[1]顾余庆,李晓文,曹伟,等.基于Segformer网络的地块尺度作物种植结构精细化识别与分类[J].江苏农业学报,2024,(02):293-302.[doi:doi:10.3969/j.issn.1000-4440.2024.02.011]
　GU Yu-qing,LI Xiao-wen,CAO Wei,et al.Refined identification and classification of crop planting structure at plot scale based on Segformer network[J].,2024,(02):293-302.[doi:doi:10.3969/j.issn.1000-4440.2024.02.011]
点击复制

基于Segformer网络的地块尺度作物种植结构精细化识别与分类()
分享到：

江苏农业学报[ISSN:1006-6977/CN:61-1281/TN]

卷:
期数:
2024年02期
页码:
293-302
栏目:
农业信息工程
出版日期:
2024-02-25

文章信息/Info

Title:
Refined identification and classification of crop planting structure at plot scale based on Segformer network
作者:
顾余庆1; 李晓文2; 3; 曹伟1; 王亚华2; 3; 周鑫鑫3; 4; 赵碧1
(1.南京国图信息产业有限公司,江苏南京210036;2.南京师范大学地理科学学院,江苏南京210023;3.南京师范大学虚拟地理环境教育部重点实验室,江苏南京210023;4.南京邮电大学地理与生物信息学院,江苏南京210023)
Author(s):
GU Yu-qing1; LI Xiao-wen2; 3; CAO Wei1; WANG Ya-hua2; 3; ZHOU Xin-xin3; 4; ZHAO Bi1
(1.Nanjing Guotu Information Industry Co., Ltd., Nanjing 210036, China;2.School of Geography, Nanjing Normal University, Nanjing 210023, China;3.Key Laboratory of Virtual Geographic Environment, Nanjing Normal University, Ministry of Education, Nanjing 210023, China;4.School of Geographic and Biologic Information, Nanjing University of Posts and Telecommunications, Nanjing 210023, China)
关键词:
种植制度; 地块尺度; 精细化识别和分类; 遥感
Keywords:
planting system; plot scale; refined identification and classification; remote sensing
分类号:
S127
DOI:
doi:10.3969/j.issn.1000-4440.2024.02.011
摘要:
防止耕地“非粮化”、稳定粮食生产是中国粮食安全的基石。为实现地块破碎化地区作物类型及种植结构精细化识别和分类,本研究以江苏省泰兴市为研究区,基于高分辨率遥感影像和多尺度融合特征显著的Segformer语义分割模型,实现地块尺度的耕地信息精细化提取;同时结合多源遥感数据构建主要植被类型归一化植被指数（NDVI）时序曲线及植被生长关键时间节点的光谱反射特征,开展地块尺度的作物种植结构分类。结果表明:基于Segformer模型的分割方法可有效识别耕地,F1系数达92.4%;基于主要植被类型多时相NDVI时序特征及植被生长关键时间节点光谱反射特征的作物种植结构分类方法能够实现地块尺度的种植结构分类,总体分类精度达82.38%。因此,本研究建立的方法可有效实现地块尺度耕地信息的精细化提取及种植结构识别和分类,为耕地保护提供技术支持。
Abstract:
Preventing the “non-grain” of cultivated land and stabilizing food production are the cornerstones of China’s food security. In order to realize the fine identification and classification of crop types and planting structure in the area of land fragmentation, this study took Taixing City, Jiangsu province as the research area, and realized the fine extraction of cultivated land information at the plot scale based on the high-resolution remote sensing images and the Segformer semantic segmentation model with significant multi-spatial scale fusion features. At the same time, the normalized difference vegetation index (NDVI) time series curve of the main vegetation types and the spectral reflectance characteristics at the key time nodes of vegetation growth were constructed by combining multi-source remote sensing data, and the classification of crop planting structure at the plot scale was carried out. The results showed that the segmentation method based on Segformer model could effectively identify cultivated land, and the F1 was 92.4%. The classification method of crop planting structure based on multi-temporal NDVI time series characteristics of main vegetation types and spectral reflection characteristics at key time nodes of vegetation growth could realize the classification of planting structure at plot scale, and the overall classification accuracy was 82.38%. Therefore, the method established in this study could effectively realize the fine extraction of cultivated land information at the plot scale and the identification and classification of planting structure, and provide technical support for cultivated land protection.

