Unknown DDoS Attack Detection with Fuzzy C-Means Clustering and Spatial Location Constraint Prototype Loss

Since its inception,the Internet has been rapidly evolving.With the advancement of science and technology and the explosive growth of the population,the demand for the Internet has been on the rise.Many applications in education,healthcare,entertainment,science,and more are being increasingly deployed based on the internet.Concurrently,malicious threats on the internet are on the rise as well.Distributed Denial of Service(DDoS)attacks are among the most common and dangerous threats on the internet today.The scale and complexity of DDoS attacks are constantly growing.Intrusion Detection Systems(IDS)have been deployed and have demonstrated their effectiveness in defense against those threats.In addition,the research of Machine Learning(ML)and Deep Learning(DL)in IDS has gained effective results and significant attention.However,one of the challenges when applying ML and DL techniques in intrusion detection is the identification of unknown attacks.These attacks,which are not encountered during the system’s training,can lead to misclassification with significant errors.In this research,we focused on addressing the issue of Unknown Attack Detection,combining two methods:Spatial Location Constraint Prototype Loss(SLCPL)and Fuzzy C-Means(FCM).With the proposed method,we achieved promising results compared to traditional methods.The proposed method demonstrates a very high accuracy of up to 99.8%with a low false positive rate for known attacks on the Intrusion Detection Evaluation Dataset(CICIDS2017)dataset.Particularly,the accuracy is also very high,reaching 99.7%,and the precision goes up to 99.9%for unknown DDoS attacks on the DDoS Evaluation Dataset(CICDDoS2019)dataset.The success of the proposed method is due to the combination of SLCPL,an advanced Open-Set Recognition(OSR)technique,and FCM,a traditional yet highly applicable clustering technique.This has yielded a novel method in the field of unknown attack detection.This further expands the trend of applying DL and ML techniques in the development of intrusion detection systems and cybersecurity.Finally,implementing the proposed method in real-world systems can enhance the security capabilities against increasingly complex threats on computer networks.

Real-Time Recognition and Location of Indoor Objects

Object recognition and location has always been one of the research hotspots in machine vision.It is of great value and significance to the development and application of current service robots,industrial automation,unmanned driving and other fields.In order to realize the real-time recognition and location of indoor scene objects,this article proposes an improved YOLOv3 neural network model,which combines densely connected networks and residual networks to construct a new YOLOv3 backbone network,which is applied to the detection and recognition of objects in indoor scenes.In this article,RealSense D415 RGB-D camera is used to obtain the RGB map and depth map,the actual distance value is calculated after each pixel in the scene image is mapped to the real scene.Experiment results proved that the detection and recognition accuracy and real-time performance by the new network are obviously improved compared with the previous YOLOV3 neural network model in the same scene.More objects can be detected after the improvement of network which cannot be detected with the YOLOv3 network before the improvement.The running time of objects detection and recognition is reduced to less than half of the original.This improved network has a certain reference value for practical engineering application.

A Novel and Efficient Method for Iris Automatic Location

An efficient and robust iris location algorithm plays a very important role in a real iris recognition system. A novel and efficient iris automatic location method is presented in this study. It includes following two steps mainly： pu- pil location and iris outer boundary location. A digital eye image was divided into many small rectangular blocks with fixed size in the pupil location, and the block with the smallest average intensity was selected as a reference area. Then image binarization was implemented taking the average intensity of the reference area as a threshold. At last the center coordinates and radius of pupil were estimated by extending the reference area to the pupil＇s boundaries in the binary iris image. In the iris outer location, two local parts of the eye image were selected and transformed into polar coordinates from Cartesian reference. In order to detect the fainter outer boundary of the iris quickly, a novel edge detector was used to locate boundaries of the two parts. The center coordinates and radius of the iris outer boundary can be estimated using the fusion of the locating results of the two local parts and the location information of the pupil. The algorithm was tested on CASIA vl.0 and MMU vl.0 digital eye image databases and experimental results show that the proposed method has satisfying performance and good robustness.

