基于部分亲和场的行走奶牛骨架提取模型

在奶牛关键点预测的基础上,通过点、线重构奶牛骨架结构,可为奶牛跛行检测、发情行为分析、运动量估测等提供重要参考。本研究基于部分亲和场,以养殖场监控摄像头拍摄的视频为原始数据,使用1 600幅图像训练了奶牛骨架提取模型,实现了奶牛站立、行走状态下关键点信息和部分亲和场信息的预测,并通过最优匹配连接对奶牛骨架结构进行准确提取。为了验证该模型的性能,采用包含干扰因素的100幅单目标奶牛和100幅双目标奶牛图像进行了测试。结果表明,该模型对单目标行走奶牛骨架提取的置信度为78.90%,双目标行走奶牛骨架提取的置信度较单目标下降了10.96个百分点。计算了不同关键点相似性(Object keypoint similarity,OKS)下的模型准确率,当OKS为0.75时,骨架提取准确率为93.40%,召回率为94.20%,说明该模型具有较高的准确率。该方法可以提取视频中奶牛骨架,在无遮挡时具有高置信度和低漏检率,当遮挡严重时置信度有所下降。该模型的单目标和双目标图像帧处理速度分别为3.30、3.20 f/s,基本相同。本研究可为多目标奶牛骨架提取提供参考。     

Water and nitrogen interaction on soil profile water extraction and ET in maize-wheat cropping system

In the present study, water and nitrogen interaction on soil profile water extraction and evapo-transpiration (ET) was investigated taking a field experiment on a clay loam soil (Typic Haplustept) at the Indian Agricultural Research Institute, New Delhi with four consecutive crops (maize-wheat-maize-wheat) taken from July 2002 to April 2004. Three levels of water regime, namely W1, W2 and W3 referring to limited, medium and maximum irrigation were applied to each crop depending on the seasonal rainfall and the critical crop growth stage. The three water regimes were used with five nitrogen levels from T1 to T5, (T1, 0% N; T2, 75% N; T3, 100% N; T4, 150% N; T5, 100% N from organic source) in a split plot design for the four crops grown in sequence.Significant water and nitrogen interaction was observed for ET and soil profile water extraction pattern. Averaged across nitrogen treatments, ET in W2 and W3 were higher by 17 and 26%, respectively than W1 in maize 2002 and by 12 and 19% in maize 2003. In case of wheat, ET in W2 and W3 were higher by 27 and 58% than W1 in 1st crop and by 37 and 70% in 2nd crop. The effect of nitrogen regime, however, was prominent in both crops of maize and wheat, with significantly higher profile soil moisture depletion in T4 of each water regime. In all cases, lowest water depletion was observed in control plots receiving 0% N.In both crops, water extraction from surface 60 cm was highest in W3 followed by W2 and W1. In maize, the % extraction from 0 to 60 cm layer varied from 71 to 76% (W1), 70-79% (W2) and 75-82% (W3), whereas the values for wheat were 70-77, 72-79 and 75-83% for W1, W2 and W3, respectively. The 90-120 cm layer contributed only 3-14% to total water extraction in both the crops. From 90 to 120 cm layer, higher extraction was observed in W1 as compared to W3. The extraction values in W1, W2 and W3 in maize were 9-13, 7-14 and 3-9, respectively, whereas the corresponding values in wheat were 8-14, 5-12 and 3-7% for the three water regimes. Effect of nitrogen treatments on water extraction from deeper layer was observed with higher extraction in highest fertilized treatment (T4) as compared to other treatments.     

锦鲤保护性组织RNA提取及引物扩增研究

从养殖锦鲤各种组织中提取完整的RNA, 是分子生物学研究锦鲤应激表达、体色变异、生长发育等功能基因表达的关键实验。在不伤害养殖锦鲤个体健康的情况下, 取用鱼体的鳍条、鳞片、鳃丝、血液4种保护性组织,可以最大程度地减少对珍贵锦鲤个体的损伤, 并达到研究相关基因表达的目的。本研究利用TR izo l法提取RNA, 获得的各样品RNA条带完整、纯度高。结果表明, 4 种保护性组织提取的RNA 经过逆转录反应, 得到了高质量的CDNA。内标基因( B- actin)、热激蛋白基因( hsp70)、金属硫蛋白基因(MT)扩增得到清晰的条带, 通过NCB I数据上的BLAST比对证明, 各序列的同源性在95%以上。利用本研究的方法, 可以充分保护养殖锦鲤个体的完好性, 用于进一步的分子生物学研究。     

