摘要
[目的]针对水稻虫害发生程度预警和田间稻飞虱虫害动态监测问题,采用光谱检测技术,运用主成分回归法建立稻飞虱百穴虫量检测模型。[方法]使用FieldSpec3光谱仪在稻田采集冠层光谱数据,同步检测地块稻飞虱百穴虫量,样本量为71个,其中51个样本为校正集,20个样本为预测集。建模波段选择350～1139nm,对原始光谱进行一阶微分处理。[结果]测量值与预测值的相关系数为0.78,预测标准偏差为161头。[结论]光谱检测可用于稻飞虱测报工作。
[Objective]Aiming at problems of early warning for occurrence of rice insect pests and dynamic monitoring of rice planthoppers in field,a detection model for rice planthopper populations was established based on PCR with spectrum detection technology. [Method]Canopy reflectance data were collected using FieldSpec 3 spectrometer in paddy field,and rice planthoppers populations were detected simultaneously. The sample size was 71,and there were 51 samples in the calibration set and 20 samples in the prediction set. The modeling band was 350-1 139 nm,and the original spectra were pretreated by first order differential. [Result]The correlation coefficient of measured values and predictive values was 0.78,and the RMSEP was 161. [Conclusion]Spectrum detection was able to be used in rice planthoppers investigation and forecasting.
出处
《安徽农业科学》
CAS
北大核心
2010年第23期12695-12696,共2页
Journal of Anhui Agricultural Sciences
基金
江苏大学现代农业装备与技术省部共建教育部重点实验室开放基金项目(NZ200803)
关键词
稻飞虱
光谱
主成分回归
虫害
水稻
Rice planthopper
Spectrum
Principal component regression （PCR）
Insect pest
Rice