Vehicle Real-time Location Based on Visual Perception Model

Vehicle recognition system (VRS) plays a very important role in the field of intelligent transportation systems.A novel and intuitive method is proposed for vehicle location.The method we provide for vehicle location is based on human visual perception model technique. The perception color space HSI in this algorithm is adopted.Three color components of a color image and more potential edge patterns are integrated for solving the feature extraction problem.A fast and automatic threshold technique based on human visual perception model is also developed.The vertical edge projection and horizontal edge projection are adopted for locating left-right boundary of vehicle and top-bottom boundary of vehicle, respectively. Very promising experimental results are obtained using real-time vehicle image sequences, which have confirmed that this proposed location vehicle method is efficient and reliable, and its calculation speed meets the needs of the VRS.

Scene recognition for mine rescue robot localization based on vision

A new scene recognition system was presented based on fuzzy logic and hidden Markov model(HMM)that can be applied in mine rescue robot localization during emergencies.The system uses monocular camera to acquire omni-directional images of the mine environment where the robot locates.By adopting center-surround difference method,the salient local image regions are extracted from the images as natural landmarks.These landmarks are organized by using HMM to represent the scene where the robot is,and fuzzy logic strategy is used to match the scene and landmark.By this way,the localization problem,which is the scene recognition problem in the system,can be converted into the evaluation problem of HMM.The contributions of these skills make the system have the ability to deal with changes in scale,2D rotation and viewpoint.The results of experiments also prove that the system has higher ratio of recognition and localization in both static and dynamic mine environments.

Automatic Location of Main Facial Features in Front-View Images

This paper presents a set of algorithms capable of locating main facial features automatically and effectively. Based on integral projection of local binary image pixels and pixel clustering techniques, a set of a priori knowledge based algorithms have succeeded in locating eyes, nose and mouth, and uprighting the tilt face. The proposed approach is superior to other methods as it takes account of photos with glasses and sha dows, therefore suitable for processing real ID type photos.

小兴安岭南麓野猪春冬季行为状态的生境选择

野猪(Sus scrofa)造成的人兽冲突是我国目前野生动物管理中急需解决的问题之一,但往往管理成本高,却效果甚微。了解野猪觅食、卧息等重要行为状态的生境选择规律,对科学管理策略的制定及避免人兽冲突极为重要。以小兴安岭南坡农林混交区域的野猪(5只雌性,8只雄性)为研究对象,基于其2019—2022年春冬季卫星定位项圈追踪数据,运用隐马尔科夫模型(HMM)识别野猪觅食、卧息和移动的三种行为状态,然后用步选择函数(SSF)研究这三种行为状态的生境选择。结果表明:野猪觅食、卧息和移动的生境需求不同,但均回避落叶针叶林,且性别差异明显,雌性野猪回避农田;地形是影响雌性野猪春季和雌雄野猪冬季三种行为状态生境选择的主要因子,地形和植被类型是影响雄性野猪春季三种行为状态生境选择的主要因子;雌性野猪,在春季,选择海拔高、坡度小的生境卧息、觅食及移动,且卧息和觅食均偏爱阴坡,冬季则选择海拔高,坡度大的阳坡卧息及觅食,偏爱在混交林和阳坡移动;雄性野猪,在春季,选择在混交林和坡度大的阴坡卧息,在混交林、落叶阔叶林和海拔低的区域觅食,在落叶阔叶林、常绿阔叶林内、海拔低、坡度小和距河流近的区域移动,在冬季,选择坡度大、海拔高、接近河流的阳坡卧息,在距河流近、坡度小的区域觅食,在常绿阔叶林、坡度小和距河流近的阳坡移动。本研究结果可为该区域的野猪毁田的预测和预警,捕获地点的选择等管控措施的制定提供依据和新思路。