火龙果茎多糖的超高压提取及抗氧化活性研究

为研究火龙果茎多糖的超高压提取工艺及抗氧化活性。在单因素试验基础上,采用Box-Behnken设计和响应面分析方法,确定超高压提取的最优工艺,并从清除ABTS自由基和DPPH自由基能力方面来评价火龙果茎多糖的体外抗氧化能力。结果表明,火龙果茎多糖的超高压提取的最佳工艺条件为:液固比10∶1(m L∶g)、超高压压力300 MPa、超高压时间4 min。在此工艺条件下多糖得率为(2.83±0.02)%。火龙果茎对ABTS自由基和DPPH自由基具有清除作用,对ABTS自由基和DPPH自由基的清除率IC_(50)分别为浓度为4.3 mg/m L和5.5 mg/m L时。     

香蕉中噻唑膦残留分析方法研究

建立香蕉中噻唑膦的超高效液相色谱串联质谱法(UPLC-MS/MS)残留测定方法。香蕉样品采用乙腈提取和PSA+C18分散固相萃取,采用UPLC-MS/MS多反应模式(MRM)进行分析测定。噻唑膦在添加水平0.01~1.0 mg/kg范围内,平均回收率为94.3%~107.5%,相对标准偏差为7.9%~11.0%。采用外标法定量,方法的检出限为0.001μg/m L,定量限为0.005 mg/kg。该方法简单、准确、重复性较好且节省溶剂,适用于香蕉中噻唑膦的残留检测。     

优化提取裸花紫珠总黄酮含量及对表皮伤口易感染菌抑菌性的研究

为了选出裸花紫珠总黄酮提取的最优方法,比较了ASE、超声、微波萃取和多功能浓缩4种提取方法在裸花紫珠总黄酮提取率上的差异,并利用微孔刃天青法测定总黄酮对表皮易感染菌的MIC值。结果发现ASE法总黄酮提取率最高,约为1.26%,其次为超声和微波萃取法,总黄酮提取率分别约为1.03%和0.98%,多功能浓缩法提取率最低,约为0.36%,且耗时最长。而且研究发现,ASE的MIC值最低。最终得出ASE提取效率最高,且提取物有较好的抑菌性,是裸花紫珠总黄酮提取的最优方法。     

基于GF-1影像面向对象分类方法的水稻种植信息提取研究

应用遥感技术提取水稻种植信息是农业遥感的重要内容。GF-1卫星WFV数据为农业信息提取提供了新的途径,面向对象的分类方法是遥感解译的重要方法。本研究以扬州市为研究区域,基于GF-1影像WFV数据,采用面向对象的分类方法,提取水稻种植信息,并实地调查验证试验结果,试图探讨GF-1数据面向对象分类方法在水稻种植信息提取中的可行性与影响提取精度的因素。结果表明,应用GF-1数据,采用面向对象的分类方法能够很好地完成扬州市水稻种植信息的提取,2016年扬州市有水稻种植面积214 524 hm~2,总体精度达到98.5%,Kappa系数0.95,面积精度达97.5%;实地考察能够提高提取精度,地形破碎程度越低,提取精度越高。     

响应面法优化超声波辅助提取薏苡仁低聚糖工艺的研究

采用响应面法优化超声波辅助提取薏苡仁低聚糖的工艺条件。在单因素试验基础上,选取液料比、超声波时间以及超声波功率3个因素结合Box-Behnken试验建立数学模型,分析考察3个因素对薏苡仁低聚糖响应值的影响程度,优化工艺参数。各因素对薏苡仁低聚糖提取率影响程度从大到小顺序依次为:超声波功率超声波时间液料比。响应面设计法优化出其最佳超声波提取条件为:超声波温度70℃,液料比33∶1(m L/g),超声波时间27 min,超声波功率450 W。在该条件下,薏苡仁低聚糖提取率为0.94%,与模型预测值0.98%接近。说明使用响应面法优化超声波辅助提取薏苡仁低聚糖的工艺条件是可行的。     

纳米TiO2在果蔬包装中的应用研究进展

统计归纳了纳米TiO_2在食品保鲜中的应用及研究热点,分析了纳米TiO_2的抑菌机制及作用,综述了纳米TiO_2在水果、蔬菜、食用菌保鲜中的应用研究进展,展望了纳米TiO_2在食品保鲜领域未来的研究方向,以期为纳米TiO_2在果蔬贮藏保鲜中的产业化应用提供理论基础和技术参考。