基于深度学习的倾斜摄影建筑物表面损毁信息提取

建筑物受损信息是地震受灾程度评估的基础,针对传统建筑物表面信息识别人工成本高、效率低等问题,受深度学习提取建筑物影像的启发,提出利用无人机倾斜摄影模型与深度学习相结合的方法提取震后建筑物表面破损信息。以2019年长宁6.0级地震为例,选用双河镇震后倾斜摄影模型切片图为数据源,对比分析面向对象分类方法、VGG-16模型和DeeplabV3+模型对建筑物表面损毁信息的提取结果。分析结果表明,针对建筑物表面破损信息的提取,尤其是细小裂缝的提取,语义分割网络DeeplabV3+模型具有较强的优势(准确率96.93%、召回率96.85%、总体精度96.89%),可实现建筑物表面破损信息的有效提取,具有较强的应用价值。     

强震临震微波动现象初步研究(二)

为了验证和检验强震临震微波动现象及其主要特征,利用甘、青、川、滇、藏等区域台网200余个宽频带数字地震台站的波形资料,建立了实时监控技术系统。通过对2012~2014年间青藏块体发生的24次5级以上地震及部分无震区域台站观测的全程实时跟踪、动态监控,检验和验证了临震微波动现象的重现性和客观性,验证了临震微波动现象的频谱、时间、空间、方向性等特征,进一步得出了临震微波动现象可能与地震强度有关,且震级越大,震前出现持续临震微波动的可能性越大的结论。     

大气散射对采用归一化植被指数进行赤潮遥感监测的影响研究

1　引言近年来赤潮灾害在我国海域频繁发生,对沿海环境、经济和人民生产生活都造成一定的影响,因此对赤潮进行实时监测并业务化运行就显得尤其重要.赤潮遥感监测是目前常用手段之一,在实践中取得了一定的效果,但到目前为止还没有一种成熟的算法能用于业务化监测赤潮的发生.很多研究人员借鉴归一化植被指数(NDVI)阈值法、海表温度(SST)异常等来判断赤潮的发生,但归一化植被指数阈值法经常会出现误判或无法区分正常与异常水体的情况.该阈值范围由先验知识得到,没有定量的理论依据.归一化植被指数是陆地遥感中用以表征地表植物覆盖程度和健…     

一种基于16S rRNA基因检测海洋浮游细菌群落的微阵列基因芯片

A better understanding of bacterioplankton community shifts following change in marine environments is critical to predict the marine ecosystem function. In order to get a snapshot of the microbial taxonomy profiling of a wide range marine area, a quick, convenient and low cost method would be favorable. In this study, we developed a 16S rRNA gene-based microarray using ARB software, which contained 447 probes targeting 160 families of marine bacteria. The specificity, sensitivity and quantitative capability of this microarray were assessed by single cloned16S rRNA genes. The reliability of this microarray was tested by eight environmental samples. The results showed that the microarray was specific, only 1.16% false results were detected in five single-clone hybridization tests. The microarray could detect DNA samples as few as 1 ng/μL and the signal intensity could reflect the relative abundance of the bacteria in the range of 1 ng/μL to 100 ng/μL of DNA concentration. Hybridization with environmental samples showed that it can discriminate bacterioplankton communities by sites and time. High throughput sequencing results from the eight samples confirmed the hybridization results. It indicated that this developed microarray could be used as a convenient tool to monitor the bacterioplankton community in marine environment.     

致病性哈维氏弧菌:海洋寄生性纤毛虫的内共生细菌

哈维氏弧菌被观察到与海洋寄生性纤毛虫刺激隐核虫具有内共生现象,其作为一种致病性细菌,可导致感染刺激隐核虫的大黄鱼产生严重的继发性细菌感染。我们通过16s宏测序技术对刺激隐核虫的细菌群落进行鉴定,并通过标准的细菌培养方法分离鉴定出哈维氏弧菌。通过透射电镜和荧光原位杂交方法,在刺激隐核虫的细胞胞质中,均观察到内共生的哈维氏弧菌的存在;而在其细胞核中,则没有相关信号存在。哈维氏弧菌与刺激隐核虫的相关关系以及内共生的哈维氏弧菌对刺激隐核虫的影响作用,仍需要深入进行研究。     

Prevalence of Bacillus sp. among the biofilm forming community on Ti surface in marine environment