新安煤矿人员定位系统升级改造的研究与应用

煤矿井下人员定位系统是加强入井人员管理、提升生命保障、促进煤矿安全生产的高端技术。新安煤矿在用的KJ251型人员定位系统安装于2011年,采用RFID定位技术实现区域定位,该技术具有实现简单、系统建设成本低的特点。随着煤矿井下智能化的推进,对人员定位提出了更高的需求,现主要存在浏览器内核版本、系统中心站软件及数据库架构满足不了当前信息化发展需求、系统数据不稳定和系统未配备唯一检卡装置等问题。结合煤矿井下特殊的作业环境,煤矿井下人员定位系统可实现对入井人员的实时监测、精确定位、轨迹回放、考勤管理、报表查询、信息网络发布、双向通信、人机交互、紧急搜救、生产调度等功能,为煤矿安全生产以及紧急救援提供第一手可靠的决策实时信息。

康乃馨扦插机器人设计与试验

针对康乃馨自动化标准扦插作业需求,解决现有人工扦插模式劳动强度大的问题,该研究设计了一种抓取-扦插一体化作业的扦插机器人系统。以康乃馨生根用插穗为对象,根据插穗物料特性参数和高产栽培农艺要求,提出辅助栽植式末端执行器作业方式和抓插协同运动规划模型,基于YOLOv5s深度学习网络构建多目标插穗检测模型,建立扦插机器人运动学模型,提出基于MD-H(modified denavit-hartenberg)规则的逆运动学建模方法和时间最优轨迹规划算法。仿真结果表明,改进后的逆解算法和轨迹规划算法的运行时间与常规算法相比分别降低了38.3%和72.5%。搭建扦插机器人试验台,进行定位误差及整机性能试验。试验结果表明,抓取作业时X、Y方向的平均定位误差分别为2.33和2.38 mm;扦插作业时X、Y、Z方向的平均定位误差分别为2.21、2.25和2.05 mm;机器人单株平均作业总时间为14.787 s,其中识别抓取平均工作周期为6.803 s,占作业总时间的46.01%,扦插作业的平均工作时间为5.104 s,占作业总时间的34.52%;插深合格率为88%,扦插直立度为92%,漏插率和根部损伤率分别为5%和2%,基本满足自动化标准扦插作业要求。研究结果可为温室标准化扦插设备的研发与应用提供参考。

基于Kirsch算子与数学形态学的马铃薯图像芽眼识别与定位方法

马铃薯图像芽眼识别是马铃薯自动切块实现大面积种植的前提,由此提出一种基于Kirsch算子与数学形态学的马铃薯图像芽眼识别与定位方法。对马铃薯图像进行灰度化、中值滤波处理,采用Kirsch算子快速算法进行边缘检测,结合数学形态学强大的处理功能巧妙地剔除马铃薯边界,实现马铃薯芽眼区域的识别与快速定位。实验表明,该方法识别率高,芽眼区域标记完整、简单快捷,具有较好的实时性和准确性,可为后续马铃薯芽眼智能化切割提供借鉴与参考。

基于3S技术下人口普查信息化建设探讨

第七次全国人口普查工作首次引入空间地理位置,采用信息化采集手段推进此次人口普查工作的开展。天津市第七次人口普查工作是在结合天津市行政区划、建筑物分布特点,人口流动情况等,充分利用3S、物联网、数据库集群、模糊识别等技术,解决了基于空间地理位置环境下建筑物快速提取、人口信息电子化采集的问题,形成了一套高效人口普查信息获取工作机制,有力地保证了天津市第七次人口普查和户籍整顿工作的顺利实施。

基于隐含空间模型降维和LDA模型的学科主题识别研究

【目的/意义】现有学科研究主题的梳理多为领域专家的定性分析和学科学者的文献梳理,一定程度会由于研究思维的局限性和获取知识的片面性造成学科研究主题误判,为有效避免漏判误判现象的发生,提出识别模型。【方法/过程】首先,运用传统LDA模型分析主题特征词;其次,结合上下文语义信息进行中文分词,形成学科主题词库;最后,结合隐含位置聚类算法发现潜在社区,提高主题识别效果。【结果/结论】提出的方法一定程度上优化了主题挖掘算法在识别短文本主题的效果,消除主观意愿。由计算机自行分类并实现科学研究前沿主题的预测,揭示前沿领域的研究热点,为致力于研究前沿学科的新兴学者提供参考价值。

