基于卷积神经网络的玉米病害识别方法研究

为解决传统的玉米病害识别方法中特征提取主观性强及误识率高的问题,提出利用卷积神经网络对玉米病害进行识别。以玉米病害图像和健康图像共5种类别的玉米图像为研究对象,并采用LeNet模型进行试验。首先,按照8∶2的比例为每种玉米病害图像选择训练集和测试集。然后,通过试验组合和对比分析的方法比较不同卷积神经网络结构设置对准确率的影响,选出最佳参数。另外,选用Adam算法代替SGD算法来优化模型,通过指数衰减法调整学习率,将L2正则项添加到交叉熵函数中,并选择Dropout策略和ReLU激励函数。最后,确定了一个10层CNN网络结构。试验结果显示,玉米花叶病、灰斑病、锈病、叶斑病和玉米健康识别率分别为95.83%、90.57%、100%、93.75%、100%,平均识别率达96%,平均计算时间为0.15 s。经试验结果比较,该模型识别效果明显高于传统方法,为玉米病害的防治提供技术支持。     

资本禀赋、技术认知与农户耕地低碳利用意愿

耕地低碳利用对于减少农业温室气体排放,促进农业持续健康发展具有积极意义。建立资本禀赋、技术认知的理论分析模型和研究假说,利用湖北省347户农户的调研数据,在测度农户资本禀赋、耕地低碳利用技术认知的基础上,采用Logit模型、中介效应模型、调节效应检验方法,就资本禀赋、技术认知与农户耕地低碳利用意愿的关系进行实证研究。结果发现,资本禀赋不仅对农户耕地低碳利用意愿在1%水平上有显著的直接正向影响,还会通过技术认知这一中介变量在1%水平上产生显著的间接正向影响,环境素养在农户自我效能认知和耕地低碳利用意愿间发挥正向调节作用。据此,建议从经济资本、社会资本、文化资本3个角度提升农户的资本禀赋,通过多种渠道提升农户对耕地低碳利用的认知水平,提高农户环境素养。     

基于机器视觉的温室番茄裂果检测

该文通过对温室番茄果实进行定位及裂果检测,可为番茄裂果率预估及后续裂果自动筛选提供参考。针对自然光照下采集的各类番茄图像,在相关颜色空间中进行阈值预分割,利用前期支持向量机训练得到的纹理特征分类器对预分割区域进行二次判别;之后在前景区域利用显著性角点分割构造边缘轮廓集,利用基于最小二乘法修正的改进霍夫变换拟合单个番茄目标;最后利用二维Gabor小波算子对拟合的单个番茄区域进行纹理特征提取及裂果判别。文中共采集82幅番茄图像,其中50幅图像作为训练集图像,32幅图像作为验证集,所提算法对测试集中总共128个番茄的果实正确检出率为91.41%,对其中35裂果的正确判别率为97.14%,裂果判别部分平均耗时21 ms。试验结果表明,该方法具有较好的鲁棒性与可靠性,对成熟期番茄裂果率的估计研究及采摘过程中裂果的自动分级筛选具有较好的指导意义,为未来实现温室番茄果实生长状态在线监测提供参考。     

Cotton Image Segmentation Algorithm Based on Fusion Method of Markov Random Field and Quantum Particle Swarm Cluster

[Objective] The aim of this study was to improve the cotton image segmentation accuracy in a picking robot image processing system. [Method] An image segmentation algorithm based on a fusion method of Markov random field and quantum particle swarm optimization clustering was proposed. The process of the proposed algorithm is as follows: first, transform the RGB (red, green, blue) images into grayscale; second, use it to segment these images; finally, the threshold of the connected area is set on the basis of the segmented image to obtain the target area. Then, the cotton front image and the cotton side image are selected from the images collected from different angles. The segmentation experiment was carried out by using this algorithm, and compared with the Otsu algorithm, the fuzzy C-means algorithm, the quantum particle swarm image segmentation algorithm and the Markov random field image segmentation algorithm. [Result] The results showed that the segmentation accuracy and peak signal to noise ratio of the proposed algorithm were 98.94% and 77.48 dB. When compared with the Otsu algorithm, fuzzy C-means algorithm, quantum particle swarm optimization algorithm and Markov random field algorithm, the average segmentation accuracy and peak signal to noise ratio of the proposed algorithm increased by 2.47%–4.56%, and 9.81–13.11 dB, respectively. [Conclusion] The proposed algorithm had higher segmentation accuracy and higher peak signal to noise ratio than the other algorithms tested.     

一株耐铝甜瓜枯萎病拮抗细菌的分离鉴定及其耐铝特性研究

从甜瓜根际酸性土壤中分离筛选到1株耐铝甜瓜枯萎病拮抗菌，命名为A2。根据表型、生理生化特征和16S rRNA基因序列相似性分析，将其鉴定为Pseudomonas sp.。菌株A2对甜瓜枯萎病病原菌的相对防效为68.3%，且拮抗能力具有遗传稳定性。相比AlCl3处理，菌株A2对Al2(SO4)3表现出了更好的耐受性，最高可耐受50mmol/L Al3+。在含有A13+的S-LB培养基中，菌株最适生长温度为30 ℃；培养基初始pH值的降低会加剧A13+对菌株A2的毒害作用。研究结果可为含活性铝的酸性土壤中甜瓜枯萎病的生物防治提供优良菌株资源和理论基础。     

Phenotypic plasticity of two invasive alien plant species inside a deciduous forest in a strict nature reserve in Italy

