摘要
为探索实现作物虫害自动监测的方法,采用图像处理和光谱分析技术,测定了斑潜蝇虫害叶片的近红外反射光谱,计算了虫害叶片的破损率,对其破损率和干鲜比与近红外分光反射率的关系分别进行了回归分析。结果表明:在某些波段,叶片的破损率和干鲜比均与近红外分光反射率有较好的相关性。叶片的干鲜比与近红外分光反射率关系的决定系数:黄瓜为R^2=0.79(在1452nm),番茄为R^2=0.70(在1450nm)。叶片的破损率与近红外分光反射率关系的决定系数,黄瓜为R^2〉0.81(在1436~1468nm),番茄为R^2〉0.69(在1436~1466nm)。试验和分析结果证明斑潜蝇虫害叶片的虫害程度能很好地被近红外光谱信息反映。
In order to find out an automatic monitoring method of plant insect pests, leaves infected by the leafminer were studied. Using image processing and spectroscopy analysis technology, near-infrared spectral reflectance of leaves infected by the leafminer was measured ; damaged degrees (DD) of leaves and the ratios of the dry and fresh leaf weights(DFLW) were calculated. The regression results between DFLW and the near-infrared spectral reflectance are as follows : R^2 value between cucumber leaves and the near-infrared spectral reflectance is 0.79 (at 1452 nm) and the tomatoes' is 0.70(at 1450 nm); the R^2 value between the cucumber's DD and the near-infrared spectral reflectance is beyond 0.81 (at 1436 ~ 1466 nm), tomato's is beyond 0. 69 (at 1436~ 1468 nm), respectively. The experiment and analysis results show that the leaf damaged degrees can be reflected by the near-infrared spectral reflectance of leafminer-infected leaves.
出处
《农业工程学报》
EI
CAS
CSCD
北大核心
2007年第2期156-159,共4页
Transactions of the Chinese Society of Agricultural Engineering
基金
国家自然科学基金项目(60374030)
北京市教育委员会共建项目建设计划资助(XK100190550)
关键词
美洲斑潜蝇
近红外反射光谱
叶片
图像处理
leafminer
near-infrared spectral reflectance
leaf
image processing