摘要：【中文】本发明涉及一种识别网络攻击的方法，包括：经由网络接收访问请求；基于镜像服务器对访问请求的镜像响应与至少一个蜜罐对访问请求的对应蜜罐响应之间的差异度来确定访问请求是否为网络攻击；其中，镜像服务器配置成实现生产服务器的镜像，蜜罐的系统配置成在系统版本上低于生产服务器的系统。其能够提供尽可能真实地模仿生产服务器的蜜罐，并通过蜜罐与网络攻击方之间的通信来识别攻击模式以及所用工具。 【EN】The invention relates to a method for identifying network attacks, which comprises the following steps: receiving an access request via a network; determining whether the access request is a network attack based on a degree of difference between a mirrored response of the mirroring server to the access request and a corresponding honeypot response of the at least one honeypot to the access request; wherein the mirror server is configured to implement mirroring of the production server, and the system of the honeypot is configured to be lower in system version than the system of the production server. It can provide honeypots simulating production servers as truly as possible and identify attack patterns and tools used through communication between honeypots and network attackers.

摘要：【中文】本发明公开了一种PCB板上的条码定位方法，对图片进行多尺度缩放，获得多个尺度下的图片；使用深度学习模型一，对每个尺度下的图片进行预测，保留预测概率值大于阈值一的矩形框；将每个尺度下检测的矩形框进行汇聚，利用非极大值抑制算法进行融合，生成候选条码框；将候选条码框输入深度学习模型二中进行预测，保留预测概率值大于阈值二的候选条码框；将保留的候选条码框利用非极大值抑制算法进行融合，输出最终的条码定位框。本发明从卷积神经网络结构优化、检测流程优化、训练样本准备，兼顾资源消耗和检测性能表现，针对pcb板上的条码区域进行定位，使条码定位准确率较高，同时平均时间消耗较短。 【EN】The invention discloses a bar code positioning method on a PCB (printed circuit board). the method comprises the steps of carrying out multi-scale zooming on a picture to obtain the picture under multiple scales; predicting the picture under each scale by using a first deep learning model, and reserving a rectangular box with a prediction probability value larger than a first threshold value; converging rectangular frames detected under each scale, and fusing by using a non-maximum suppression algorithm to generate candidate bar code frames; inputting the candidate bar code frame into a deep learning model II for prediction, and reserving the candidate bar code frame with the prediction probability value larger than a threshold value II; and fusing the reserved candidate bar code frames by using a non-maximum suppression algorithm, and outputting a final bar code positioning frame. The method optimizes the structure of the convolutional neural network, optimizes the detection process, prepares the training sample, considers the resource consumption and the detection performance, and positions the barcode region on the pcb, so that the barcode positioning accuracy is high, and the average time consumption is short.

摘要：【中文】本发明公开了一种聚苯胺@银纳米线/聚酰亚胺多孔梯度复合薄膜、制备方法及应用，包括共混物的制备，第一次旋涂，第二次旋涂，固化，模板的去除，表面亲水性处理，分散液中处理，去离子水和乙醇洗涤多次，真空干燥，最后得到聚苯胺@银纳米线/聚酰亚胺梯度复合薄膜。与现有技术相比，本发明的有益效果为：其多孔结构有效增加了复合薄膜的比表面积，其梯度结构保证了复合薄膜良好的力学性能；负载聚苯胺后的梯度复合薄膜，实现了敏感材料三维网络的构筑，对目标气体的吸附活性位点大大增加；具有对NO₂响应灵敏、柔性好、力学强度高等优点。 【EN】The invention discloses a polyaniline @ silver nanowire/polyimide porous gradient composite film, a preparation method and application thereof. Compared with the prior art, the invention has the beneficial effects that: the porous structure effectively increases the specific surface area of the composite film, and the gradient structure ensures the composite filmThe film has good mechanical properties; the gradient composite film loaded with polyaniline realizes the construction of a three-dimensional network of a sensitive material, and greatly increases the adsorption active sites of target gas; having NO in pair₂Sensitive response, good flexibility, high mechanical strength and the like.

摘要：【中文】本发明公开一种防水砂浆，其中，所述防水砂浆包括水泥砂浆、金属皂及水泥基渗透结晶型防水剂。本发明通过水泥砂浆、金属皂及水泥基渗透结晶型防水剂混合，形成防水砂浆。这样，将防水砂浆涂覆于墙面或地面后，可以理解的，本发明的技术方案能够提高砂浆料的防水性能。 【EN】The invention discloses waterproof mortar, wherein the waterproof mortar comprises cement mortar, metal soap and a cement-based capillary crystalline waterproofing agent. The waterproof mortar is formed by mixing cement mortar, metal soap and a cement-based capillary crystalline waterproofing agent. Therefore, after the waterproof mortar is coated on the wall surface or the ground, the technical scheme of the invention can improve the waterproof performance of the mortar.