基于YOLOv3不同场景辣椒采摘机器人识别定位研究

针对辣椒采摘受环境光、枝叶遮挡和果实重叠的影响问题,构建了基于YOLOv3模型和realsense深度相机的识别定位系统,研究不同补光位置、枝叶遮挡和果实重叠程度对辣椒识别和定位精度的影响规律。结果表明:模型召回率Recall达0.98,平均精度均值mAP达0.95,精确率precision达0.854;不同光照场景下,识别成功率由高到低依次为正向光、顶光、侧光和背光;轻度枝叶遮挡和轻微果实重叠时,模型识别成功率均保持在96%左右,综合定位误差最大为0.024m,满足辣椒采摘机器人识别和定位精度需求。

基于ERNIE及改进DPCNN的棉花病虫害问句意图识别

针对目前没有公开的棉花病虫害相关问句数据集且问句较短、类型多样等问题,本研究通过查阅文献及咨询相关领域专家,构建了棉花病虫害问句数据集CQCls,定义了78种棉花病虫害实体和9种问句类型;同时提出了一种基于ERNIE预训练模型的棉花病虫害问句意图识别模型,首先通过ERNIE模型将输入问句映射到向量空间,使用融合词位置信息的DPCNN模型进行特征向量的抽取,与基础的DPCNN模型相比,通过融合词位置信息能有效提高模型的表达能力,然后经过Softmax得到最终结果。实验结果表明,本研究提出的意图识别模型相较于其他模型取得了较好的结果,宏平均和加权平均的F1分数值分别为97.45%和97.31%;在文本语料数据内容复杂多样且文本格式不规范的DMSCD数据集上,训练结果中不同类别的F1分数的权重平均也能达到73.42%,进一步证明了该模型的有效性及泛化能力。

基于点云轨迹和压缩多普勒的跨场景手势识别

毫米波雷达能够用于各种感知任务,如活动识别、手势识别、心率感知等。手势识别作为其中的研究热点,可实现无接触人机交互。目前大多数手势识别研究使用点云或距离多普勒图通过神经网络进行识别感知,但是这些方法存在一些问题。首先,这些方法鲁棒性较差,被感知人员或其位置发生变化都会影响接收到的毫米波信号,降低感知精度。其次,这些方法将完整的距离多普勒图输入神经网络进行识别,由于图中存在较多与感知任务无关的区域,模型复杂且难以专注于感知任务。为解决这些问题,首先从连续多帧点云数据中建立手势轨迹,然后将连续多帧距离多普勒图进行局部切割并压缩获得二维局部多普勒图,最后将点云轨迹和二维局部多普勒图分别经过神经网络特征提取后,对特征进行拼接,通过全连接神经网络进行分类。实验结果表明,所提方法专注于手势,能够达到98%的识别准确率,在人员变化和位置变化情况下对新用户和在新位置的识别准确率分别能够达到93%和92%,高于现有方法。

基于实例分割的大场景下茶叶嫩芽轮廓提取与采摘点定位

为了解决野外大场景下茶叶嫩芽识别与采摘点精确定位问题,提出了一种基于实例分割的Yolov5s-segment改进算法。该算法首先引入P2微小目标检测层,解决原始Yolov5s-segment网络P3、P4、P5检测层对于小目标检测能力不佳的问题。其次,在主干网络末端增加CBAM(convolutional block attention module)注意力机制模块,提升模型的抗干扰能力,实现野外自然光照环境下茶叶嫩芽轮廓特征提取。最后,根据嫩芽轮廓特征进行采摘点精确定位。研究结果表明,相较于原始Yolov5s-segment模型,改进模型的精确度、召回率、F1分数、平均精度均值mAP50和mAP50-95分别提升了7.0、8.9、8.1、8.3、7.3个百分点。使用该方法可以准确提取大场景下的单芽、一芽一叶、一芽两叶三种类型茶叶嫩芽轮廓,并且实现采摘点的精确定位。研究结果为名优茶智能化快速采摘提供了一定的理论基础。

