摘要
水稻灯诱害虫的识别与计数在水稻田间害虫监测中是非常重要的。由于水稻害虫被黑光灯诱集后姿态各异,存在虫体残缺现象,增加了图像自动识别的难度。在获取水稻灯诱害虫非粘连图像基础上,利用模板匹配和K折交叉验证方法进行多目标水稻灯诱害虫的识别。首先,提取每个水稻害虫图像中包括颜色、形态和纹理共156个特征参数;然后,利用主成分分析法进行数据降维,选取前6个主成分作为害虫特征参数;最后,根据每种灯诱害虫的姿态确定模板数,通过模糊C均值获得聚类中心作为模板参数,分别利用单模板和多模板匹配方法进行水稻害虫的识别。结果表明,针对姿态各异且有虫体残缺的多目标水稻灯诱害虫,多模板和单模板匹配法的识别率分别为83.1%和59.9%。
Identification and count of rice light-trap pests are very important in monitoring rice pests. The pests trapped by black light lamps show different postures and incomplete bodies, which increase the difficulty of image automatic identification. Template matching and K-fold cross validation methods were used to identify multi-objective rice light- trap pests based on non-touching pest images. Firstly, one hundred and fifty-six features including color, shape and texture features were extracted from each pest image. Secondly, the principal component analysis was employed for reducing data dimensionality, and first six principal components were selected as pests＇ features. Then, the template number was determined according to the gesture of each pest species and the template parameters were obtained from the cluster centers by fuzzy C-mean clustering method. Finally, single and multi-template matching methods were used to identify rice pests. The results showed that the accurate rate of multi-template matching and single template matching were 83.1% and 59.9 % respectively for rice light-trap pests with multiple postures and some incomplete bodies.
出处
《中国水稻科学》
CAS
CSCD
北大核心
2012年第5期619-623,共5页
Chinese Journal of Rice Science
基金
国家自然科学基金资助项目(31071678)
浙江省重大科技项目(2010C12026)
中央级公益性科研院所基本科研业务费专项资金资助项目(2009RG0042)
宁波市科技项目(2010C10044)
浙江理工大学研究生创新基金资助项目(YCX-S11022)
关键词
水稻灯诱害虫
害虫识别
图像处理
特征提取
模板匹配
rice light-trap pest
pest identification
image processing
feature extraction
template matching
