{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,12]],"date-time":"2026-06-12T16:37:28Z","timestamp":1781282248557,"version":"3.54.1"},"reference-count":62,"publisher":"MDPI AG","issue":"9","license":[{"start":{"date-parts":[[2021,4,21]],"date-time":"2021-04-21T00:00:00Z","timestamp":1618963200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Rural Development Administration, Korea","award":["PJ01350004"],"award-info":[{"award-number":["PJ01350004"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Remote Sensing"],"abstract":"<jats:p>When sufficient time-series images and training data are unavailable for crop classification, features extracted from convolutional neural network (CNN)-based representative learning may not provide useful information to discriminate crops with similar spectral characteristics, leading to poor classification accuracy. In particular, limited input data are the main obstacles to obtain reliable classification results for early crop mapping. This study investigates the potential of a hybrid classification approach, i.e., CNN-random forest (CNN-RF), in the context of early crop mapping, that combines the automatic feature extraction capability of CNN with the superior discrimination capability of an RF classifier. Two experiments on incremental crop classification with unmanned aerial vehicle images were conducted to compare the performance of CNN-RF with that of CNN and RF with respect to the length of the time-series and training data sizes. When sufficient time-series images and training data were used for the classification, the accuracy of CNN-RF was slightly higher or comparable with that of CNN. In contrast, when fewer images and the smallest training data were used at the early crop growth stage, CNN-RF was substantially beneficial and the overall accuracy increased by maximum 6.7%p and 4.6%p in the two study areas, respectively, compared to CNN. This is attributed to its ability to discriminate crops from features with insufficient information using a more sophisticated classifier. The experimental results demonstrate that CNN-RF is an effective classifier for early crop mapping when only limited input images and training samples are available.<\/jats:p>","DOI":"10.3390\/rs13091629","type":"journal-article","created":{"date-parts":[[2021,4,21]],"date-time":"2021-04-21T21:25:10Z","timestamp":1619040310000},"page":"1629","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":55,"title":["Potential of Hybrid CNN-RF Model for Early Crop Mapping with Limited Input Data"],"prefix":"10.3390","volume":"13","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-8474-1006","authenticated-orcid":false,"given":"Geun-Ho","family":"Kwak","sequence":"first","affiliation":[{"name":"Department of Geoinformatic Engineering, Inha University, Incheon 22212, Korea"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chan-won","family":"Park","sequence":"additional","affiliation":[{"name":"Research Policy Bureau, Rural Development Administration, Jeonju 54875, Korea"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kyung-do","family":"Lee","sequence":"additional","affiliation":[{"name":"National Institute of Agricultural Sciences, Rural Development Administration, Wanju 55365, Korea"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sang-il","family":"Na","sequence":"additional","affiliation":[{"name":"National Institute of Agricultural Sciences, Rural Development Administration, Wanju 55365, Korea"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ho-yong","family":"Ahn","sequence":"additional","affiliation":[{"name":"National Institute of Agricultural Sciences, Rural Development Administration, Wanju 55365, Korea"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9778-3624","authenticated-orcid":false,"given":"No-Wook","family":"Park","sequence":"additional","affiliation":[{"name":"Department of Geoinformatic Engineering, Inha University, Incheon 22212, Korea"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2021,4,21]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","unstructured":"Weiss, M., Jacob, F., and Duveiller, G. (2020). Remote sensing for agricultural applications: A meta-review. Remote Sens. Environ., 236.","DOI":"10.1016\/j.rse.2019.111402"},{"key":"ref_2","doi-asserted-by":"crossref","unstructured":"Kim, N., Ha, K.J., Park, N.W., Cho, J., Hong, S., and Lee, Y.W. (2019). A comparison between major artificial intelligence models for crop yield prediction: Case study of the Midwestern United States, 2006\u20132105. ISPRS Int. J. Geo-Inf., 8.","DOI":"10.3390\/ijgi8050240"},{"key":"ref_3","first-page":"829","article-title":"Application method of unmanned aerial vehicle for crop monitoring in Korea","volume":"34","author":"Na","year":"2018","journal-title":"Korean J. Remote Sens."