rse.2022.113206.
Breiman, L. (2001). Random forests. Machine Learning, 45, 5-32. doi: 10.1023/A:1010933404324.
Chauhan, K.K., & Lunagaria, M.M. (2025). Remote sensing-based crop identification and acreage estimation of Rabi Wheat in Anand, Gujarat. International Environment and Climate Change, 15(1), 1-11 doi: 10.9734/ijecc/2025/v15i14668.
Chen, C., Liaw, A., & Breiman, L. (2004). Using random forest to learn imbalanced data. University of California, Berkeley, 110(1-12), 24.
Chen, T., & Guestrin, C. (2016). Xgboost: A scalable tree boosting system. Proceedings of the 2nd ACM Sigkdd International Conference on Knowledge Discovery and Data Mining, doi: 10.1145/2939672.2939785.
Gili, M.R., Ashourloo, D., Aghighi, H., Matkan, A. A., & SHakiba, A. (2022). Crop classification with deep convolutional neural network based on crop feature. Environmental Sciences, 20(4), 37-52. doi: 10.48308/envs.2022.1126
d’Andrimont, R., Verhegghen, A., Lemoine, G., Kempeneers, P., Meroni, M., & van der Velde, M. (2021). From parcel to continental scale – A first European crop type map based on Sentinel-1 and LUCAS Copernicus in-situ observations. Remote Sensing of Environment, 266, 112708. doi: 10.1016/j.rse.2021.112708.
Darvish Hendi, M., & Amiri Tokaldany, E. (2024). Agricultural water management challenges in Qazvin Plain Irrigation Network. Soil and Water Research, 54(12), 1945-1962. doi: 10.22059/ijswr.2023.365841.669581 [In Persian]
Dong, Q., Chen, X., Chen, J., Zhang, C., Liu, L., Cao, X.,Zang, Y., Zhu, X., & Cui, X. (2020). Mapping winter wheat in North China using Sentinel 2A/B data: A method based on phenology-time weighted dynamic time warping. Remote Sensing, 12(8), 1274. doi: 10.3390/rs12081274.
Esmaeilnezhad, r., & Zeinalzadeh, K. (2020). Evaluation of land use changes using remote sensing and GIS in Nazlou Chai sub basin. Soil Management and Sustainable Production, 9(4), 159-172. doi: 10.22069/ejsms.2020.16657.1892 [In Persian]
Fan, J., Wang, X., Wu, L., Zhou, H., Zhang, F., Yu, X., Xiang, L., & Xiang, X. (2018). Comparison of Support Vector Machine and Extreme Gradient Boosting for predicting daily global solar radiation using temperature and precipitation in humid subtropical climates: A case study in China. Energy Conversion and Management, 164. doi: 10.1016/j.enconman.2018.02.087.
Friedman, J.H. (2001). Greedy function approximation: a gradient boosting machine. Annals of Statistics, 29(5) , 1189-1232. doi: 10.1214/aos/1013203451
Ghodsi, Z., Kheirkhah Zarkesh, M.M., & Ghermezcheshmeh, B. (2021). Comparison of accuracy between Support Vector Machine and Random Forest Classifiers for land use and crop mapping using multi-temporal Sentinel-2 Images. Iranian Remote Sensing and GIS, 12(4), 73-92. doi: 10.52547/gisj.12.4.73 [In Persian]
Gholamy, A., Kreinovich, V., & Kosheleva, O. (2018). Why 70/30 or 80/20 relation between training and testing sets: A pedagogical explanation. International Journal of Intelligent Systems Technologies and Applications, 11(2), 105-111. doi: 10.6148/IJITAS.201806_11(2).0003.
Goldblatt, R., You, W., Hanson, G., & Khandelwal, A.K. (2016). Detecting the boundaries of urban areas in india: A dataset for pixel-based image classification in google earth engine. Remote Sensing, 8(8), 634. doi: 10.3390/rs8080634.
Hastie, T., Tibshirani, R., and Friedman, J. (2009). The Elements of statistical learning. 1st Edition: Springer. 536 pages.
Huang, Y., Reddy, K.N., Fletcher, R.S., & Pennington, D. (2018). UAV low-altitude remote sensing for precision weed management. Weed Technology, 32(1), 2-6. doi: 10.1017/wet.2017.89
Khosravi, I. (2024). Crop mapping from Landsat-8 images time series using Machine-Learning Methods (Case Study: Marvdasht in Fars Province). Geography and Environmental Planning, 35(2), 45-66. doi: 10.22108/gep.2024.138615.1601 [In Persian]