参考文献/References:

[1]胡琼,吴文斌,宋茜,等. 农作物种植结构遥感提取研究进展[J]. 中国农业科学,2015,48(10):1900-1914.
[2]SALMON J M, FRIEDL M A, FROLKING S, et al. Global rain-fed,irrigated, and paddy croplands:a new high resolution mapderived from remote sensing, crop inventories and climate data[J]. International Journal of Applied Earth Observation & Geoinformation,2015,38:321-334.
[3]陈仲新,任建强,唐华俊,等. 农业遥感研究应用进展与展望[J]. 遥感学报,2016,20(5):748-767.
[4]刘焕军,邱政超,孟令华,等. 黑土区田块尺度遥感精准管理分区[J]. 遥感学报,2017,21(3):470-478.
[5]刘哲,刘帝佑,朱德海,等. 作物遥感精细识别与自动制图研究进展与展望[J]. 农业机械学报,2018,49(12):1-12.
[6]WU M, HUANG W, NIU Z, et al. Fine crop mapping by combining high spectral and high spatial resolution remote sensing data in complex heterogeneous areas[J]. Computers and Electronics in Agriculture,2017,139:1-9.
[7]YU B, SHANG S. Multi-year mapping of maize and sunflower in Hetao irrigation district of China with high spatial and temporal resolution vegetation index series[J]. Remote Sensing,2017,9(8):855.
[8]杨颖频,吴志峰,骆剑承,等. 时空协同的地块尺度作物分布遥感提取[J]. 农业工程学报,2021,37(7):166-174.
[9]张鹏,胡守庚. 地块尺度的复杂种植区作物遥感精细分类[J]. 农业工程学报,2019,35(20):125-134.
[10]ZHENG Y, WU B, ZHANG M, et al. Crop phenology detection using high spatio-temporal resolution data fused from SPOT5 and MODIS products[J]. Sensors,2016,16(12):2099.
[11]GAO F, ANDERSON M C, ZHANG X, et al. Toward mapping crop progress at field scales through fusion of Landsat and MODIS imagery[J]. Remote Sensing of Environment,2017,188:9-25.
[12]HAO P, ZHAN Y, WANG L, et al. Feature selection of time series MODIS data for early crop classification using random forest:a case study in Kansas, USA[J]. Remote Sensing,2015,7(5):5347-5369.
[13]王娜,李强子,杜鑫,等. 单变量特征选择的苏北地区主要农作物遥感识别[J]. 遥感学报,2017,21(4):519-530.
[14]SONOBE R, TANI H, WANG X, et al. Parameter tuning in the support vector machine and random forest and their performances in cross-and same-year crop classification using TerraSAR-X[J]. International Journal of Remote Sensing,2014,35(23):7898-7909.
[15]PELLETIER C, VALERO S, INGLADA J, et al. Assessing the robustness of Random Forests to map land cover with high resolution satellite image time series over large areas[J]. Remote Sensing of Environment,2016,187:156-168.
[16]DONG J, XIAO X, MENARGUEZ M A, et al. Mapping paddy rice planting area in northeastern Asia with Landsat 8 images, phenology-based algorithm and Google Earth Engine[J]. Remote Sensing of Environment,2016,185:142-154.
[17]邓刘洋,沈占锋,柯映明,等. 基于地块尺度多时相遥感影像的冬小麦种植面积提取[J]. 农业工程学报,2018,34(21):157-164.
[18]田海峰,邬明权,牛铮,等. 基于Radarsat-2 影像的复杂种植结构下旱地作物识别[J]. 农业工程学报,2015,31(23):154-159.
[19]黄启厅,覃泽林,曾志康. 多星数据协同的地块尺度作物分类与面积估算方法研究[J]. 地球信息科学学报,2016,18(5):708-717.
[20]韩衍欣,蒙继华. 面向地块的农作物遥感分类研究进展[J]. 国土资源遥感,2019,31(2):1-9.
[21]肖艳,姜琦刚,王斌,等. 基于ReliefF 和PSO 混合特征选择的面向对象土地利用分类[J]. 农业工程学报,2016,32(4):211-216.
[22]XIE E, WANG W, YU Z, et al. SegFormer:simple and efficient design for semantic segmentation with transformers[J]. Advances in Neural Information Processing Systems,2021,34:12077-12090.

相似文献/References:

[1]周炜,张岳芳,朱普平,等.种植制度对长江下游稻田温室气体排放的影响[J].江苏农业学报,2017,(02):340.[doi:doi:10.3969/j.issn.1000-4440.2017.02.016]
　ZHOU Wei,ZHANG Yue-fang,ZHU Pu-ping,et al.Effects of different cropping patterns on greenhouse gases emissions from rice fields in the lower reaches of Yangtze River[J].,2017,(02):340.[doi:doi:10.3969/j.issn.1000-4440.2017.02.016]

备注/Memo

备注/Memo:
收稿日期：2023-11-16基金项目：国家自然科学基金项目(42201504、41971404);江苏省自然资源厅科技项目（2021013）作者简介：顾余庆（1979-）,男,硕士,高工,主要从事自然资源保护利用和规划研究。（E-mail）guyuqing@gtmap.cn通讯作者：赵碧,（E-mail）zhaobi1929@163.com
更新日期/Last Update: 2024-03-17