融合2D-3D卷积特征识别哺乳母猪姿态转换

母猪姿态转换影响仔猪存活率,且动作幅度与持续时间存在差异,准确识别难度大。该研究提出一种融合2D-3D卷积特征的卷积网络(2D+3D-CNet,2D+3D Convolutional Network)识别深度图像母猪姿态转换。以视频段为输入,引入注意力机制SE模块和3D空洞卷积,以提升3D卷积网络姿态转换的时空特征提取能力,用2D卷积提取母猪的空间特征;经特征融合后,动作识别分支输出母猪转换概率,姿态分类分支输出4类姿态概率,结合这两个输出结果识别8类姿态转换,减少了人工标注数据集的工作量;最后设计动作分数,优化母猪姿态转换的时间定位。在测试集上,2D+3D-CNet姿态转换识别精度为97.95%、召回率为91.67%、测试速度为14.39帧/s,精度、召回率和时间定位精度均高于YOWO、FRCNN-HMM和MOC-D方法。该研究结果实现了母猪姿态转换高精度识别。     

基于隐马尔科夫模型的深度视频哺乳母猪高危动作识别

哺乳母猪的高危动作和仔猪存活率有密切关系,能直接体现其母性行为能力,而这些高危动作又与其姿态转换的频率、持续时间等密切相关。针对猪舍环境下,环境光线变化、母猪与仔猪黏连、猪体形变等给哺乳母猪姿态转换识别带来的困难。该文以梅花母猪为研究对象,以Kinect2.0采集的深度视频图像为数据源,提出基于Faster R-CNN和隐马尔科夫模型的哺乳母猪姿态转换识别算法,通过FasterR-CNN产生候选区域,并采用维特比算法构建定位管道;利用Otsu分割和形态学处理提取疑似转换片段中母猪躯干部、尾部和身体上下两侧的高度序列,由隐马尔科夫模型识别姿态转换。结果表明,对姿态转换片段识别的精度为93.67%、召回率为87.84%。研究结果可为全天候母猪行为自动识别提供技术参考。     

龙桑和垂桑胚胎发育观察及其异常现象

以桑科植物的两个变种龙桑与垂桑的雌株为材料(种植园内无雄株),采用常规石蜡切片法,对单性结实桑科植物的胚胎发育过程进行观察.结果表明:单性结实的垂桑和龙桑胚胎发育过程基本相似.大孢子卵细胞不经过授粉受精直接开始发育,发育成类似于合子的胚状体结构,再进行分裂,依次经棒状形、球形、心形、鱼雷形,至果实成熟时形成子叶胚.其胚珠倒生,胚囊发育类型属于蓼形胚囊,生殖方式属于二倍体孤雌生殖.在种子发育过程中发现,珠心、珠被细胞解体、大孢子卵细胞不正常分裂等异常现象.     

基于SupReMe影像重建和RF的玉米冠层LAI反演

针对Sentinel-2卫星影像拥有3个对植被生长状况非常敏感、空间分辨率为20m的红边波段（705、740、783nm）,其空间分辨率与可见光和近红外波段10m的空间分辨率不一致,使Sentinel-2影像应用受到限制的问题,基于多光谱多分辨率估计的超分辨率（Super-resolution for multispectral multiresoltion estimation, SupReMe）算法将空间分辨率20m的6个波段重建为10m;以重建后的影像为数据源,耦合PROSAIL辐射传输模型和随机森林模型反演玉米冠层叶面积指数（LAI）,并以野外实测LAI验证其反演精度。结果表明,采用SupReMe算法对Sentinel-2影像进行重建后,在保持光谱特性不变的同时提高了影像的空间细节;基于重建影像和原始影像的LAI反演决定系数R2分别为0.70、0.68,均方根误差RSME分别为0.240、0.262。研究表明,利用SupReMe算法重建后的Sentinel-2卫星影像,能够在提高玉米冠层LAI反演空间分辨率的同时提高反演精度,在挖掘高分辨率农作物生长信息方面具有很大潜力。     

Tape-Arabidopsis Sandwich - a simpler Arabidopsis protoplast isolation method

Background  

Protoplasts isolated from leaves are useful materials in plant research. One application, the transient expression of recombinant genes using Arabidopsis mesophyll protoplasts (TEAMP), is currently commonly used for studies of subcellular protein localization, promoter activity, and in vivo protein-protein interactions. This method requires cutting leaves into very thin slivers to collect mesophyll cell protoplasts, a procedure that often causes cell damage, may yield only a few good protoplasts, and is time consuming. In addition, this protoplast isolation method normally requires a large number of leaves derived from plants grown specifically under low-light conditions, which may be a concern when material availability is limited such as with mutant plants, or in large scale experiments.     

