Bio-replicated forming of the biomimetic drag-reducing surfaces in large area based on shark skin

On the investigation of biomimetic drag-reducing surface, direct replication of the firm scarfskins on low-resistance creatures to form biomimetic drag-reducing surfaces with relatively vivid morphology relative to the living prototype is a new attempt of the bio-replicated forming technology. Taking shark skin as the bio-replication template, the hot embossing method was applied to the micro-replication of its outward morphology. Furthermore, the skins were jointed together to form the drag-reducing surface in large area. The results of the resistance measurements in a water tunnel according to the flat-plate sample pieces have shown that the biomimetic shark-skin coating fabricated by the bio-replicated forming method has significant drag reduction effect, and that the drag reduction efficiency reached 8.25% in the test conditions.     

无指盘臭蛙栖息地环境微生物及皮肤分泌物初步分析

在贵州水城县、云南官渡区、四川会理县和四川冕宁县采集了无指盘臭蛙动物样品,收集了其栖息地的水样及土样、以及环境因子.对所获土样和水样中的微生物进行了计数,并鉴定了优势细菌的种类及其丰度.进行了细菌总数和非优势细菌比例对无指盘臭蛙皮肤分泌物体重比关系的］初步分析.结果表明无指盘臭蛙皮肤分泌物体重比与土样中的细菌总数极显著地线性相关(p＜0.01),而分泌物体重比与土壤非优势细菌比例仅呈现粗略的对数关系(p＜0.1).     

两栖动物皮肤抗菌肽的研究进展

两栖动物的皮肤抗菌肽是抵御外界微生物侵袭的有效屏障之一.此文综述了两栖动物皮肤抗菌肽的最新研究进展,重点介绍了天然抗菌肽的分离、纯化、基因克隆与表达的过程,并展望了其在农业和临床上的应用前景.     

6种化学消毒剂对养殖海水中浮游生物的影响

 采用不同质量浓度的聚维酮碘、二溴海因、福尔马林、富氯、漂粉精和次氯酸钠等6种化学消毒剂,对蓄水沉淀池养殖海水进行消毒处理,处理前后分别测定浮游生物种类和数量变化。结果显示：在合理的浓度范围内,各种化学消毒剂对浮游生物均有一定程度的影响,和质量浓度成正相关变化。含氯消毒剂作用效果最好,有效氯达到10 g·m^-3时,漂粉精对浮游动物杀灭率最高,达到97.3%,次氯酸钠次之,为91.4%。本文研究结果对对虾养殖用水的处理和循环利用有重要的指导意义,有利于部分切断对虾白斑综合症病毒的水平传播途径,减少病害的发生,促进对虾养殖业的健康发展。     

基于YIQ颜色空间的自适应肤色检测算法

选取YIQ颜色空间,通过选择多种拍摄环境、光照条件的训练样本图像,统计其中肤色像素点和非肤色像素点中颜色分量I的分布概率,构建皮肤颜色分布模型,并利用贝叶斯规则构造肤色分类器.同时,计算测试图像准肤色区域的平均亮度,作为自适应阈值进行肤色判别.实验表明,该方法对不同亮度图像和不同肤色类型都有很好的检测效果.     

翁方纲的诗学及“学术诗”

阐述了翁方纲的生平与学术成就。认为翁方纲的“肌理说”及其在创作上的反映是：（1）对王士祯“神韵”的批判及牵制;（2）以“理”和学问为基础的诗学;（3）学术诗：以考据为诗。最后,提出了关于翁方纲的诗学及“学术诗”的结论。     

Involvement of NADPH oxidase NtrbohD in the rapid production of H2O2 induced by ABA in cultured tobacco cell line BY-2

The mechanisms for the production of hydrogen peroxide (H₂O₂) induced by abscisic acid (ABA) were investigated in suspension culture cells of tobacco BY-2 cells. The results showed that the immediate generation of H₂O₂, which was mainly derived from superoxide dismutase-catalyzed dismutation of superoxide radical, was significantly induced by ABA. Furthermore, treatment of the cultured tobacco cells with ABA resulted in a time-dependent quick increase in plasma membrane (PM) NADPH oxidase activity, which coincided on time and magnitude with the elevation in ABA-induced accumulation of H₂O₂. Moreover, these enhanced effects were pronouncedly inhibited by two NADPH oxidase inhibitor, diphenylene iodonium and imidazole, suggesting that PM NADPH oxidase is involved in the rapid accumulation of ₂O₂ in cultured tobacco cells. In addition, analysis of the expression level of NtrbohD, a PM NADPH oxidase gene in tobacco, by RT-PCR and protein gel blot revealed that the gene at both mRNA and protein levels was upregulated by ABA, indicating that NtrbohD participates in the ABA-stimulated rapid production of H₂O₂ in tobacco culture cells. Taken together, these findings suggest that ABA induces the rapid accumulation of reactive oxygen species via NADPH oxidase in suspension culture cells of tobacco, and that NADPH oxidase and H₂O₂ appear to be important components in ABA signal transduction pathway in plants.     

中华芦荟成熟叶组织培养的研究

以中华芦荟成熟叶为外植体,接种于9组不同2,4-二氯苯氧乙酸(2,4-D)和6-苄基腺嘌呤(6-BA)浓度配比的脱分化培养基中,诱导外植体的脱分化形成愈伤组织。研究表明,中华芦荟叶愈伤组织的最佳培养基为MS+0.5mg/L(2,4-D)+3mg/L(6-BA)+30g/L蔗糖+10g/L琼脂+1.0g/L活性炭;诱导不定芽的适宜培养基为MS+1mg/L(2,4-D)+3mg/L(6-BA)+30g/L蔗糖+10g/L琼脂+0.7g/L活性炭;及诱导生根的适宜培养基为1/2MS+0.2mg/Lα-萘乙酸(NAA)+20g/L蔗糖+10g/L琼脂+0.5g/L活性炭,生根率达100%.     

一种基于K-均值聚类方法的肤色分割算法

随着智能化信息处理技术的发展,人脸检测与定位越来越受到重视.本文在利用颜色信息进行皮肤检测的过程中采用YCbCr颜色空间,利用高斯肤色模型,得到图像中任意像素属于肤色的概率,同时借鉴了模式识别中广泛应用的K-均值的思想,设计了基于K-均值聚类方法的肤色分割算法,完成了图像的二值化.     

视频序列中运动人脸的检测与特征定位

针对视频序列中运动人脸的检测与特征定位问题,给出了一种算法,即通过计算并比较人脸区域而不是整幅视频帧的峰值信噪比（PSNR）,先从视频序列中找到人脸区域清晰度高且尽可能大的视频帧,然后再进行人脸的准确检测与特征定位。首先利用Adaboost方法检测出人脸的大致范围,再根据肤色模型确定人脸的具体位置并从图像中提取出人脸部分,然后利用基于帧间差的人脸区域的PSNR判断图像清晰度,从而定位出人脸区域清晰度高且尽可能大的视频帧,最后进行人脸检测和特征定位。对实际视频序列的计算表明,给出的算法在时间效率上较高,人脸的检测与特征定位效果较好。