Crop classification during the early stages is challenging because of the striking similarity in spectral and texture features among various crops. To improve classification accuracy, this study proposes a novel fusion-based deep learning approach. The approach integrates textural and spectral features from a fused dataset generated by merging Landsat 8-9 and Sentinel-2A data using the Gram-Schmidt fusion approach. The textural features were extracted using the multi-patch Gray Level Co-occurrence Matrix (GLCM) technique. The spectral features, namely the Enhanced Vegetation Index (EVI) and Normalized Difference Vegetation Index (NDVI), were obtained using the spectral index method. The five machine learning methods (deep neural network, 1D convolutional neural network, decision tree, support vector machine, and random forest) were trained using textural and spectral parameters to develop classifiers. The proposed approach achieves promising results using deep neural network (DNN), with an accuracy of 0.89, precision of 0.88, recall of 0.91, and F1-score of 0.90. These results demonstrate the effectiveness of the fusion-based deep learning approach in enhancing classification accuracy for early-stage crops.

Department of Biotechnology COMSATS University Islamabad Vehari Campus 61100 Vehari Pakistan

Department of Computer Science COMSATS University Islamabad Vehari Campus Vehari 61100 Pakistan

Department of Environmental Sciences COMSATS University Islamabad Vehari Campus Vehari 61100 Pakistan

Department of Plant Production College of Food and Agriculture Sciences King Saud University 11451 Riyadh Saudi Arabia

Department of Silviculture Faculty of Forestry and Wood Technology Mendel University Zemědělská 3 613 00 Brno Czech Republic

Department of Soil and Environmental Sciences Faculty of Crop Production Sciences The University of Agriculture Peshawar 25130 Pakistan

Institute of Agricultural Engineering Wroclaw University of Environmental and Life Sciences 37B Chełmonskiego Street 51 630 Wrocław Poland

Zobrazit více v PubMed

Cai, Y., Guan, K., Peng, J., et al. (2018). A high-performance and in-season classification system of field-level crop types using time-series Landsat data and a machine learning approach.

Cao, J., Cai, X., Tan, J., et al. (2021). Mapping paddy rice using Landsat time series data in the Ganfu Plain irrigation system, Southern China, from 1988–2017.

Demarez, V., Helen, F., Marais-Sicre, C., et al. (2019). In-season mapping of irrigated crops using Landsat 8 and Sentinel-1 time series.

Demelash, M., Tesfaw, B., & Tibebe, D. (2021). Crop field classification using data fusion of unmanned aerial vehicle (UAV) and Sentinel 2A satellite: The case of Oda Dhawata Kebele Cluster farmland, Oromia Region, Ethiopia. Research Square. 10.21203/rs.3.rs-1166571/v1

Fei, H., Fan, Z., Wang, C., et al. (2022). Cotton classification method at the county scale based on multi-features and random forest feature selection algorithm and classifier.

Hao, P. Y., Tang, H. J., Chen, Z. X., Meng, Q. Y., & Kang, Y. P. (2020). Early-season crop type mapping using 30-m reference time series.

Haryanto, T., Suhartanto, H., Arymurthy, A. M., & Kusmardi, K. (2021). Conditional sliding windows: An approach for handling data limitation in colorectal histopathology image classification.

Hsiou, D. C., Huang, F., Tey, F. J., et al. (2022). An automated crop growth detection method using satellite imagery data.

Hu, Q., Sulla-Menashe, D., Xu, B., et al. (2019). A phenology-based spectral and temporal feature selection method for crop mapping from satellite time series.

Inglada, J., Vincent, A., Arias, M., et al. (2016). Improved early crop type identification by joint use of high temporal resolution SAR and optical image time series.

Iqbal, N., Mumtaz, R., Shafi, U., et al. (2021). Gray level co-occurrence matrix (GLCM) texture based crop classification using low altitude remote sensing platforms. PubMed PMC

Khojastehnazhand, M., & Roostaei, M. (2022). Classification of seven Iranian wheat varieties using texture features.

Kobayashi, N., Tani, H., Wang, X., et al. (2020). Crop classification using spectral indices derived from Sentinel-2A imagery.

Koley, S., & Chockalingam, J. (2022). Sentinel 1 and Sentinel 2 for cropland mapping with special emphasis on the usability of textural and vegetation indices.

Kordi, F., & Yousefi, H. (2022). Crop classification based on phenology information by using time series of optical and synthetic-aperture radar images.

Li, G., Cui, J., Han, W., Zhang, H., Huang, S., Chen, H., & Ao, J. (2022). Crop type mapping using time-series Sentinel-2 imagery and U-net in early growth periods in the Hetao irrigation district in China.

Luo, P., Liao, J., & Shen, G. (2020). Combining spectral and texture features for estimating leaf area index and biomass of maize using Sentinel-1/2, and Landsat-8 data.

Maponya, M. G., Van Niekerk, A., & Mashimbye, Z. E. (2020). Pre-harvest classification of crop types using a Sentinel-2 time-series and machine learning.

Mekhalfa, F., & Yacef, F. (2021). Supervised learning for crop/weed classification based on color and texture features. arXiv preprint arXiv:2106.10581

Messina, G., Peña, J. M., Vizzari, M., et al. (2020). A comparison of UAV and satellites multispectral imagery in monitoring onion crop. An application in the ‘Cipolla Rossa di Tropea’ (Italy).

