| 41 | 0 | 47 |
| 下载次数 | 被引频次 | 阅读次数 |
灌区是我国粮食生产的重要区域，开展灌区耕地地块精细提取与作物识别对保障国家粮食安全和提升农业管理水平具有重要意义。在丘陵灌区复杂地形与破碎耕地条件下，基于单一遥感数据往往难以兼顾地块边界精度与作物时序特征，导致分类精度受限。针对这一问题，本文以云南昭通大型灌区为例，利用GF-1影像和CPVF(Cultivated Parcel Vectorization Framework)框架提取耕地地块边界，然后在Sentinel-2影像上计算地块尺度NDVI等关键物候指标，最后将各地块的物候曲线与不同作物的标准物候规则进行匹配，实现灌区主要作物的地块级识别。结果表明，该方法总体分类精度为75.35%，其中小麦和水稻识别精度较高，玉米次之，马铃薯因生长周期灵活而精度偏低；2020—2024年玉米始终为主导作物，马铃薯为第二大作物，二者间的相互转化最频繁，水稻面积虽小但呈增长趋势。研究结果证明GF-1与Sentinel-2相结合能有效提升复杂丘陵灌区的作物分类精度，可为区域农业监测、粮食安全评估及耕地精细化管理提供新思路。
Abstract：Irrigation districts are key grain-producing areas in China,and accurate extraction of cultivated land parcels is important to gather with crop mapping of great significance for safeguarding national food security and improving agricultural management.Inhilly irrigation districts with complex terrain and highly fragmented farmland,classification methods based on a single remote sensing data source often cannot simultaneously ensure parcel boundary accuracy and capture crop temporal characteristics,thus limiting classification performance.To address this problem,taking the large irrigation district in Zaotong,Yunnan Province as a case study,this paper proposes a parcel-level fine-grained crop type identification method integrating GF-1 and Sentinel-2 imagery.GF-1 imagery and the CPVF(Cultivated Parcel Vectorization Framework) are first used to extract cultivated land parcel boundaries.Then,parcel-level NDVI and other key phenological indicators are derived from Sentinel-2 time-series imagery.Finally,the phenological curve of each parcel is matched with crop-specific reference phenological profiles to identify the main crop at the parcel scale.The proposed method achieves an overall classification accuracy of 75.35%.Wheat and rice obtain the highest classification accuracies,followed by maize,while maize shows low accuracy due to its flexible growing season.Cropping structure analysis from 2020 to 2024 indicates that maize has remained the dominant crop,with potato as the second largest crop;rice occurs relatively small but continuously increasing share,and conversions between maize and potato are the most frequent.These results demonstrate that combining GF-1 and Sentinel-2 data effectively improves crop classification accuracy in complex hilly irrigation districts and provides a new approach for regional agricultural monitoring,food security assessment,and refined cultivated land management.
[1]中华人民共和国中央政府.水利部、国家发改委正式印发实施“十四五”重大农业节水供水工程实施方案[EB/OL].(2021-08-16)[2026-01-10].http://nssd.mwr.gov.cn/xyyw/202209/t20220903_1254751.html.
[2]Li Huapeng,Zhang Ce,Zhang Shuqing,et al.Iterative deep learning(IDL)for agricultural landscape classification using fine spatial resolution remotely sensed imagery[J].International Journal of Applied Earth Observation and Geoinformation,2021,102:102437.DOI:10.1016/j.jag.2021.102437.
[3]孔嘉鑫,张昭臣,张健.基于多源遥感数据的植物物种分类与识别:研究进展与展望[J].生物多样性,2019,29(7):796-812.
[4] Zhang Dongyun,Wu Tianjun,Luo Jiancheng,et al.Precise crop planting structure mapping method based on spatial-temporal collaboration of remote sensing[J].National Remote Sensing Bulletin,2024,28(8):2014-2029.
[5] Kibret K S,Marohn C,Cadisch G.Use of MODIS EVI to map crop phenology,identify cropping systems,detect land use change and drought risk in Ethiopia:an application of Google Earth Engine[J].European Journal of Remote Sensing,2020,53(1):176-191.
[6] Longchamps L,Philpot W.Full-season crop phenology monitoring using two-dimensional normalized difference pairs[J].Remote Sensing,2023,15(23):5565.DOI:10.3390/rs15235565.
[7] Gao Feng,Zhang Xiaoyang.Mapping crop phenology in near real-time using satellite remote sensing:challenges and opportunities[J].Journal of Remote Sensing,2021,2021(2):1-14.