Identification of favorable alleles in the <Emphasis Type="Italic">non</Emphasis>-<Emphasis Type="Italic">yellow coloring 1</Emphasis> gene by association mapping in maize

Association mapping was conducted to explore favorable alleles of the chlorophyll-related non-yellow coloring 1 (NYC1) gene under light and dark using an association panel of 146 maize inbred lines. A total of 14 polymorphic sites were identified to be significantly associated with at least one of the chlorophyll-related traits at the seedling stage. Four single nucleotide polymorphisms (SNPs) (S320, S2951, S3901, and S3355) from the NYC1 gene were respectively strongly associated with chlorophyll b (chlb), the ratio of chlorophyll a to chlorophyll b (chl_ratio), chlorophyll a degradation (chla_deg), and total chlorophyll degradation (total_chl_deg). SNPs S320 (C/A) in exon 1, and S2951 (A/G) in intron 8 was related to chlb, with 6.01 and 8.89% of phenotypic variation under light treatment, respectively. Under dark treatment, SNP S3901 (C/T), located in 3′ untranslated region (3′UTR), was associated with chl_ratio, explaining 7.01% of the observed phenotypic variation, whereas SNP S3355 (C/G) in intron 9 explained 6.48 and 5.18% of phenotypic variations in chla_deg and total_chl_deg, respectively. Taken together, these results indicated that the NYC1 gene plays an important role in chlorophyll content and other related traits, and different sites act on chlorophyll metabolism under different light intensities in maize seedlings. Furthermore, these findings improve our understanding of the genetic basis of chlorophyll metabolism under different light conditions.     

温室大棚钢骨架自动弯折和自动焊接技术及装备

介绍了钢骨架自动弯折和自动焊接技术及装备的必要性和重要性,以及国内外发展现状和趋势。针对目前骨架制作难、安全性差、现有设备移动不便及消耗大量人力物力等一系列问题,在国内外现有钢骨架制作技术及装备的基础上,自主开发研制一种新型的并适合我国国情的钢骨架自动弯折和自动焊接技术及装备。重点介绍钢骨架自动弯折和自动焊接技术及装备的主要结构组成、工作原理、工作过程和关键技术特点,并对其开发应用前景进行展望。      

A typology of agricultural production systems: Capability building trajectories of three Asian economies

Analysing the agricultural sectors of Taiwan, Malaysia and Singapore, this paper examines the capability‐building process that encourages productivity and innovation. It describes and explains the origins and subsequent evolution of three forms of agricultural production system, each generating different farming capabilities and distinct forms of competitive advantages. The paper argues that Taiwan's rice‐oriented agricultural production system stimulates both productivity and innovation, helping Taiwanese farmers raise their income level and living standards. The active deployment of state institutions and a malleable labour force, evidenced in the Malaysian palm oil industry, is effective in raising farming productivity but not the ability to innovate. Singapore's aquaculture‐oriented agricultural production system is somewhat useful in stimulating productivity and innovation. Yet the city‐state's inherent lack of space and open international trade regime have circumscribed the potential of its aquaculture industry. It now relies on the regional ethnic Chinese business networks to expand the aquaculture industry's knowledge base and its industrial commons. The principles discussed in this paper provide policy lessons, or at least some initial guidance, for other developing economies aspiring to modernise their agricultural sector.     

表型组学及多组学联合分析在植物种质资源精准鉴定中的应用

种质资源是指具有实际或潜在利用价值的、携带生物遗传信息的载体。表型组学是近年来新兴的高通量、高分辨率的表型分析技术和平台。随着各种先进的表型组学研究工具和平台的研发,表型组学在种质资源筛选和评价中起着越来越重要的作用:在植物优异农艺性状鉴定、抗逆性研究、突变体研究和分子标记辅助育种研究中,高能量、高分辨率的鉴定技术对基因与环境互作的表型鉴定,为种质资源的规模化、批量化鉴定评价提供了基础和条件,为发掘优异种质和优良等位基因奠定了基础。