},{"key":"ref_4","doi-asserted-by":"crossref","unstructured":"Kwak, G.H., and Park, N.W. (2019). Impact of texture information on crop classification with machine learning and UAV images. Appl. Sci., 9.","DOI":"10.3390\/app9040643"},{"key":"ref_5","doi-asserted-by":"crossref","unstructured":"Immitzer, M., Vuolo, F., and Atzberger, C. (2016). First experience with Sentinel-2 data for crop and tree species classifications in central Europe. Remote Sens., 8.","DOI":"10.3390\/rs8030166"},{"key":"ref_6","doi-asserted-by":"crossref","unstructured":"B\u00f6hler, J.E., Schaepman, M.E., and Kneub\u00fchler, M. (2018). Crop classification in a heterogeneous arable landscape using uncalibrated UAV data. Remote Sens., 10.","DOI":"10.3390\/rs10081282"},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"12859","DOI":"10.3390\/rs71012859","article-title":"In-season mapping of crop type with optical and X-band SAR data: A classification tree approach using synoptic seasonal features","volume":"7","author":"Villa","year":"2015","journal-title":"Remote Sens."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"5347","DOI":"10.3390\/rs70505347","article-title":"Feature selection of time series MODIS data for early crop classification using random forest: A case study in Kansas, USA","volume":"7","author":"Hao","year":"2015","journal-title":"Remote Sens."},{"key":"ref_9","doi-asserted-by":"crossref","unstructured":"Ji, S., Zhang, C., Xu, A., Shi, Y., and Duan, Y. (2018). 3D convolutional neural networks for crop classification with multi-temporal remote sensing images. Remote Sens., 10.","DOI":"10.3390\/rs10010075"},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"Wei, S., Zhang, H., Wang, C., Wang, Y., and Xu, L. (2019). Multi-temporal SAR data large-scale crop mapping based on U-Net model. Remote Sens., 11.","DOI":"10.3390\/rs11010068"},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Zhao, H., Chen, Z., Jiang, H., Jing, W., Sun, L., and Feng, M. (2019). Evaluation of three deep learning models for early crop classification using Sentinel-1A imagery time series\u2014A case study in Zhanjiang, China. Remote Sens., 11.","DOI":"10.3390\/rs11222673"},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"244","DOI":"10.1016\/j.rse.2017.04.026","article-title":"Early season large-area winter crop mapping using MODIS NDVI data, growing degree days information and a Gaussian mixture model","volume":"195","author":"Skakun","year":"2017","journal-title":"Remote Sens. Environ."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"179","DOI":"10.1016\/j.rse.2006.04.001","article-title":"The use of small training sets containing mixed pixels for accurate hard image classification: Training on mixed spectral responses for classification by a SVM","volume":"103","author":"Foody","year":"2006","journal-title":"Remote Sens. Environ."},{"key":"ref_14","doi-asserted-by":"crossref","unstructured":"Deng, F., Pu, S., Chen, X., Shi, Y., Yuan, T., and Pu, S. (2018). Hyperspectral image classification with capsule network using limited training samples. Sensors, 18.","DOI":"10.3390\/s18093153"},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"617","DOI":"10.1080\/01431160701352154","article-title":"The application of artificial neural networks to the analysis of remotely sensed data","volume":"29","author":"Mas","year":"2008","journal-title":"Int. J. Remote Sens."},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"1","DOI":"10.2480\/agrmet.D-15-00010","article-title":"Pixel-based crop classification in Peru from Landsat 7 ETM+ images using a random forest model","volume":"72","author":"Tatsumi","year":"2016","journal-title":"J. Agric. Meteorol."