广西甘蔗地冬季黏虫发生调查初报

于2014—2018年冬季,在广西桂南地区、桂中、桂西和桂东南地区选择不同类型甘蔗地,调查黏虫发生及为害情况。结果表明,新植蔗和收割后焚烧蔗叶的宿根蔗地极少发现黏虫,而收割后蔗叶(直接或粉碎)还田的宿根蔗地发现有不同龄期的幼虫为害,老蔗地亦发现老熟幼虫或蛹,但虫口密度低。数据显示,广西桂南地区冬季甘蔗地黏虫虫口密度最大。其中,2017年南宁市武鸣区虫口密度最高,达8.9头/m^2,其次是桂东南地区,桂中地区和桂西地区极少或未发现黏虫。     

基于深度学习的病虫害智能化识别系统

我国农作物种植覆盖面广、分散度高,病虫害发生种类多、区域性发生规律复杂,传统的人工鉴定技术从效率、能力与精度方面均难以满足新形势下重大病虫测报要求。针对这一实践需求,以测报灯下害虫图像数据库(约18万张)、田间病虫害图像数据库(约32万张)为基础,构建了基于深度学习方法的病虫害种类特征自动学习、特征融合、识别和位置回归计算框架,并研发了移动式病虫害智能化感知设备和自动识别系统。通过近2年的精确度和实操运行效率检验,该系统在自然状态下对16种灯下常见害虫的识别率为66%～90%,对38种田间常见病虫害(症状)的识别率为50%～90%。随基础数据库的不断丰富、神经网络深层特征提取的不断完善,该系统有望进一步提高识别准确率,从而真正实现田间病虫害识别自动化、智能化和高效率。     

Cover Image,Volume 75, Issue 6

The cover image is based on Research Article Field evaluation of an unmanned aerial vehicle (UAV) sprayer: effect of spray volume on deposition and the control of pests and disease in wheat by Guobin Wang et al., DOI: 10.1002/ps.5321 . Illustration Credit: Guobin Wang, Yubin Lan, and Yuxing Han.

     

First report of Pseudomonas aeruginosa causing tumor disease of Populus koreana in China

Journal of Plant Diseases and Protection - In May 2018, a bacterial tumor disease was observed on Populus koreana in Shuangyang county, Changchun city, Jilin Province. Based on physiological...     

First report of Aplosporella javeedii causing branch blight disease of Mulberry (Morus alba) in China

Journal of Plant Diseases and Protection - Mulberry has tremendous economic importance and cultivated widely across China. In June 2018, branch blight disease was found on mulberry (Morus alba) in...