摘要：【中文】本发明公开了一种考勤与学情分析方法，采集视频图像，对视频图像进行硬件解码；在边缘端对视频图像进行人脸特征值提取，上传所述人脸特征值至服务器进行考勤；在边缘端对视频图像中的人体进行人体姿态估计，进行学情行为分析；将所述人脸特征值与所述学情行为分析结果通过时间戳以及位置进行匹配；将匹配了所述人脸特征值的所述学情行为分析结果上传至所述服务器。本方案实现了对学生的自动考勤及学生的学情情况自动分析，可利用教室中原本架设的相机进行使用，减少部署开销。此外，由于采用边缘计算替代传统服务器方案将计算提到前端得到计算结果，在网络数据传输上仅传输AI计算结果而非视频数据可以减少大量的带宽提高速度。 【EN】The invention discloses an attendance and academic situation analysis method, which comprises the steps of collecting video images and carrying out hardware decoding on the video images; extracting a face characteristic value of the video image at an edge end, and uploading the face characteristic value to a server for attendance checking; estimating the human body posture in the video image at the edge end, and analyzing the learning situation behavior; matching the face characteristic value with the emotional behavior analysis result through a time stamp and a position; and uploading the emotional behavior analysis result matched with the face characteristic value to the server. The scheme realizes automatic attendance checking of students and automatic analysis of learning situation of the students, and the originally erected camera in the available classroom is used, so that the deployment overhead is reduced. In addition, since the edge calculation is adopted to replace the traditional server scheme to make the calculation to the front end to obtain the calculation result, only the AI calculation result is transmitted on the network data transmission instead of the video data, so that a large amount of bandwidth can be reduced, and the speed is increased.

摘要：【中文】本发明公开了一种级联式人脸图像优选方法，通过多任务级联卷积神经网进行人脸检测及坐标估计；过滤分辨率低于阈值的图片；过滤人脸姿态过歪的图片；通过分辨率得分、人脸头部姿态得分、人脸遮挡得分、表情得分、清晰度得分、光照强度得分，对图片进行综合分计算，综合分最高的图片为最优人脸；对人脸图像进行追踪，推送最优人脸，当最优人脸被更新时或者同一人脸连续预设数量帧内没有更新最优人脸且综合分大于预设分数时，推送最优人脸至识别模块。除了第一次出现人脸需要经过完整的优选步骤以外，在级联中途可以提前筛除不合格的图像，加快优选进度，减少内存资源占用。 【EN】The invention discloses a cascading human face image optimization method, which carries out human face detection and coordinate estimation through a multitask cascading convolution neural network; filtering pictures with resolution lower than a threshold value; filtering the picture with the facial posture being too askew; performing comprehensive score calculation on the pictures through resolution score, face head posture score, face shielding score, expression score, definition score and illumination intensity score, wherein the picture with the highest comprehensive score is the optimal face; the face image is tracked, an optimal face is pushed, and when the optimal face is updated or the optimal face is not updated in a continuous preset number of frames of the same face and the comprehensive score is larger than the preset score, the optimal face is pushed to the recognition module. Except for the fact that the complete optimization step is needed when the face appears for the first time, unqualified images can be screened out in the cascade midway, the optimization progress is accelerated, and the occupation of memory resources is reduced.

摘要：【中文】本发明属于人防工程防护设备技术领域，具体而言，本发明涉及一种城市地下综合管廊吊装口钢结构人防防护设备，包括门框、铰页、门扇、闭锁和电动驱动装置；所述门框安装在吊装口的基体上；所述门扇为钢结构门扇，通过所述铰页与所述门框铰接；所述闭锁安装在所述门扇上；所述电动驱动装置分别与所述闭锁、所述门扇相连。本发明可为综合管廊工程吊装口提供高效可靠的孔口防护措施，保障工程内部人员及设备安全，对于保存战时城市实力，提高重要城市输送管线的生存能力具有重要的指导意义和应用价值，在我国综合管廊建设中具有广阔的推广和应用前景，具有重大的战备效益和显著的经济效益。 【EN】The invention belongs to the technical field of civil air defense engineering protection equipment, and particularly relates to steel structure civil air defense protection equipment for a hoisting port of an urban underground comprehensive pipe gallery, which comprises a door frame, hinges, door leaves, a lock and an electric driving device, wherein the door frame is provided with a hinge; the door frame is arranged on the base body of the hoisting opening; the door leaf is a steel structure door leaf and is hinged with the door frame through the hinge; the lock is arranged on the door leaf; the electric driving device is respectively connected with the lock and the door leaf. The invention can provide efficient and reliable orifice protection measures for the lifting port of the comprehensive pipe gallery engineering, ensures the safety of personnel and equipment in the engineering, has important guiding significance and application value for saving the real capacity of the wartime city and improving the viability of the important urban conveying pipeline, has wide popularization and application prospects in the construction of the comprehensive pipe gallery in China, and has important combat readiness benefits and obvious economic benefits.