Li, S., Gong, Q., & Yang, S. (2019). A sustainable, regional agricultural development measurement system based on dissipative structure theory and the entropy weight method: a case study in Chengdu, China. Sustainability, 11(19), 5313. doi: 10.3390/su11195313
Li, W., Zhang, H., Li, W., & Ma, T. (2022). Extraction of winter wheat planting area based on multi-scale fusion. Remote Sensing, 15(1), 164. doi: 10.3390/rs15010164.
Li, Y., Zhu, X., Pan, Y., Gu, J., Zhao, A., & Liu, X. (2014). A Comparison of model-assisted estimators to infer land cover/use class area using satellite imagery. Remote Sensing, 6(9), 8904-8922. doi: 10.3390/rs6098904.
Saei Jamalabad, M., Akbari, A., & Mojardi, B. (2018). Winter wheat classification by multi-temporal optimized image analysis based on Random Forest algorithm. Geomatic Science and Technology, 8(2), 133-150. dor: 20.1001.1.2322102.1397.8.2.9.8 [In Persian]
Moumni, A., & Lahrouni, A. (2021). Machine learning‐Based classification for crop‐type mapping using the fusion of high‐resolution satellite imagery in a Semiarid Area. Scientifica, 2021(1), 8810279. doi: 10.1155/2021/8810279.
Neetu, & Ray, S. (2019). Exploring machine learning classification algorithms for crop classification using Sentinel 2 data. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 42, 573-578. doi: 10.5194/isprs-archives-XLII-3-W6-573-2019.
Padma VVL, A., Suhail, M., Lutfullo, I., & Shodiyor, B. (2024). A comparative study of three supervised algorithms for mixed crop classification. E3S Web of Conference, Pp. 590-595, doi: 10.1051/e3sconf/202459001004.
Phalke, A. R., Özdoğan, M., Thenkabail, P.S., Erickson, T., Gorelick, N., Yadav, K., & Congalton, R.G. (2020). Mapping croplands of Europe, Middle East, Russia, and Central Asia using Landsat, Random Forest, and Google Earth Engine. ISPRS Photogrammetry and Remote Sensing, 167, 104-122. doi: 10.1016/j.isprsjprs.2020.06.022.
Pourhasan, N., Shah-Hosseini, R., & Seydi, S.T. (2021). Deep learning-based classification method for crop mapping using time series satellite images. Geomatics Science and Technology, 11(1), 129-142. dor: 20.1001.1.2322102.1400.11.1.10.8 [In Persian]
Ramezani Etedali, H., & Ahmadi, M. (2024). Investigating the relationship between drought indices and maize yield using random forest method (Case study: Qazvin plain irrigation network). Nivar, 126, 127-137. doi: 10.30467/nivar.2024.467444.1299 [In Persian]
Rasti, S., Mahdavifardnh, M., Shaykh Ghaderi, H., Nasiri, A., & Taktaz, N.Z. (2022). Improving classification accuracy by combining multi-season images of Sentinel 1 and 2 in order to prepare a land use map in the cloud space of Google Earth Engine (Case study: Guilan province). Geography and Human Relationships, 5(3), 357-373. doi: 10.22034/gahr.2022.336692.1696 [In Persian]
Sahin, E.K. (2020). Assessing the predictive capability of ensemble tree methods for landslide susceptibility mapping using XGBoost, gradient boosting machine, and random forest. SN Applied Sciences, 2(7), 1308. doi: 10.1007/s42452-020-3060-1.
Salehi Shafa, N., Babazadeh, H., Aghayari, F., Saremi, A., Ghafouri, M.R., Safavi, M., & Panahdar, A. (2022). Formulation of an optimized Cropping pattern in order to manage groundwater level changes in Shahriar Plain. Water and Soil Management and Modelling, 3(2), 217-235. doi: 10.22098/mmws.2022.11792.1169 [In Persian]
Sarker, I.H. (2021). Machine learning: Algorithms, real-world applications and research directions. SN Computer Science, 2(3), 160. doi: 10.1007/s42979-021-00592-x.
Seifi Majdar, R., & Ghassemian, H. (2017). A probabilistic SVM approach for hyperspectral image classification using spectral and texture features. Remote Sensing, 38(15), 4265-4284. doi: 10.1080/01431161.2017.1317941.