基于BERT位置感知的旅游三元组知识抽取方法

直接获取文本中的三元组,往往存在语义联系较弱、距离过长和一词多义的问题,因此,提出基于BERT预训练的位置感知的两阶段旅游三元组知识抽取方法。利用BERT-Span模型通过边界预测方法对旅游实体进行识别;利用旅游数据中的字、语义、位置和实体类型特征,构建融合位置感知注意力和头尾实体类型的关系抽取模型。在山西旅游数据集上进行实验,实验结果表明提出的方法优于基准模型的F1值。

一种级联式高压开关柜局部放电定位方法

针对高压开关柜局部放电定位技术存在误差大、效率低、难以实现定位到柜的问题,文中基于开关柜中局部放电TEV信号进行了定位技术研究,提出了TEV传感器级联式定位方法。通过分布式传感器有线级联的方式、双母线分时比较方法与TEV传感器状态开关切换策略,使单个双通道采集卡能对多面开关柜上的TEV传感器信号脉冲进行采集比较,从而快速、高效、准确地在成排高压开关柜中定位出局部放电源所在的开关柜。基于该方法设计了高压开关柜局部放电级联式定位系统,采用总线级联式的模块化设计,不需增加高速采集通道数量即可满足任意扩展TEV检测节点数量的需求,最大化地减少了系统成本。通过搭建验证平台验证了该定位方法的定位效果,证明了级联式定位系统的工作有效性,实现了局部放电定位到柜的目的。

基于多传感器融合的电力变压器内部放电定位与辨识技术

针对电力设备状态监测中传感器信息融合的难题,应用油中溶解气体、特高频、光纤超声等监测传感器构建了适用于变压器的融合传感阵列,提出了变压器内部放电的辨识策略。采用超声信号到达时间差估算实现了放电源定位,建立了多传感器监测特征融合算法,并结合K邻近(K nearest neighbors,KNN)算法实现了放电模式辨识。实验结果表明:放电源定位平均误差为45.1 mm,放电类型辨识准确率达到90.6%;所构建的多传感器融合感知策略与系统,可实现变压器内部放电源的准确定位与辨识,具有一定的应用推广价值。

基于机器视觉的岩块自动化识别检测方法

在隧道施工过程中,岩体参数的获取是实现隧道掘进机参数调整和智能决策的前提,因此,要对掘进过程中获得的岩块进行采样和检测,而岩块识别和检测目前主要由人工完成。本文针对岩块的自动化识别和检测问题,提出了一种基于机器视觉的岩块自动化识别检测方法,通过融合岩块区域检测和语义分割算法能够快速准确获取岩块的形心坐标和过形心最小直径。首先,使用YOLOv3网络对岩块进行识别,实现岩块的区域检测。其次,针对每个区域的岩块采用FCN-DenseNet网络进行语义分割和图像处理,并对全卷积神经网络进行改进,减少了语义分割模型的参数量,提高了语义分割效率,提升了岩块轮廓获取的精度和速度。最后,根据获得的岩块轮廓点,计算其形心坐标及过形心的最小直径,为机械臂抓取和岩块点荷载强度的计算提供支持。搭建实验平台,完成机械臂手眼标定和深度相机坐标下岩块图像与岩块点云对齐,获取岩块形心坐标在机械臂坐标下的位置。实验结果表明,本文所提算法能够快速准确地获取岩块的形位参数,对10次实验中的102块岩块识别检测成功率为91.18%,在所有完成识别检测岩块中的吸取成功率为92.47%,可以应用于岩体的自动化检测,提高岩体检测的效率和智能化水平。