摘要：【中文】提供了水性聚合物分散体和包含这种水性聚合物分散体的水性涂料组合物，并提供了具有改进的早期耐水泡性以及令人满意的耐水纹性和良好耐久性的涂料。 【EN】Aqueous polymer dispersions and aqueous coating compositions comprising such aqueous polymer dispersions are provided, and coatings having improved early blister resistance as well as satisfactory water mark resistance and good durability are provided.

摘要：【中文】本发明提供一种基于神经网络的海量流量数据类型的检测方法，涉及信息处理技术领域。该方法首先对原始流量进行流量类型标注，作为原始训练数据集，并对其以会话为单位进行切割，每个会话单独组成流量数据包序列，并在长度上加以处理，得到若干条等长数据包，将该数据包数据进行图形化操作，并按照时间顺序堆叠成流量图像三维数据；然后将预处理好的流量图像三维数据送入以3D卷积神经网络为基础的流量分类模型中，训练保存该模型，并检测模型准确性。将待分类的流量数据进行相同预处理操作，送入训练好的模型，得到分类结果。本发明提供的检测海量流量数据类型的方法，能够在接受海量流量数据的同时，快速准确的对流量数据类型做出分类。 【EN】The invention provides a method for detecting mass flow data types based on a neural network, and relates to the technical field of information processing. The method comprises the steps of firstly, carrying out flow type marking on original flow to serve as an original training data set, cutting the original training data set by taking a conversation as a unit, independently forming a flow data packet sequence by each conversation, processing the flow data packet sequence in length to obtain a plurality of equal-length data packets, carrying out graphical operation on the data packet data, and stacking the data packet data into flow image three-dimensional data according to a time sequence; and then sending the preprocessed flow image three-dimensional data into a flow classification model based on a 3D convolutional neural network, training and storing the model, and detecting the accuracy of the model. And carrying out the same preprocessing operation on the flow data to be classified, and sending the flow data to the trained model to obtain a classification result. The method for detecting the type of the mass flow data can receive the mass flow data and can quickly and accurately classify the type of the mass flow data.

摘要：【中文】本发明提供一种基于渐进式生成对抗网络的视频摘要方法，涉及信息处理技术领域。该方法首先按照一定帧率将视频切分成图片的集合，将视频数据转化为图片数据；然后建立渐进式生成对抗网络模型，逐渐增加模型的网络层，从低分辨率训练到高分辨率，提取关键帧；同时，选择精度模式或者收敛模式，以决定模型以何种方式在某一分辨率上停止训练；最终，给出视频所有帧的标签，以标记出视频关键帧所在；利用该标签可提取关键帧，合成摘要短视频。本发明提供的视频摘要方法，采用一种无监督训练方式，不需要对视频做出关键帧的人工标记，同时渐进式的训练，可充分利用局部信息，降低了训练的复杂性，增加了训练结果的稳定性。 【EN】The invention provides a video abstraction method based on a progressive generation countermeasure network, and relates to the technical field of information processing. Firstly, cutting a video into a set of pictures according to a certain frame rate, and converting video data into picture data; then establishing a progressive generation confrontation network model, gradually increasing the network layer of the model, training from low resolution to high resolution, and extracting key frames; meanwhile, selecting a precision mode or a convergence mode to determine the mode of the model to stop training at a certain resolution; finally, labels of all frames of the video are given to mark the positions of the video key frames; the key frames can be extracted by utilizing the label to synthesize the abstract short video. The video abstraction method provided by the invention adopts an unsupervised training mode, does not need to manually mark key frames for videos, and can fully utilize local information by progressive training, thereby reducing the training complexity and improving the stability of training results.

摘要：【中文】本发明提供一种基于向量化的恶意流量攻击类型的判断方法，涉及信息处理技术领域。该方法首先把原始流量数据以会话为单位进行长度统一，然后送入自动编码机模型提取大小一致长度适中的会话数据，将处理后的会话数据转化为图像集合，送入卷积神经网络训练得到向量，并与对应统计特征提取的向量进行拼接，得到新的流量向量集合，并放到临时数据库中；对临时数据库中的向量以余弦距离为依据进行聚类，找出聚类中心的向量，构建行为向量库；预处理后的待检测流量向量与行为向量库中向量以余弦距离为依据进行相似度检测，获得待检测流量的流量类型。本发明方法通过向量距离的计算，提高了检测的效率，有效的降低了相似的恶意流量特征的误判程度。 【EN】The invention provides a vectorization-based method for judging malicious traffic attack types, and relates to the technical field of information processing. Firstly, unifying the length of original flow data by taking a conversation as a unit, then sending the original flow data into an automatic coding machine model to extract conversation data with consistent size and moderate length, converting the processed conversation data into an image set, sending the image set into a convolutional neural network to train to obtain a vector, splicing the vector with the vector extracted by corresponding statistical characteristics to obtain a new flow vector set, and putting the new flow vector set into a temporary database; clustering vectors in the temporary database by taking cosine distance as a basis, finding out a vector of a clustering center, and constructing a behavior vector library; and performing similarity detection on the preprocessed flow vector to be detected and the vector in the behavior vector library by taking the cosine distance as a basis to obtain the flow type of the flow to be detected. The method improves the detection efficiency and effectively reduces the misjudgment degree of similar malicious flow characteristics by calculating the vector distance.

摘要：【中文】本发明涉及一种沥青路面渗水系数试验中密封材料快速刮涂装置，其包括外壳总成、内壳总成、转动机构、密封材料刮涂机构，所述外壳总成上部设置转动机构，外壳总成内部设置内壳总成，所述密封材料刮涂机构位于外壳总成和内壳总成之间，密封材料刮涂机构通过转轴与转动机构连接，密封材料刮涂机构在转动机构的驱动下绕转轴转动，内壳总成包括内壳体、压缩出料机构及底座，内壳体设置有密封料排出孔。本发明结构设计科学合理，实现快速精确刮涂环形密封区域的目的，有效提高沥青路面渗水系数试验的效率和结果的准确性。 【EN】The invention relates to a sealing material rapid scraping device in an asphalt pavement water seepage coefficient test, which comprises an outer shell assembly, an inner shell assembly, a rotating mechanism and a sealing material scraping mechanism, wherein the rotating mechanism is arranged at the upper part of the outer shell assembly, the inner shell assembly is arranged inside the outer shell assembly, the sealing material scraping mechanism is positioned between the outer shell assembly and the inner shell assembly, the sealing material scraping mechanism is connected with the rotating mechanism through a rotating shaft, the sealing material scraping mechanism rotates around the rotating shaft under the driving of the rotating mechanism, the inner shell assembly comprises an inner shell, a compression discharging mechanism and a base, and the inner shell is provided with a sealing material discharging hole. The invention has scientific and reasonable structural design, realizes the purpose of quickly and accurately scraping and coating the annular sealing area, and effectively improves the efficiency of the asphalt pavement water permeability coefficient test and the accuracy of the result.