Moazzam, S. I., Khan, U. S., Qureshi, W. S., Tiwana, M. I., Rashid, N., Alasmary, W. S., Iqbal, J., & Hamza, A. (2021). A patch-image based classification approach for detection of weeds in sugar beet crop.

Mulla, D. J. (2013). Twenty five years of remote sensing in precision agriculture: Key advances and remaining knowledge gaps.

Muneer, A., & Fati, S. M. (2020). Efficient and automated herbs classification approach based on shape and texture features using deep learning.

Nasrallah, A., Baghdadi, N., Mhawej, M., et al. (2018). A novel approach for mapping wheat areas using high resolution Sentinel-2 images. PubMed PMC

Pan, L., Xia, H., Yang, J., et al. (2021). Mapping cropping intensity in Huaihe basin using phenology algorithm, all Sentinel-2 and Landsat images in Google Earth Engine.

Rahmati, A., Zoej, M. V., & Dehkordi, A. T. (2022). Early identification of crop types using Sentinel-2 satellite images and an incremental multi-feature ensemble method (Case study: Shahriar, Iran).

Saeed, M. F., Jamal, A., Ahmad, I., et al. (2020). Storage conditions deteriorate cotton and wheat seeds quality: An assessment of farmers’ awareness in Pakistan.

Seydi, S. T., Akhoondzadeh, M., Amani, M., & Mahdavi, S. (2021). Wildfire damage assessment over Australia using Sentinel-2 imagery and MODIS land cover product within the Google Earth Engine cloud platform.

Seydi, S. T., Amani, M., & Ghorbanian, A. (2022). A dual attention convolutional neural network for crop classification using time-series Sentinel-2 imagery.

Seydi, S. T., Abatzoglou, J. T., AghaKouchak, A., Pourmohamad, Y., Mishra, A., & Sadegh, M. (2024). Predictive understanding of links between vegetation and soil burn severities using physics-informed machine learning.

Sun, Z., Di, L., & Fang, H. (2019). Using long short-term memory recurrent neural network in land cover classification on Landsat and Cropland Data Layer time series.

Tang, K., Zhu, W., Zhan, P., et al. (2018). An identification method for spring maize in Northeast China based on spectral and phenological features.

Wang, Y., Zhang, Z., Feng, L., et al. (2021). A new attention-based CNN approach for crop mapping using time series Sentinel-2 images.

Wang, L., Wang, J., Zhang, X., et al. (2022). Deep segmentation and classification of complex crops using multi-feature satellite imagery.

Wei, C., Li, H., Shi, J., et al. (2022a). Row anchor selection classification method for early-stage crop row-following.

Wei, M., Wang, H., Zhang, Y., et al. (2022b). Investigating the potential of Sentinel-2 MSI in early crop identification in Northeast China.

Weilandt, F., Behling, R., Goncalves, R., Madadi, A., Richter, L., Sanona, T., Spengler, D., & Welsch, J. (2023). Early crop classification via multi-modal satellite data fusion and temporal attention.

Wu, Y., Wu, Y., Wang, B., & Yang, H. (2022). A remote sensing method for crop mapping based on multiscale neighborhood feature extraction.

Xia, L., Zhao, F., Chen, J., et al. (2022). A full resolution deep learning network for paddy rice mapping using Landsat data.

Yang, S., Gu, L., Li, X., et al. (2020). Crop classification method based on optimal feature selection and hybrid CNN-RF networks for multi-temporal remote sensing imagery.

Yi, Z., Jia, L., & Chen, Q. (2020). Crop classification using multi-temporal Sentinel-2 data in the Shiyang River Basin of China.

Yi, Z., Jia, L., Chen, Q., et al. (2022). Early-season crop identification in the Shiyang River basin using a deep learning algorithm and time-series Sentinel-2 data.

Yuan, W., Wijewardane, N. K., Jenkins, S., et al. (2019). Early prediction of soybean traits through color and texture features of canopy RGB imagery. PubMed PMC

Zhang, M., Lin, H., Wang, G., et al. (2018). Mapping paddy rice using a convolutional neural network (CNN) with Landsat 8 datasets in the Dongting Lake area, China.

Zhang, H., Kang, J., Xu, X., et al. (2020). Accessing the temporal and spectral features in crop type mapping using multi-temporal Sentinel-2 imagery: A case study of Yi’an County, Heilongjiang province, China.

Zhang, H., Du, H., Zhang, C., et al. (2021). An automated early-season method to map winter wheat using time-series Sentinel-2 data: A case study of Shandong, China.

Zhao, H., Chen, Z., Jiang, H., et al. (2019a). Evaluation of three deep learning models for early crop classification using Sentinel-1A imagery time series—A case study in Zhanjiang, China.

Zhao, L., Shi, Y., Liu, B., Hovis, C., Duan, Y., & Shi, Z. (2019b). Finer classification of crops by fusing UAV images and Sentinel-2A data.

Zhong, L., Hu, L., & Zhou, H. (2019). Deep learning based multi-temporal crop classification.