摘要：【中文】本发明提供一种用于评估基线模型的方法，包括如下步骤：提取在预定历史时间段内的每个时间点处的、由所述基线模型预测的第一基线和历史流量；基于所述第一基线的值和所述历史流量的值判断在所述预定历史时间段内的流量变化是否为正常流量变化；在所述流量变化是所述正常流量变化的情况下，基于所述第一基线的值和所述历史流量的值确定评估参数；以及基于所述评估参数确定是否需要训练所述基线模型。 【EN】The invention provides a method for evaluating a baseline model, comprising the steps of: extracting a first baseline and historical flow predicted by the baseline model at each time point within a predetermined historical time period; determining whether a flow change within the predetermined historical period of time is a normal flow change based on the value of the first baseline and the value of the historical flow; determining an evaluation parameter based on the value of the first baseline and the value of the historical flow rate if the flow rate change is the normal flow rate change; and determining whether training of the baseline model is required based on the evaluation parameters.

摘要：【中文】本发明涉及运维领域，尤其涉及一种告警信息的处理方法及装置，服务器获取当前告警信息对应的攻击类型、攻击目标标识和攻击记录信息，然后确定该攻击类型对应的n个攻击特征维度，其次，对攻击记录信息进行解析，确定与上述特征维度对应的特征值。针对n个攻击特征维度中的任意一个攻击特征维度的特征值，确定该特征值与该攻击特征维度对应的参考特征值是否一致，当n个攻击特征维度中至少一个攻击特征维度的特征值与所述攻击特征维度对应参考特征值不一致时，服务器确定当前告警信息为无效告警信息。该方法可以较为准确地识别出无效告警信息，以便于过滤无效告警信息，提高运维工作效率。 【EN】The invention relates to the field of operation and maintenance, in particular to a method and a device for processing alarm information. And aiming at the characteristic value of any one attack characteristic dimension in the n attack characteristic dimensions, determining whether the characteristic value is consistent with the reference characteristic value corresponding to the attack characteristic dimension, and when the characteristic value of at least one attack characteristic dimension in the n attack characteristic dimensions is inconsistent with the reference characteristic value corresponding to the attack characteristic dimension, determining that the current alarm information is invalid alarm information by the server. The method can accurately identify the invalid alarm information so as to filter the invalid alarm information and improve the operation and maintenance work efficiency.

摘要：【中文】本发明提供一种日志分析方法及装置，确定待测试日志属于N类参照集中的第一类参照集；根据该参照集的第一主题矩阵，确定待测试日志与该参照集的第一主题距离；根据该参照集的第一词向量矩阵和第一互斥矩阵，确定待测试日志与该参照集的第一词距离；根据第一主题距离和第一词距离，确定待测试日志与该参照集的互斥相似度；在互斥相似度不大于第一阈值时，确定待测试日志为异常日志。该方案通过将待测试日志与已知N类参照集进行匹配，可快速确定待测试日志所属的类型，极大缩短确定待测试日志正常与否的时间；在确定相似度的过程中，通过将词集间的互斥关系引入，对于确定待测试日志正常与否的问题，极大程度上提高了对其判断的准确性。 【EN】The invention provides a log analysis method and a log analysis device, which are used for determining whether a log to be tested belongs to a first type reference set in N types of reference sets; determining the first theme distance between the log to be tested and the reference set according to the first theme matrix of the reference set; determining a first word distance between the log to be tested and the reference set according to the first word vector matrix and the first mutual exclusion matrix of the reference set; determining the mutual exclusion similarity between the log to be tested and the reference set according to the first subject distance and the first word distance; and when the mutual exclusion similarity is not greater than a first threshold value, determining the log to be tested as an abnormal log. According to the scheme, the log to be tested is matched with the known N-type reference set, so that the type of the log to be tested can be quickly determined, and the time for determining whether the log to be tested is normal or not is greatly shortened; in the process of determining the similarity, the mutual exclusion relationship among the word sets is introduced, so that the accuracy of judging whether the log to be tested is normal or not is greatly improved.