生姜主要病虫害及其防治技术分析

生姜的根状茎为使用部分,其中包含姜油酮、姜油醇、姜油酚、蛋白质、粗脂肪、糖,并且钙、铁的含量也较为丰富,是人们日常生活当中不可或缺的调味品,还能够加工成为姜片、糖姜片和盐姜片等诸多食品,其在医药方面的应用也较为广泛,备受人们的推崇与喜爱。我国在生姜栽培方面的历史较为悠久,技术也较为成熟,每年生姜出口贸易都能为我国带来非常可观的经济效益。     

Pathogenicity of three species of entomopathogenic nematodes to some major stored-product insect pests

Entomopathogenic nematodes (Nematoda: Heterorhabditidae and Steinernematidae) are commonly used biological control agents of insects in cryptic habitats, but their potential for suppressing stored-product insects in these habitats has not been explored previously. Here, we provide data from the first step in a program to evaluate entomopathogenic nematodes in the genus Steinernema as biological control agents of stored-product pests by determining their pathogenicity to some of the major stored-product pest species. When evaluated against larvae, pupae and adults of six pest species (Plodia interpunctella, Ephestia kuehniella, Oryzaephilus surinamensis, Tenebrio molitor, Tribolium castaneum, and Trogoderma variabile), and the adults of two additional pest species (Sitophilus oryzae and Rhyzopertha dominica), Steinernema riobrave was either the most pathogenic or of similar pathogenicity compared to S. carpocapsae and S. feltiae. A dose of 10 infective juveniles of S. riobrave caused 80% or higher mortality against larvae of P. interpunctella, E. kuehniella, T. castaneum, and O. surinamensis, pupae of T. castaneum and T. molitor, and adults of T. molitor and the two moth species. All stages of Trogoderma variabile exhibited 70% or higher mortality. Adults of S. oryzae and R. dominica exhibited low susceptibility with 15% and 35% mortality, respectively. On the basis of these results, S. riobrave was selected for further evaluation under more field-like conditions.     

基于多特征融合和SVM分类器的植物病虫害检测方法

针对农业领域植物病虫害检测问题,提出一种基于高清视频图像融合特征的支持向量机(SVM)的检测方法,实现农业生产中植物病虫害的快速检测。对每幅植物叶片图像的颜色、HSV、纹理和方向梯度直方图四种特征采用基于特征包的多特征融合方法,形成特征向量,并利用SVM分类器进行训练分类。对单特征与融合特征的SVM分类器性能进行试验比较,所提出的方法具有较高的准确率。     

森林资源亚健康状态的卫星遥感预警技术研究

森林病虫害等扰动类型造成的亚健康林木监测预警工作不能及时到位,导致防治工作长期处于灾后救灾的被动局面。基于2019年5~9月份的多时相GF-1 WFV数据,应用比值植被指数和红绿植被指数,准实时地监测逆生长、叶冠胁迫或失色等“灾害”信息。结果表明：虽然树木叶片枯黄、萎蔫等叶绿素降解并逐渐转化成叶黄素和叶红素需要一定的过程,或“灾害症状”有时具有滞后性,但高频次遥感动态监测结果对于指导森林灾害地面踏查,提高监测覆盖率和科学性,防范大面积灾害,具有积极作用。国产GF-1和GF-6 WFV遥感数据的高重访周期能为月度森林资源生长过程的监测提供坚实的数据保障,满足公顷级树叶长势退化预警监测的需要。     

基于TACNN的玉露香梨叶虫害识别

为了解决玉露香梨叶虫害种类多、扩散速度快、人工识别梨叶虫害耗时长的问题,提出了能够在自然环境下对玉露香梨叶虫害图像自动识别的Tiny-Alexnet卷积神经网络（Tiny-Alexnet Convolution Neural Network,TACNN）的虫害识别模型。分析了Alexnet模型的网络结构,并将实地采集的玉露香叶片虫害图像进行统一处理,为避免全连接层卷积核参数过大而产生的过拟合现象,通过优化全连接层,设置不同神经元节点和实验参数,得出了Mid-Alexnet、TACNN两种虫害识别模型。实验结果表明：TACNN较Alexnet和Mid-Alexnet模型有较高的识别准确率,该模型能够有效地提取梨叶虫害特征,类别平均准确率为81.18%,实现了对金龟子、梨木虱、梨瘿蚊三种虫害的准确区分。该模型在玉露香梨叶虫害识别方面具有良好的性能,可实现自然环境下玉露香梨叶虫害的精准识别。     

