改进YOLOv5测量田间小麦单位面积穗数

单位面积穗数是决定小麦产量的主要因素之一。针对人工清点小麦穗数的方法容易受主观因素影响、效率低和图像处理方法鲜有进行系统部署等问题,提出一种注意力模块（Convolutional Block Attention Module,CBAM）与YOLOv5相结合的CBAM-YOLOv5网络结构,通过对特征图进行自适应特征细化,实现更准确的单位面积穗数测量。该研究以本地采集小麦图像数据和网络公开小麦图像数据为数据集,设置输入图像分辨率为1 280,得到CBAM-YOLOv5模型,可以达到0.904的F1分数和0.902的平均精度,测试集计数的平均相对误差为2.56%,平均每幅图像耗时0.045 s,综合对比,CBAM-YOLOv5模型具有显著优势。模型放置于服务器,结合手机端软件和辅助装置,形成单位面积穗数测量系统,实现育种小区麦穗图像实时采集、处理和计数,计数的平均相对误差为2.80%,抗环境干扰性强。该研究方法与装置可以实现田间小麦单位面积穗数的实时在线检测,降低主观误差,具有较高的准确率及较强的鲁棒性,为小麦单位面积穗数快速、准确估测提供一种新的技术和装备支撑。     

减量侧深施肥对里下河地区单季粳稻产量和品质的影响

以优质食味粳稻南粳9108和南粳5718为材料,以常规施肥为对照（CK,施氮量300 kg/hm²、基蘖氮肥与穗氮肥比7∶3）,分别设计6个侧深施肥处理[基蘖氮肥为常规施肥处理的100%（T1）、90%（T2）、85%（T3）、80%（T4）、75%（T5）及70%（T6）,穗氮肥量同CK],探讨侧深施肥对里下河地区单季粳稻产量和品质的影响。结果表明,与CK相比,T1处理显著提高水稻产量,2个品种产量分别提高了9.49%和8.23%,经济效益显著增加,同时改善了稻米的加工品质和营养品质,但未改善稻米的食味品质;T3处理2个品种产量与CK相比无显著差异,但提高了稻米的加工品质、外观品质和食味品质,是一种优质稳产的施肥方式。     

Nitrogen Deep Placement Combined with Straw Mulch Cultivation Enhances Physiological Traits,Grain Yield and Nitrogen Use Efficiency in Mechanical Pot-Seedling Transplanting Rice

To assess the effects of straw return coupled with deep nitrogen (N) fertilization on grain yield and N use efficiency (NUE) in mechanical pot-seedling transplanting (MPST) rice, the seedlings of two rice cultivars, i.e., Yuxiangyouzhan and Wufengyou 615 transplanted by MPST were applied with N fertilizer at 150 kg/hm² and straw return at 6 t/hm² in early seasons of 2019 and 2020. The experiment comprised of following treatments: CK (no fertilizer and no straw return), MDS (deep N fertilization and straw return), MBS (broadcasting fertilizer and straw return), MD (deep N fertilization without straw return), MB (broadcasting fertilizer without straw return). Results depicted that the MDS treatment significantly increased the rice yield by 41.69%?72.22% due to total above-ground biomass, leaf area index and photosynthesis increased by 54.70%?55.80%, 38.52%?52.17% and 17.89%?28.40%, respectively, compared to the MB treatment. In addition, the MDS treatment enhanced the total N accumulation by 37.74%?43.69%, N recovery efficiency by 141.45%?164.65%, N agronomic efficiency by 121.76%? 134.19%, nitrate reductase by 46.46%?60.86% and glutamine synthetase by 23.56%?31.02%, compared to the MB treatment. The average grain yield and NUE in both years for Yuxiangyouzhan were higher in the MDS treatment than in the MD treatment. Hence, deep N fertilization combined with straw return can be an innovative technique with improved grain yield and NUE in MPST in South China.     

Dissecting Genetic Basis of Deep Rooting in Dongxiang Wild Rice

Deep rooting is an important trait in rice drought resistance. Genetic resources of deep-rooting varieties are valuable in breeding of water-saving and drought-resistant rice. In the present study, 234 BC₂F₇ backcross introgression lines were derived from a cross of Dongye 80 (an accession of Dongxiang wild rice as the donor parent) and R974 (an indica restorer line as the recurrent parent). A genetic linkage map containing 1 977 bin markers was constructed by ddRADSeq for QTL analysis. Thirty-one QTLs for four root traits (the number of deep roots, the number of shallow roots, the total number of deep roots and the ratio of deep roots) were assessed on six rice chromosomes in two environments (2020 Shanghai and 2021 Hainan). Two of the QTLs, qDR5.1 and qTR5.2, were located on chromosome 5 in a 70-kb interval. They were detected in both environments. qDR5.1 explained 13.35% of the phenotypic variance in 2020 Shanghai and 12.01% of the phenotypic variance in 2021 Hainan. qTR5.2 accounted for 10.88% and 10.93% of the phenotypic variance, respectively. One QTL (qRDR2.2) for the ratio of deep roots was detected on chromosome 2 in a 210-kb interval and accounted for 6.72% of the phenotypic variance in 2020. The positive effects of these three QTLs were all from Dongxiang wild rice. Furthermore, nine and four putative candidate genes were identified in qRDR2.2 and qDR5.1/qTR5.2, respectively. These findings added to our knowledge of the genetic control of root traits in rice. In addition, this study will facilitate the future isolation of candidate genes of the deep-rooting trait and the utilization of Dongxiang wild rice in the improvement of rice drought resistance.     