ABSTRACT

Invasive alien plant species (IAPs) represent one of the main biological threats to biodiversity worldwide. Information about their phenotypic plasticity are needed to increase awareness about their future invasive potential. A study about phenotypic plasticity in response to contrasting light regimes and its quantification by a plasticity index (PI) of two IAPs (Ailanthus altissima and Robinia pseudoacacia) inside a Strict Nature Reserve was conducted. R. pseudoacacia showed a 70% higher PI, with a strongly greater value at morphological leaf level, associated with a greater ability to survive and grow in forest understory, explaining its greater widespread. Otherwise, A. altissima showed its highest PI at physiological level, which was associated with the ability to colonize and grow in environments with high-light regimes. Based on these results, the conservative management has limited the presence of A. altissima by its lower ability to grow in forest understory. In fact, the small-scale gaps in the forest infrastructure, that could allow its recruitment, are originated only from the death of a single tree or small group of trees. Regarding R. pseudoacacia, it is critical to maintain this type of management because any disturbances resulting in large openings could further promote its presence inside the Reserve.     

长期施肥处理对双季晚稻叶绿素荧光特征及籽粒产量的影响

研究施肥模式对水稻叶绿素含量及荧光参数的调节作用,对提高水稻光能利用效率及籽粒产量具有重要意义。本研究以水稻品种赣929为试验材料,比较长期定位施磷钾肥(PK)、施氮磷钾肥(NPK)、等养分条件下70%氮磷钾肥配合施用30%有机肥(70F+30M)、50%氮磷钾肥配合施用50%有机肥(50F+50M)、30%氮磷钾肥配合施用70%有机肥(30F+70M)条件下水稻叶绿素荧光参数和产量变化特征。结果表明,水稻旗叶叶绿素含量从齐穗期到成熟期呈下降趋势,且表现为30F+70M50F+50M70F+30MNPKPK。水稻叶绿素荧光参数Fv/Fm随水稻生育期的推进呈降低趋势,其中PK处理Fv/Fm值最低,而施氮处理比不施氮处理Fv/Fm值增加了2.85%～4.18%。叶绿素荧光参数qL、ΦPSII在NPK处理中表现为齐穗期较高, 20 d之后显著降低;在50F+50M、30F+70M处理中表现为齐穗期较低, 20 d以后显著增加;在PK处理、70F+30M处理中一直处于较高水平。NPQ变化趋势与qL基本相反。ETR-PAR光响应曲线拟合结果表明, 70F+30M处理ETRmax、?和Ek值在齐穗期和20 d后值均最高。综合看来,等养分条件下配施30%有机肥具有最优的叶绿素荧光指标组合及籽粒产量;而配施超过50%有机肥由于前期热耗散增大使得用于光合作用的光能份额减少,而施用氮磷钾肥处理则由于后期的衰老使得光能利用效率下降。     

Hydrodynamic method for estimating production carrying capacity of coastal finfish cage aquaculture in Southeast Asia

This paper presents the development of a simple and generally applicable hydrodynamic method for the estimation of production carrying capacity (PCC) of coastal finfish cage aquaculture. Dimensional analysis was used to find significant and general interdependencies between the hydrodynamics at fish farm locations and particulate wastes deposited on the seafloor by fish farms. Modeled ratios of deposition to emission of particulate wastes underneath fish farms were found to be primarily a function of the flow Reynolds numbers at the farming locations and the non-dimensional settling velocity of emitted wastes. In the non-dimensional model, farming conditions include daily feed rate, proportion of unconsumed feed, and carbon content in feed and fish feces. The relationship can be used to estimate the PCC of floating net cages imposing a threshold value for deposition. Results of in-situ assessments of the benthic impacts of several fish farms in an aquaculture site in the northwest of Bali, Indonesia were used to validate and demonstrate the effectiveness of the method. Predicted results were able to clearly identify fish farms operating beyond ecologically sustainable carrying capacity. The proposed method has broad applicability and could help make decisions regarding the estimation of production potential of individual farms in pristine areas, for providing first estimates in sites that have scarce data, and for assessment, expansion, and optimization of the currently operating aquaculture sites in Southeast Asia, China and potentially other data-poor island nations. As the method relies on dynamic models, it enables straightforward assessments over the entire aquaculture region.     

2018年进口奶牛检疫情况分析

随着十多年的超速发展,我国奶牛业已由单一品种的散养方式逐步转变为高产品种的规模化、集约化养殖方式。然而,我国奶牛整体遗传素质不高,多为进口高产奶牛与地方奶牛的低代次改良品种,以致原奶质量参差不齐,严重制约了我国奶业的健康发展。因此,大规模引进高品质的奶牛品种,已成为我国优化奶牛基因、推动畜牧业产业化发展的重要抓手,但同时也为检疫工作带来了挑战。本文通过分析2018年我国进口奶牛数量、来源国、疫病检出、进境口岸及隔离检疫场分布等数据,结合各来源国的疫病防控特点,有针对性地提出检疫要点,为我国全面提升进口奶牛的检疫水平、切实做好奶牛群体遗传改良计划、加快推进畜牧业转型升级提供参考。     

DG/T 082-2019《粪污固液分离机》推广鉴定大纲解读

推广鉴定大纲是进行农机鉴定工作时的技术要求,是推广鉴定项目实施的最主要依据。为了加大对推广鉴定大纲的宣传,便于农机生产企业、用户、管理部门和鉴定机构理解与使用,本文对DG/T 082-2019《粪污固液分离机》推广鉴定大纲中推广鉴定的产品型式、鉴定前的准备工作、鉴定内容和鉴定结论4个方面的主要内容进行了解读。