| 田间冬小麦抽穗期长势分析——基于可见-近红外光 |
| |
| 引用本文： | 刘仁杰,房俊龙,李民赞,孙红,吴李烜,赵毅,张猛.田间冬小麦抽穗期长势分析——基于可见-近红外光[J].农机化研究,2016(4):141-146. |
| |
| 作者姓名： | 刘仁杰 房俊龙 李民赞 孙红 吴李烜 赵毅 张猛 |
| |
| 作者单位： | 1. 东北农业大学电气与信息学院，哈尔滨 150030; 中国农业大学现代精细农业系统集成研究教育部重点实验室，北京 100083;2. 东北农业大学电气与信息学院,哈尔滨,150030;3. 中国农业大学现代精细农业系统集成研究教育部重点实验室,北京,100083 |
| |
| 基金项目： | “十二五”国家科技支撑计划重大项目 |
| |
| 摘 要： | 为了快速估测大田冬小麦叶绿素含量指标,指导冬小麦抽穗期追肥管理,基于光谱分析技术在可见光和近红外波段(325~1 075 nm)处,对陕西省杨凌区揉谷镇粮食基地的冬小麦进行长势检测、分析。试验在1 000 m×600 m区域内划分为30个采样区进行数据采集,使用ASD Field Spec Hand Held光谱辐射仪(Analytical Spectral Devices.,USA)采集冬小麦的冠层光谱反射率数据,使用SPAD-5 0 2 Plus便携式叶绿素仪测量小麦倒一叶和倒二叶的叶绿素指标(SPAD值),使用G738 CM型手持式GPS记录采样点的位置信息。分别进行冠层光谱反射率小麦倒一叶和倒二叶的预处理,结果表明:冠层反射光谱倒二叶的SPAD值相关系数高于倒一叶。基于相关性分析,选取4个敏感波段538、661、740、850 nm分别与预处理前后的光谱数据进行多元线性回归分析,结果表明:预处理后的模型精度较高,建模精度R2=0.8 3,验证建模精度R2=0.7。同时,绘制了大田作物长势分布图,可为冬小麦抽穗期追肥提供支持。
|
| 关 键 词： | 冬小麦 无损检测 叶绿素指标 可见近红外光谱 |
Analysis of Winter Wheat Heading Fields Growing-Based on Vis/NIR Spectroscopy |
| |
| Abstract: | In order to estimate the chlorophyll content indicators of field winter wheat in real -time and guide fertilizer management in heading period .The detection and analysis was conducted in Yang Ling of Shan Xi province .The test area of winter wheat was 1000 ×600 m2 and it was divided into 30 sampling plots to collect data .The spectral technology was applied .The Analytical Spectral Devices Field Spec HandHeld ( USA ) spectral radiometer was used to collect spectral reflectance data of winter wheat .The visible and NIR band ( 325 ~1050 nm ) reflectance of winter wheat canopy was measured.The portable chlorophyll meter (SPAD-502plus) was applied to measure SPAD index of the pour wheat leaf and the second countdown leaf of each plant .And handheld GPS (G738CM) was used to record the location of sampling points .The preprocessing between the reflectance and the wheat leaf SPAD index was analyzed .It was showed that the relationship between spectral data the second countdown leaves SPAD values higher than the pour leaves .In addition , four sensitive wavelengths were selected at the 538、661、740、850 nm based on correlation analysis .And four sensitive wavelengths respectively proceed to multiple linear regression analysis for the before and after preprocessing .The result showed that the after preprocessing modeling accuracy was 0.83 and validation accuracy was 0.7.The distribution map was drawn by GPS coordinates and modeling prediction result .With the application of spectral technology , it provides a feasible method to detect the winter wheat growth status at heading stage . |
| |
| Keywords: | winter wheat non-destructive measurement chlorophyll index visible/near infrared spectra |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