基于模糊聚类分析的草莓图像分割方法研究

为改善草莓采摘机器视觉系统中果实图像的分割效果,对普通均值聚类的分割方法理论进行分析,针对草莓果实图像的特点将模糊-均值聚类算法引入分割算法,大大改善草莓果实图像的分割效果。     

福建乌龙茶产业可持续发展的对策建议

通过对福建乌龙茶产业及相关产业链发展现状的调研,分析该产业存在问题,查阅相关资料,提出福建乌龙茶产业的可持续发展不仅要注重茶园的生态建设和休闲旅游开发,提高茶叶的品质及安全,将茶产业与旅游业有机结合,实现茶产业从第一产业向第三产业的跨越升级,更要借鉴国际茶产业应用精深加工开发茶叶副产品的先进经验,重视精深加工技术的研发,提取茶中含有的各种既有“有效成分”又有“活性成分”化学物质,发挥茶的医疗效能,弘扬中医和茶文化结合的荼疗生态文化,为“十二五”实现福建乌龙茶产业的可持续发展提出对策建议。     

土壤肥沃耕层构建模式

详细介绍土壤的物质组成及作用,论述构建土壤肥沃耕层对农业生产的重要意义及我国土壤耕层构建模式的发展历程,介绍一种将深翻和秸秆还田结合起来的土壤肥沃耕层构建新模式,为提高土壤耕层的肥力提供参考。     

都柳江不同河段急流一深潭—沙（砾）滩系统底质酶活性研究

本文以贵州省境内人工干扰较少的都柳江为研究对象,对急流-深潭-沙（砾）滩系统中底质酶活性进行研究。结果表明：（1）都柳江不同河段底质中酶活性的的总体变化趋势为：过氧化氢酶活性沙（砾）滩〉急流〉深潭,磷酸酶和脲酶活性都是深潭〉急流〉沙（砾）滩,脱氢酶活性变化规律不明显;在急流—深潭—沙（砾）滩系统中,过氧化氢酶活性和脲酶活性在深潭和急流中的最大值都在下游,沙（砾）滩的最大值在中游,磷酸酶活性的最大值均在上游,而脱氢酶活性在深潭和急流的最大值都在中游,在沙（砾）滩中的最大值在下游。（2）都柳江的河道形态结构因子与深潭中底质酶活性有一定关系,而与急流和沙（砾）滩中底质酶活性的关系不明显。（3）急流和深潭中的氮素水平与底质中的酶活性有一定的关系,其中与脲酶活性的关系尤为突出;磷素水平与底质中酶活性没有明显的相关性;电导率和COD与底质酶活性有着明显的相关性;溶解氧与急流中的过氧化氢酶有极显著正相关。研究结果为河道生物修复提供了一定的理论依据。     

深基坑支护中冲孔灌注桩质量控制研究

灌注桩+锚索支护型式被越来越多地应用于深基坑支护中。文章结合冲孔灌注桩在深圳GL河口调蓄池工程—深基坑支护中的应用实践,介绍了冲孔灌注桩施工工艺、施工过程质量控制要点,对施工准备、冲孔、确定终孔、清孔、钢筋笼制作和吊装、浇筑水下混凝土等各施工环节进行了梳理和探讨,并对施工监理工作程序和监理控制要点做了论述,借此与广大同行进行交流、学习。     

休闲期深翻覆盖配施氮磷肥对旱地小麦群体动态、叶面积的影响及其与产量的相关性

采用大田试验方法,研究了休闲期深翻覆盖配施氮磷肥对旱地小麦群体动态、叶面积的变化及其与产量相关性的影响。结果表明,氮磷比为1∶0.5和1∶0.75时,增加施氮量可提高各生育期群体茎数、单株叶面积及成穗率,可显著提高可育小穗数、干物质量,降低不可育小穗数,从而提高穗数、穗粒数及产量;施氮量为150 kg/hm2时,增加施磷量可提高各生育期群体茎数、单株叶面积及成穗率,可显著提高可育小穗数、干物质量,可提高产量及其构成因素,其中,氮磷比为1∶1较1∶0.5提高成穗率13.10%,从而提高穗数22.7%,提高穗粒数3.79%,提高产量18.92%;施氮量为180 kg/hm2时,各生育期群体茎数、单株叶面积、成穗率、穗长、可育小穗数、不可育小穗数、干物质量以氮磷比为1∶0.75最高,1∶1居中,1∶0.5最低,其中,氮磷比为1∶0.75较1∶0.5提高成穗率11.43%,提高穗数19.35%,提高穗粒数4.95%,提高产量17.31%;施氮量为180 kg/hm2时,氮磷比为1∶0.75对各生育期群体茎数、叶面积、成熟期农艺性状、产量及其构成因素的影响较施氮量为150 kg/hm2、氮磷比为1∶1时效果更好。相关分析表明,各生育期群体茎数与穗数、穗粒数密切相关,且与穗数关系更密切;生育中后期单株叶面积与穗数、穗粒数相关较密切,且与穗数关系更密切。总之,休闲期深翻覆盖配施氮肥180 kg/hm2、氮磷比为1∶0.75时更有利于促进群体有效分蘖,提高成穗率;有利于提高叶面积,从而提高干物质量;有利于提高穗数、穗粒数,最终提高产量。