[8]Md-Tahir H,Mahmood H S,Husain M,et al.Localized crop classification by NDVI time series analysis of remote sensing satellite data;applications for mechanization strategy and integrated resource management[J].AgriEngineering,2024,6(3):2429-2444.
[9]Wardlow B D,Egbert S L.Large-area crop mapping using time-series MODIS 250 m NDVI data:an assessment for the U.S.Central Great Plains[J].Remote Sensing of Environment,2008,112(3):1096-1116.
[10]Jamil M D,Abbas M Z,Saeed M F,et al.Improved early-stage crop classification using a novel fusion-based machine learning approach with Sentinel-2 A and Landsat 8-9 data[J].Environmental Monitoring and Assessment,2025,197(9):982.DOI:10.1007/s10661-025-14420-9.
[11]Zeng Tian,Zhang Zengxiang,Zhao Xiaoli,et al.Evaluation of the 2010 MODIS collection 5.1 land cover type product over China[J].Remote Sensing,2015,7(2):1981-2006.
[12]Foerster S,Kaden K,Foerster M,et al.Crop type mapping using spectral-temporal profiles and phenological information[J].Computers and Electronics in Agriculture,2012,89:30-40.
[13]Aneece I P,Thenkabail P,Mccormick R L,et al.Machine learning and new-generation spaceborne hyperspectral data advance crop type mapping[J].Photogrammetric Engineering&Remote Sensing,2024,90(11):687-698.
[14] De Oliveira Júnior J G,Esquerdo J C D M,Lamparelli R A C.Optimum combination of spectral variables for crop mapping in heterogeneous landscapes based on Sentinel-2 time series and machine learning[J].ISPRS Annals of the Photogrammetry,Remote Sensing and Spatial Information Sciences,2024,X-3-2024:85-92.
[15]Wang Cong,Zhang Xinyu,Wang Wenjing,et al.Understanding the potentials of early-season crop type mapping by using Landsat-8,Sentinel-1/2,and GF-1/6 data[J].Computers and Electronics in Agriculture,2024,224:109239.DOI:10.1016/j.compag.2024.109239.
[16]Wang Jiayue,Cai Zhiwen,Wang Wenjing,et al.Integrating multisource Gaofen images and object-based methods for crop type identification in South China[J].Scientia Agricultura Sinica,2023,56(13):2474-2490.
[17]Yang Yingpin,Huang Qiting,Wu Wei,et al.Geo-parcel based crop identification by integrating high spatial-temporal resolution imagery from multi-source satellite data[J].Remote Sensing,2017,9(12):1298.DOI:10.3390/rs9121298.
[18]Jin Xiuliang,Kumar L,Li Zhenhai,et al.A review of data assimilation of remote sensing and crop models[J].European Journal of Agronomy,2018,92:141-152.
[19] Blickensdörfer L,Schwieder M,Pflugmacher D,et al. Mapping of crop types and crop sequences with combined time series of Sentinel-1,Sentinel-2 and Landsat 8 data for Germany[J].Remote Sensing of Environment,2022,269:112831.DOI:10.1016/j.rse.2021.112831.
[20] Kambli M,Palkar B.Categorisation of vegetation using machine learning and remote sensing methods[J].Journal of Image and Graphics,2025,13(2):174-189.
[21]Feng Fukang,Gao Maofang,Liu Ronghua,et al.A deep learning framework for crop mapping with reconstructed Sentinel-2time series images[J].Computers and Electronics in Agriculture,2023,213:108227.DOI:10.1016/j.compag.2023.108227.
[22]Song Xiaopeng,Huang Wenli,Hansen M C,et al.An evaluation of Landsat,Sentinel-2,Sentinel-1 and MODIS data for crop type mapping[J].Science of Remote Sensing,2021,3:100018.DOI:10.1016/j.srs.2021.100018.
[23]康育鹏.基于GF-6 WFV和Sentinel-1/2数据红边特征的农作物分类研究:以衡水市为例[D].焦作:河南理工大学,2023.
[24]郑利娟.基于高分一/六号卫星影像特征的农作物分类研究[D].北京:中国科学院大学(中国科学院遥感与数字地球研究所),2017.
[25]Chen Chen,Dong Taifeng,Wang Zhaohai,et al.Exploring optimal features and image analysis methods for crop type classification from the perspective of crop landscape heterogeneity[J].Remote Sensing Applications:Society and Environment,2024,36:101308.DOI:10.1016/j.rsase.2024.101308.
[26] Maolan K,Rusuli Y,Zhang Xuhui,et al.Sentinel-2 image based smallholder crops classification and accuracy assessment by UAV data[J].Geocarto International,2024,39(1):2361733.DOI:10.1080/10106049.2024.2361733.