一种基于YOLOv4-TIA的林业害虫实时检测方法

针对现有基于深度学习的林业昆虫图像检测方法存在检测精度低和检测速度慢的问题,提出一种结合改进PANet结构与三分支注意力机制的目标检测方法YOLOv4-TIA。通过对样本数量较少的昆虫类别进行数据增强,实现样本均衡分布。利用三分支注意力机制改进YOLOv4中的CSPDarkNet53骨干网络,同时通过旋转操作和残差变换建立维度间的依存关系,以提高有效的特征通道权重,在PANet结构上增加将跳跃连接与跨尺度连接相结合的特征融合方式,从而获取更丰富的语义信息和位置信息。在此基础上,采用Focal loss函数优化分类损失,解决正负样本不均衡的问题。实验结果表明,该方法的精确率和召回率分别达到85.9%和91.2%,相比SSD、Faster R-CNN、YOLOv4方法,其在保证检测速度的同时,能够有效提高检测精度,且实现对林业害虫的实时精确监测。     

基于计算机视觉的果树害虫的形态特征分析

为解决农作物病虫害的诊断问题,选择京郊常见果树的三种害虫:桃蛀螟成虫、华北大黑鳃金龟成虫、桃六点天蛾,使用计算机视觉技术清除背景,获取该害虫的形状参数、偏心率等形态特征。通过判别分析,建立三种害虫形态特征的判别评分函数,并测试了该判别评分函数的有效性。实验结果表明,形状参数、偏心率、似圆度特征能反映这三种害虫的形态特点,可有效区分这三种害虫。     

知识管理框架ReKM在农业病虫害诊断中的应用

在农业病虫害诊断领域，传统的专家系统往往在设计时就硬编码好了知识结构，在不重新改动代码的情况下，具有很差的扩展性。介绍了一个基于本体和案例推理的可重构知识管理框架ReKM。该框架利用本体作为知识结构的描述，由知识存储层、本体层和应用层组成。利用本体模型的可重构特性，一定程度上克服了传统专家系统可扩展性不足的缺点。文中详细阐述了框架中基于案例的检索算法和简单推理的概念。最后介绍了基于该框架开发的一个系统在某农科院病虫害诊断领域的应用情况和今后研究的方向。     

Trehalose and proline failed to enhance cold tolerance of the cowpea weevil,Callosobruchus maculatus (F.) (Col.: Bruchidae)

The cowpea weevil, Callosobruchus maculatus (F.) (Col.: Bruchidae) is a cosmopolitan field-to-store pest ranked as the major post-harvest pest of cowpea in tropical regions. The cold tolerance of an insect species can vary as a result of abiotic features including food resources. In this study, C. maculatus larvae were fed with proline and trehalose (10, 20, and 40 mmol) treated cowpea seeds to determine the effects of these potential cryoprotectants on the supercooling (SCP) and cold hardiness of the upcoming adult beetles. The SCPs of the control, proline-fed and trehalose-fed adults non-significantly changed from −18.2 °C for the control to −17.2 °C for trehalose-fed adults. The cold hardiness (24 h at 0, -5.0, −7.5, −10.0, and −12.5 °C) of the adults was almost the same for control and treatments. Median lethal times (LT₅₀; lethal time for 50% mortality) were 6.3, 6.0, and 5.4 h, respectively. Moreover, feeding the larvae with proline and trehalose-treated seeds did not affect the proline and trehalose contents of the adult beetles. Our results showed that C. maculatus could not tolerate subzero temperatures well above their SCP, indicating that this species might be a chill-susceptible insect.     

关于“储粮害虫”与“仓库害虫”名称的商榷

本文从实际和学科的发展两方面论证了采用“仓库害虫”名称的原因和优点。