Potential Applications of Milk Fractions and Valorization of Dairy By-Products: A Review of the State-of-the-Art Available Data, Outlining the Innovation Potential from a Bigger Data Standpoint

The unique composition of milk makes this basic foodstuff into an exceptional raw material for the production of new ingredients with desired properties and diverse applications in the food industry. The fractionation of milk is the key in the development of those ingredients and products;hence continuous research and development on this field, especially various levels of fractionation and separation by filtration, have been carried out. This review focuses on the production of milk fractions as well as their particular properties, applications and processes that increase their exploitation. Whey proteins and caseins from the protein fraction are excellent emulsifiers and protein supplements. Besides, they can be chemically or enzymatically modified to obtain bioactive peptides with numerous functional and nutritional properties. In this context, valorization techniques of cheese-whey proteins, by-product of dairy industry that constitutes both economic and environmental problems, are being developed. Phospholipids from the milk fat fraction are powerful emulsifiers and also have exclusive nutraceutical properties. In addition, enzyme modification of milk phospholipids makes it possible to tailor emulsifiers with particular properties. However, several aspects remain to be overcome;those refer to a deeper understanding of the healthy, functional and nutritional properties of these new ingredients that might be barriers for its use and acceptability. Additionally, in this review, alternative applications of milk constituents in the non-food area such as in the manufacture of plastic materials and textile fibers are also introduced. The unmet needs, the cross-fertilization in between various protein domains,the carbon footprint requirements, the environmental necessities, the health and wellness new demand, etc., are dominant factors in the search for innovation approaches;these factors are also outlining the further innovation potential deriving from those “apparent” constrains obliging science and technology to take them into account.

基于三尺度嵌套残差结构的交通标志快速检测算法

智能驾驶对交通标志自动检测的实时性及鲁棒性有着极高要求.目标检测中YOLOv3-tiny检测算法是轻量网络,实时性好、但准确率不高.将YOLOv3-tiny检测算法作为基础网络,提出了一种三尺度嵌套残差结构的交通标志快速检测算法.首先,在基础网络上采用逐像素相加的跨层连接,并未增加特征图的通道数,同时网络中形成1个小残差结构.其次,通过同样的跨层连接方式,增加了1层空间分辨率更高的预测输出,使得该尺度输出包含更丰富的空间信息,进而构成大残差结构.最终,将2个残差结构进行嵌套,形成了1个三尺度预测的嵌套残差网络模型,使得Tiny检测算法的部分主网络位于这2个残差结构中,起到3次调参的作用.实验结果表明:提出的算法能够快速鲁棒地检测真实场景中的交通标志.在德国交通标志检测数据集(German traffic sign detection benchmark, GTSDB)上交通标志总F1值为91.77%、检测时间为5 ms;在长沙理工大学中国交通标志检测数据集(CSUST Chinese traffic sign detection benchmark, CCTSDB)上指示、禁令、警告三大类交通标志F1值分别为92.41%,93.91%,92.03%,检测时间为5 ms.

改进实时目标检测算法的电力巡检鸟巢检测

本文提出一种基于深度学习算法的实时目标检测模型,用于电力无人机巡检中鸟巢的自动检测。通过基于距离的K-means聚类算法,对数据集的标记框重新聚类,获得了更适用于识别不同杆塔在多种所处环境下的鸟巢的锚点集合。检测结果表明,使用新集合的算法均值平均精度提高至0.896,同时召回率和平均交并比均有提高;且运用本文算法可对巡检视频进行实时化处理(单帧处理时间低于30ms),便于后续问题的实时分析及处理,满足电力巡检智能化、常态化应用需求。

析探春“拣高枝儿飞”

大多数作品在分析探春时都认为,庶出的探春"只拣高枝儿飞"和王夫人、王熙凤等当权派站在一起,背弃她的生母赵姨娘。但通过分析可以看出,探春与赵姨娘由于文化品味以及理想追求不同,决定了母女二人不可能有真正的理解和沟通。但是二人并没有抛开母女亲情。从探春三次流泪即可说明探春"只拣高枝儿飞"的观点是不正确的。《红楼梦》真正为我们展现的是探春与赵姨娘之间的本质差异,及她们之间扯不断的母女亲情与矛盾冲突。

一种动态的自适应前瞻嵌套规划算法

针对传统自适应前瞻规划算法在计算时需要将前瞻首末速度设置为固定值,且过渡衔接部分无法动态调整等问题,提出一种动态调整过渡部分的自适应前瞻嵌套规划算法。以给定误差为约束,建立拐角处圆弧转接模型。基于线性加减速模型,将自适应前瞻分段与嵌套衔接的最佳角度定为自适应前瞻分段角度,并限制自适应前瞻的末速度以实现嵌套规划。将动态调整引入前瞻规划中,动态调整过渡圆弧半径以实现全局最优。实验表明,通过动态的自适应前瞻嵌套规划算法,可以对路径进行有效规划,既提升了路径的最大允许速度、平均轮廓精度,又降低了加工耗时。