[27]Li Qingting,Wang Cuizhen,Zhang Bing,et al.Object-based crop classification with Landsat-MODIS enhanced time-series data[J].Remote Sensing,2015,7(12):16091-16107.
[28] Waldner F,Diakogiannis F I.Deep learning on edge:extracting field boundaries from satellite images with a convolutional neural network[J].Remote Sensing of Environment,2020,245:111741.DOI:10.1016/j.rse.2020.111741.
[29]Dong Yong,Wang Hongyan,Zhang Yuan,et al.Accurate parcel extraction combined with multi-resolution remote sensing images based on SAM[J].Agriculture,2025,15(9):976.DOI:10.3390/agriculture15090976.
[30] Liu Xiuyu,Zhang Jinshui,Duan Yaming,et al.CPVF:vectorization of agricultural cultivation field parcels via a boundaryparcel multi-task learning network in ultra-high-resolution remote sensing images[J].ISPRS Journal of Photogrammetry and Remote Sensing,2025,226:267-299.
[31] Congalton R G,Green K.Assessing the accuracy of remotely sensed data[M].2nd ed.Boca Raton:CRC Press,2008.
[32]Yao Jinxi,Wu Ji,Xiao Chengzhi,et al. The classification method study of crops remote sensing with deep learning,machine learning,and Google Earth Engine[J].Remote Sensing,2022,14(12):2758.DOI:10.3390/rs14122758.
[33] Mazzia V,Khaliq A,Chiaberge M.Improvement in land cover and crop classification based on temporal features learning from Sentinel-2 data using recurrent-convolutional neural network(R-CNN)[J].Applied Sciences,2020,10(1):238.DOI:10.3390/app10010238.
[34]Luo Ke,Lu Linlin,Xie Yanhua,et al.Crop type mapping in thecentral part of the North China Plain using Sentinel-2 time series and machine learning[J].Computers and Electronics in Agriculture,2023,205:107577.DOI:10.1016/j.compag.2022.107577.
[35]Li Ziqian,Xuan Fu,Dong Yi,et al.Performance of GEDI data combined with Sentinel-2 images for automatic labelling of wall-to-wall corn mapping[J].International Journal of Applied Earth Observation and Geoinformation,2024,127:103643.DOI:10.1016/j.jag.2023.103643.
[36]Franch B,Bautista A S,Fita D,et al.Within-field rice yield estimation based on Sentinel-2 satellite data[J].Remote Sensing,2021,13(20):4095.DOI:10.3390/rs13204095.
[37]Pan Yaozhong,Li Le,Zhang Jinshui,et al.Winter wheat area estimation from MODIS-EVI time series data using the crop proportion phenology index[J].Remote Sensing of Environment,2012,119:232-242.
[38]Belgiu M,Bijker W,Csillik O,et al.Phenology-based sample generation for supervised crop type classification[J].International Journal of Applied Earth Observation and Geoinformation,2021,95:102264.DOI:10.1016/j.jag.2020.102264.
[39]Sun Chen,Zhou Jing,Ma Yuchi,et al. A review of remote sensing for potato traits characterization in precision agriculture[J].Frontiers in Plant Science Sustainable and Intelligent Phytoprotection,2022,13:871859.DOI:10.3389/fpls.2022.871859.
[40]Li Qiao,Fu Xueliang,Li Honghui,et al. Advancing county-level potato cultivation area extraction:a novel approach utilizing multi-source remote sensing imagery and the Shapley additive explanations-sequential forward selection-random forest model[J].Agriculture,202 5,15(1):92.DOI:10.3 390/agriculture15010092.
[41]Gardner A S,Gaston K J,Maclean I M D.Combining qualitative and quantitative methodology to assess prospects for novel crops in a warming climate[J].Agricultural Systems,2021,190:103083.DOI:10.1016/j.agsy.2021.103083.
[42]云南省人民政府.云南省国民经济和社会发展第十四个五年规划和二〇三五年远景目标纲要[EB/OL].(2021-02-09)[2026-01-10].https://www.yn.gov.cn/zwgk/zcwj/yzf/202102/t20210209_217052.html.
[43]周兴王,杨琦,田学礼,等.高寒山区马铃薯套种玉米高产高效配套栽培技术[J].大麦与谷类科学,2014(3):36-37.
基本信息:
中图分类号:TP751;S127
引用信息:
[1]汪妮,罗菊花,张春雨,等.基于GF-1与Sentinel-2的地块级作物类型精细识别方法[J].地理与地理信息科学().
基金信息:
国家自然科学基金面上项目（42571445）
2026-09-24
2026-09-24
2026-09-24