},{"key":"ref_17","doi-asserted-by":"crossref","unstructured":"Ma, L., Fu, T., Blaschke, T., Li, M., Tiede, D., Zhou, Z., Ma, X., and Chen, D. (2017). Evaluation of feature selection methods for object-based land cover mapping of unmanned aerial vehicle imagery using random forest and support vector machine classifiers. ISPRS Int. J. Geo-Inf., 6.","DOI":"10.3390\/ijgi6020051"},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Kim, Y., Park, N.W., and Lee, K.D. (2017). Self-learning based land-cover classification using sequential class patterns from past land-cover maps. Remote Sens., 9.","DOI":"10.3390\/rs9090921"},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"102","DOI":"10.1016\/j.isprsjprs.2013.08.007","article-title":"Impact of feature selection on the accuracy and spatial uncertainty of per-field crop classification using support vector machines","volume":"85","author":"Michel","year":"2013","journal-title":"ISPRS J. Photogramm. Remote Sens."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"756","DOI":"10.1016\/j.rse.2018.11.031","article-title":"dPEN: Deep Progressively Expanded Network for mapping heterogeneous agricultural landscape using WorldView-3 satellite imagery","volume":"221","author":"Sidike","year":"2019","journal-title":"Remote Sens. Environ."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"4544","DOI":"10.1109\/TGRS.2016.2543748","article-title":"Spectral\u2013spatial feature extraction for hyperspectral image classification: A dimension reduction and deep learning approach","volume":"54","author":"Zhao","year":"2016","journal-title":"IEEE Trans. Geosci. Remote Sens."},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"5408","DOI":"10.1109\/TGRS.2018.2815613","article-title":"Hyperspectral image classification with deep learning models","volume":"56","author":"Yang","year":"2018","journal-title":"IEEE Trans. Geosci. Remote Sens."},{"key":"ref_23","doi-asserted-by":"crossref","unstructured":"Song, H., Kim, Y., and Kim, Y. (2019). A patch-based light convolutional neural network for land-cover mapping using Landsat-8 images. Remote Sens., 11.","DOI":"10.3390\/rs11020114"},{"key":"ref_24","first-page":"681","article-title":"Combining 2D CNN and bidirectional LSTM to consider spatio-temporal features in crop classification","volume":"35","author":"Kwak","year":"2019","journal-title":"Korean J. Remote Sens."},{"key":"ref_25","first-page":"1273","article-title":"A convolutional neural network model with weighted combination of multi-scale spatial features for crop classification","volume":"35","author":"Park","year":"2019","journal-title":"Korean J. Remote Sens."},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"1797","DOI":"10.1109\/LGRS.2014.2309695","article-title":"Vehicle detection in satellite images by hybrid deep convolutional neural networks","volume":"11","author":"Chen","year":"2014","journal-title":"IEEE Geosci. Remote Sens. Lett."},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"2337","DOI":"10.1109\/TGRS.2017.2778300","article-title":"Rotation-insensitive and context-augmented object detection in remote sensing images","volume":"56","author":"Li","year":"2018","journal-title":"IEEE Trans. Geosci. Remote Sens."},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"8","DOI":"10.1109\/MGRS.2017.2762307","article-title":"Deep learning in remote sensing: A comprehensive review and list of resources","volume":"5","author":"Zhu","year":"2017","journal-title":"IEEE Geosci. Remote Sens. Mag."},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"430","DOI":"10.1016\/j.rse.2018.11.032","article-title":"Deep learning based multi-temporal crop classification","volume":"221","author":"Zhong","year":"2019","journal-title":"Remote Sens. Environ."},{"key":"ref_30","doi-asserted-by":"crossref","unstructured":"Cris\u00f3stomo de Castro Filho, H., Ab\u00edlio de Carvalho J\u00fanior, O., Ferreira de Carvalho, O.L., Pozzobon de Bem, P., dos Santos de Moura, R., Olino de Albuquerque, A., Rosa Silva, C., Guimaraes Ferreira, P.H., Fontes Guimaraes, R., and Trancoso Gomes, R.A. (2020). Rice crop detection using LSTM, Bi-LSTM, and machine learning models from sentinel-1 time series. Remote Sens., 12.","DOI":"10.3390\/rs12162655"},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"778","DOI":"10.1109\/LGRS.2017.2681128","article-title":"Deep learning classification of land cover and crop types using remote sensing data","volume":"14","author":"Kussul","year":"2017","journal-title":"IEEE Geosci. Remote Sens. Lett."