Prevalence of bacterial species involved in biomineralization of manganese on titanium (Ti) surfaces in marine environment was revealed in this research work. This study involves one year sea water exposure of Ti and their periodical biofilm characterization was carried out to quantify the manganese oxidizing bacterial (MOB) presence in the biofilm formed on titanium surfaces. The total viable count study of Ti coupons exposed to sea water for one year resulted in 60% of the MOB in overall biofilm population. The biochemical characterization of MOB isolates were performed for the genus level identification of the seven bacterial isolates. Further, the seven strains were subjected to 16S rRNA gene sequencing. Evolutionary analysis was performed using MEGA 7 to obtain closely related strains within the groups. The manganese oxidizing ability of the bacterial isolates were determined with Leucoberbelin Blue Assay (LBB) and Atomic Absorption Spectroscopy studies (AAS). The results show that among the isolated marine MOB species, Bacillus sp. and Leptothrix sp. have the maximum Mn oxidizing property. The microtitre plate assay was performed to determine the biofilm forming ability of the isolated marine MOB species. All the results have confirmed the prevalence of Bacillus sp. among the biofilm colonizers on Ti surfaces when exposed in sea water.     

3种典型荒漠灌木内生固氮菌及固氮酶基因nifH多样性分析

具有特殊生境的内生固氮菌,对改善植物营养、增强植物抗逆性及群落稳定性具有重要作用,是一类潜力巨大、尚待开发的微生物资源。以新疆3种典型荒漠灌木多枝柽柳（Tamarix ramosissima）、黑果枸杞（Lycium ruthenicum）、盐穗木（Halostachys caspica）为材料,从根和枝条组织中分离筛选获得137个内生固氮菌菌株,采用16S rDNA PCR-RFLP、16S rDNA序列测定、BOXAIR-PCR指纹图谱、nifH PCR-RFLP等方法分析其遗传多样性及系统发育关系。结果表明：分离菌株经16SrDNA PCR-RFLP酶切图谱聚类划分为9个遗传类型。序列测定和系统发育分析显示,代表菌株分属于拉恩氏菌属（Rahnella）、假单胞菌属（Pseudomonas）、泛菌属（Pantoea）、芽孢杆菌属（Bacillus）、寡养单胞菌属（Stenotrophomonas）和申氏杆菌属（Shinella）,其中Rahnella为优势种群。一个测定菌株的16SrDNA序列同源性低于97%,提示可能为一个潜在的新种。BOXAIR-PCR分析优势种群的基因组结构特征,获得12种指纹图谱类型,显示出该种群具有丰富的基因组多样性。nifH PCR-RFLP分析分离内生固氮菌菌株固氮酶基因nifH的分子多态性,将其划分为3种基因型,体现了nifH具有高度保守性,同时在不同固氮微生物种群和菌株间也存在一定的多样性。研究结果丰富了固氮微生物物种资源库和基因库,对于内生固氮菌资源的保护和利用奠定了科学基础。     

Phylogenetic diversity of planktonic bacteria in the Chukchi Borderland region in summer

Planktonic bacteria are abundant in the Chukchi Borderland region. However, little is known about their diversity and the roles of various bacteria in the ocean. Seawater samples were collected from two stations K2S and K4S where sea ice was melting obviously. The analysis of water samples with fluorescence in situ hybridization (FISH) showed that DMSP-degrading bacteria accounted for 13% of the total bacteria at the station K2S. No aerobic anoxygenic phototrophic (AAP) bacteria were detected in both samples. The bacterial communities were characterized by two 16S rRNA gene clone libraries. Sequences fell into four major lineages of the domain Bacteria, including Proteobacteria (Alpha, Beta and Gamma subclasses), Bacteroidetes, Actinobacteria and Firmicutes. No significant difference was found between the two clone libraries. SAR11 and Rhodobacteraceae clades of Alphaproteobacteria and Pseudoalteromonas of Gammapro-teobacteria constituted three dominant fractions in the clone libraries. A total of 191 heterotrophic bacterial strains were isolated and 76% showed extracellular proteolytic activity. Phylogenetic analysis reveals that the isolates fell into Gammaproteobacteria, Bacteroidetes, Actinobacteria and Firmicutes. The most common genus in both the bacterial isolates and protease-producing bacteria was Pseudoalteromonas. UniFrac data showed suggestive differences in bacterial communities between the Chukchi Borderland and the northern Bering Sea.