Sen, R., Goswami, S., & Chakraborty, B. (2019). Jeffries-Matusita distance as a tool for feature selection. International Conference on Data Science and Engineering (ICDSE). doi: 10.1109/ICDSE47409.2019.8971800
Shao, Z., Ahmad, M.N., & Javed, A. (2024). Comparison of Random Forest and XGBoost classifiers using integrated optical and SAR features for mapping urban impervious surface. Remote Sensing, 16(4), 665. doi: 10.3390/rs16040665.
Shi, D., & Yang, X. (2015). Support Vector Machines for land cover mapping from remote sensor imagery. Pp. 265-279. In Monitoring and Modeling of Global Changes: A Geomatics Perspective. SPRINGERREMO. doi: 10.1007/978-94-017-9813-6_13.
Simsek, F. (2025). Comparison of Agricultural Crop Type Classifications with Different Machine Learning Algorithms (RF-SVM-ANN-XGBoost) by Generating Ground Truth Data from Farmer Declaration Parcels. Engineering and Geosciences, 10(2), 207-220. doi: 10.26833/ijeg.1552141.
Sun, W., Yang, G., Chen, C., Chang, M., Huang, K., Meng, X., & Liu, L. (2020). Development status and literature analysis of China’s earth observation remote sensing satellites. Remote Sensing, 24(5), 479-510. doi: 10.11834/jrs.20209464.
Tuia, D., Volpi, M., Copa, L., Kanevski, M., & Munoz-Mari, J. (2011). A survey of active learning algorithms for supervised remote sensing image classification. IEEE Journal of Selected Topics in Signal Processing, 5(3), 606-617. doi: 10.1109/JSTSP.2011.2139193.
Ugirumurera, J., Bensen, E.A., Severino, J., & Sanyal, J. (2024). Addressing bias in bagging and boosting regression models. Scientific Reports, 14(1), 18452. doi: 10.1038/s41598-024-68907-5.
Viskovic, L., Kosovic, I.N., & Mastelic, T. (2019, 19-21 Sept. 2019). Crop classification using multi-spectral and multitemporal satellite imagery with machine learning. 2019 International Conference on Software, Telecommunications and Computer Networks (SoftCOM), doi: 10.23919/SOFTCOM.2019.8903738.
Wang, K., Cheng, L., & Yong, B. (2020). Spectral-similarity-based Kernel of SVM for hyperspectral image classification. Remote Sensing, 12(13), 2154. doi: 10.3390/rs12132154.
Wang, Y., Qi, Q., & Liu, Y. (2018). Unsupervised segmentation evaluation using area-weighted variance and Jeffries-Matusita distance for remote sensing Images. Remote Sensing, 10(8), 1193. doi: 10.3390/rs10081193.
Wuyun, D., Sun, L., Sun, Z., Chen, Z., Hou, A., Teixeira Crusiol, L. G., Reymond, L., Chen, R., & Zhao, H. (2022). Mapping fallow fields using Sentinel-1 and Sentinel-2 archives over farming-pastoral ecotone of Northern China with Google Earth Engine. GIScience and Remote Sensing, 59(1), 333-353. doi: 10.1080/15481603.2022.2026638.
Yadav, S., Sharma, S., & Chaudhary, P. (2024) Review on advancement in machine learning and Deep Learning techniques for crop classification. SCRS Publication. doi: 10.56155/978-81-955020-7-3-53.
Yaramasu, R., Bandaru, V., & Pnvr, K. (2020). Pre-season crop type mapping using deep neural networks. Computers and Electronics in Agriculture, 176, 105664. doi: 10.1016/j.compag.2020.105664.
Zhan, W., Luo, F., Luo, H., Li, J., Wu, Y., Yin, Z., Wu, Y., & Wu, P. (2024). Time-series-based spatiotemporal fusion network for improving crop type mapping. Remote Sensing, 16(2), 235. doi: 10.3390/rs16020235.
Zheng, B., Myint, S.W., Thenkabail, P.S., & Aggarwal, R.M. (2015). A support vector machine to identify irrigated crop types using time-series Landsat NDVI data. Applied Earth Observation and Geoinformation, 34, 103-112. doi: 10.1016/j.jag.2014.07.002.
 
Water and Soil Management and Modelling
Volume 5, Issue 3
September 2025
Pages 54-73

Files

History

  • Receive Date: 23 December 2024
  • Revise Date: 10 March 2025
  • Accept Date: 15 March 2025

Share

How to cite

Statistics