},{"key":"ref_32","doi-asserted-by":"crossref","unstructured":"Xie, B., Zhang, H.K., and Xue, J. (2019). Deep convolutional neural network for mapping smallholder agriculture using high spatial resolution satellite image. Sensors, 19.","DOI":"10.3390\/s19102398"},{"key":"ref_33","first-page":"811","article-title":"Performance evaluation of machine learning and deep learning algorithms in crop classification: Impact of hyper-parameters and training sample size","volume":"34","author":"Kim","year":"2018","journal-title":"Korean J. Remote Sens."},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"4764","DOI":"10.3390\/s120404764","article-title":"Multiple classifier system for remote sensing image classification: A review","volume":"12","author":"Du","year":"2012","journal-title":"Sensors"},{"key":"ref_35","doi-asserted-by":"crossref","unstructured":"Yang, S., Gu, L., Li, X., Jiang, T., and Ren, R. (2020). Crop classification method based on optimal feature selection and hybrid CNN-RF networks for multi-temporal remote sensing imagery. Remote Sens., 12.","DOI":"10.3390\/rs12193119"},{"key":"ref_36","doi-asserted-by":"crossref","first-page":"1086","DOI":"10.1080\/2150704X.2019.1649736","article-title":"Hyperspectral image classification based on convolutional neural network and random forest","volume":"10","author":"Wang","year":"2019","journal-title":"Remote Sens. Lett."},{"key":"ref_37","doi-asserted-by":"crossref","first-page":"113","DOI":"10.1109\/JSTARS.2019.2953234","article-title":"Very high resolution remote sensing imagery classification using a fusion of random forest and deep learning technique\u2014Subtropical area for example","volume":"13","author":"Dong","year":"2020","journal-title":"IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens."},{"key":"ref_38","doi-asserted-by":"crossref","first-page":"3411","DOI":"10.1007\/s11042-018-5986-5","article-title":"DCNR: Deep cube CNN with random forest for hyperspectral image classification","volume":"78","author":"Li","year":"2019","journal-title":"Multimed. Tools Appl."},{"key":"ref_39","doi-asserted-by":"crossref","unstructured":"Inglada, J., Vincent, A., Arias, M., and Marais-Sicre, C. (2016). Improved early crop type identification by joint use of high temporal resolution SAR and optical image time series. Remote Sens., 8.","DOI":"10.3390\/rs8050362"},{"key":"ref_40","first-page":"621","article-title":"Field crop classification using multi-temporal high-resolution satellite imagery: A case study on garlic\/onion field","volume":"33","author":"Yoo","year":"2017","journal-title":"Korean J. Remote Sens."},{"key":"ref_41","unstructured":"(2021, January 09). Environmental Geographic Information Service (EGIS). Available online: http:\/\/egis.me.go.kr."},{"key":"ref_42","doi-asserted-by":"crossref","first-page":"5","DOI":"10.1023\/A:1010933404324","article-title":"Random forests","volume":"45","author":"Breiman","year":"2001","journal-title":"Mach. Learn."},{"key":"ref_43","doi-asserted-by":"crossref","unstructured":"Guidici, D., and Clark, M.L. (2017). One-Dimensional convolutional neural network land-cover classification of multi-seasonal hyperspectral imagery in the San Francisco Bay Area, California. Remote Sens., 9.","DOI":"10.3390\/rs9060629"},{"key":"ref_44","doi-asserted-by":"crossref","first-page":"436","DOI":"10.1038\/nature14539","article-title":"Deep learning","volume":"521","author":"LeCun","year":"2015","journal-title":"Nature"},{"key":"ref_45","doi-asserted-by":"crossref","unstructured":"Yoo, C., Lee, Y., Cho, D., Im, J., and Han, D. (2020). Improving local climate zone classification using incomplete building data and Sentinel 2 images based on convolutional neural networks. Remote Sens., 12.","DOI":"10.3390\/rs12213552"},{"key":"ref_46","doi-asserted-by":"crossref","unstructured":"Feng, Q., Zhu, D., Yang, J., and Li, B. (2019). Multisource hyperspectral and LiDAR data fusion for urban land-use mapping based on a modified two-branch convolutional neural network. ISPRS Int. J. Geo-Inf., 8.","DOI":"10.3390\/ijgi8010028"},{"key":"ref_47","doi-asserted-by":"crossref","first-page":"4806","DOI":"10.1109\/TGRS.2016.2551720","article-title":"Target classification using the deep convolutional networks for SAR images","volume":"54","author":"Chen","year":"2016","journal-title":"IEEE Trans. Geosci. Remote Sens."},{"key":"ref_48","first-page":"79","article-title":"Combining UAV-based plant height from crop surface models, visible, and near infrared vegetation indices for biomass monitoring in barley","volume":"39","author":"Bendig","year":"2015","journal-title":"Int. J. Appl. Earth Obs. Geoinf."},{"key":"ref_49","first-page":"2579","article-title":"Visualizing data using t-SNE","volume":"9","author":"Hinton","year":"2008","journal-title":"J. Mach. Learn. Res."},{"key":"ref_50","doi-asserted-by":"crossref","first-page":"5585","DOI":"10.1109\/TGRS.2017.2710079","article-title":"Deep fully convolutional network-based spatial distribution prediction for hyperspectral image classification","volume":"55","author":"Jiao","year":"2017","journal-title":"IEEE Trans. Geosci. Remote Sens."},{"key":"ref_51","unstructured":"(2021, January 09). Scikit-Learn: Machine Learning in Python. Available online: https:\/\/scikit-learn.org."},{"key":"ref_52","unstructured":"(2021, January 09). TensorFlow. Available online: https:\/\/tensorflow.org."},{"key":"ref_53","unstructured":"(2021, January 09). Keras Documentation. Available online: https:\/\/keras.io."},{"key":"ref_54","doi-asserted-by":"crossref","first-page":"627","DOI":"10.14358\/PERS.70.5.627","article-title":"Thematic map comparison: Evaluating the statistical significance of differences in classification accuracy","volume":"70","author":"Foody","year":"2004","journal-title":"Photogramm. Eng. Remote Sens."},{"key":"ref_55","doi-asserted-by":"crossref","unstructured":"Yi, Z., Jia, L., and Chen, Q. (2020). Crop classification using multi-temporal Sentinel-2 data in the Shiyang River Basin of China. Remote Sens., 12.","DOI":"10.5194\/egusphere-egu2020-20926"},{"key":"ref_56","doi-asserted-by":"crossref","unstructured":"Ren, T., Liu, Z., Zhang, L., Liu, D., Xi, X., Kang, Y., Zhao, Y., Zhang, C., Li, S., and Zhang, X. (2020). Early identification of seed maize and common maize production fields using Sentinel-2 images. Remote Sens., 12.","DOI":"10.3390\/rs12132140"},{"key":"ref_57","doi-asserted-by":"crossref","first-page":"593","DOI":"10.1109\/LGRS.2018.2878773","article-title":"Data augmentation for hyperspectral image classification with deep CNN","volume":"16","author":"Li","year":"2018","journal-title":"IEEE Geosci. Remote Sens. Lett."},{"key":"ref_58","unstructured":"Zhu, X. (2005). Semi-Supervised Learning Literature Survey, Department of Computer Sciences, University of Wisconsin. Technical Report 1530."},{"key":"ref_59","unstructured":"Settles, B. (2010). Active Learning Literature Survey, Department of Computer Sciences, University of Wisconsin. Technical Report 1648."},{"key":"ref_60","doi-asserted-by":"crossref","unstructured":"Hu, J., Shen, L., and Sun, G. (2018, January 18\u201322). Squeeze-And-Excitation Networks. Proceedings of the 2018 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Salt Lake City, UT, USA.","DOI":"10.1109\/CVPR.2018.00745"},{"key":"ref_61","doi-asserted-by":"crossref","first-page":"631","DOI":"10.1109\/JSTARS.2020.3033944","article-title":"Improved metric learning with the CNN for very-high-resolution remote sensing image classification","volume":"14","author":"Shi","year":"2021","journal-title":"IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens."},{"key":"ref_62","doi-asserted-by":"crossref","unstructured":"Park, S., and Park, N.W. (2020). Effects of class purity of training patch on classification performance of crop classification with convolutional neural network. Appl. Sci., 10.","DOI":"10.3390\/app10113773"}],"container-title":["Remote Sensing"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2072-4292\/13\/9\/1629\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,11]],"date-time":"2025-10-11T05:50:53Z","timestamp":1760161853000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2072-4292\/13\/9\/1629"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,4,21]]},"references-count":62,"journal-issue":{"issue":"9","published-online":{"date-parts":[[2021,5]]}},"alternative-id":["rs13091629"],"URL":"https:\/\/doi.org\/10.3390\/rs13091629","relation":{},"ISSN":["2072-4292"],"issn-type":[{"value":"2072-4292","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,4,21]]}}}