Virtual edge based coverage hole detection algorithm in WSN(WCNC'13)
基于深度学习在海缆表面缺陷检测中的应用
基于深度学习在海缆表面缺陷检测中的应用作者：张浩吴陈徐影来源：《电脑知识与技术》2022年第15期摘要：傳统的海缆表面缺陷检测是通过人工观察的方式，易出现高错误率、速度较慢等现象。
为此，文章提出了一种基于深度学习在海缆表面缺陷检测的系统。
为了提高整体的检测效率，该系统通过Faster - RCNN网络对图片缺陷进行检测。
为了减小模型的复杂程度，采用了RPN Loss+ Fast RCNN Loss的联合训练方法，通过一次的反向传播更新模型参数。
同时，为了增强模型对海缆表面缺陷的检测能力，在主干特征提取网络部分，利用Resnet50中四个不同阶段的特征层构建特征金字塔网络，以适应不同缺陷的检测。
实验结果表明，缺陷检出率达到98%，分类准确率达到97.9%。
经过验证，该系统检测效果优于传统方式，可以满足工业化生产要求。
关键词：深度学习;海缆;表面缺陷;检测中图分类号：TP311 文献标识码：A文章编号：1009-3044（2022）15-0088-04海缆是用绝缘体包裹的电缆或光缆，铺设在海底的一种通信线路[1]。
因为需要特殊的工作环境，所以需要更加严格的标准。
对于传统的人工检测方法，受到有限的空间与时间限制，存在很多不足之处。
通常人眼能较好地辨别缺陷尺寸需要大于0.5mm，且需要较大的光学形变（64灰度级），同时当待测物体的运动速度到达3m/s的临界点时，人眼将不能够发现缺陷存在的具体形态[2];同时，必须在强光源下才能快速找到海缆表面的缺陷;加之长时间工作，极易引起眼疾，而且会产生比较高的误检率及漏检率。
因此，在现实工业生产中急需一种改善传统检测的方案。
1 相关工作得益于现代图像处理和模式识别等技术的发展，以机器视觉的表面缺陷检测方法在工业产品品质把控环节中得到了较为广泛的应用。
文献[3]中分析了机器视觉瑕疵检测精度可达到微米级，机器快门采集时间可以达到微秒级，识别和判断都能够保持稳定，使得检测效率得到提升。
整个超构透镜器件的仿真边界条件
整个超构透镜器件的仿真边界条件下载提示：该文档是本店铺精心编制而成的，希望大家下载后，能够帮助大家解决实际问题。
文档下载后可定制修改，请根据实际需要进行调整和使用，谢谢!本店铺为大家提供各种类型的实用资料，如教育随笔、日记赏析、句子摘抄、古诗大全、经典美文、话题作文、工作总结、词语解析、文案摘录、其他资料等等，想了解不同资料格式和写法，敬请关注!Download tips: This document is carefully compiled by this editor. I hope that after you download it, it can help you solve practical problems. The document can be customized and modified after downloading, please adjust and use it according to actual needs, thank you! In addition, this shop provides you with various types of practical materials, such as educational essays, diary appreciation, sentence excerpts, ancient poems, classic articles, topic composition, work summary, word parsing, copy excerpts, other materials and so on, want to know different data formats and writing methods, please pay attention!超构透镜器件仿真边界条件的探究在超构透镜器件的设计与仿真过程中，正确设置仿真边界条件至关重要。
利用遗传算法设计可见光波段全能反射器(英文)
利用遗传算法设计可见光波段全能反射器(英文)蒋立勇;郑改革;董秋云;李相银【期刊名称】《光子学报》【年(卷),期】2009(38)6【摘要】介绍了遗传算法在设计宽频带全能反射器中的应用.设计的关键是通过遗传算法寻找到不同一维准周期光子晶体间的最佳组合方式以及光学厚度以便形成一个光子晶体异质结.利用传输矩阵法分析了一维准周期系统中的电磁传输特性.计算结果表明,准周期光子晶体的全方向反射带宽受晶胞单元以及周期数的影响.根据这些规律,用遗传算法成功地优化了光子晶体异质结的结构并得到两种适用于可见光波段的高性能全能反射器.例如结构为(HLLHL)13(HL)13(LHL)15[注:nL=1.46,nH=2.6,dL=0.218λ0/nL且dH=0.201λ0/nH]的反射器在可见光波段内的全方向反射带宽达到了88.42%(446nm～779nm).【总页数】6页(P1432-1437)【关键词】全能反射器;准周期;一维光子晶体;传输矩阵法;遗传算法【作者】蒋立勇;郑改革;董秋云;李相银【作者单位】南京理工大学理学院,南京210094;解放军理工大学理学院,南京211101【正文语种】中文【中图分类】O411.3【相关文献】1.基于一维光子晶体的可见光波段大角度反射器研究 [J], 邹泽华;李祥霞;田东康;杨毅彪;费宏明2.一种利用可见光波段无人机遥感的林下植被覆盖识别方法 [J], 朱雄斌;汪小钦;周小成3.可见光激光二极管开发现状与市场——600nm波段可见光LD [J], 何兴仁4.利用SPR效应测定可见光波段金膜折射率 [J], 曹振新5.利用MODIS可见光波段反演陆地气溶胶光学厚度 [J], 郭强;唐家奎;何文通;田媛;于新菊因版权原因，仅展示原文概要，查看原文内容请购买。
减小光刻中驻波效应的新方法研究
减小光刻中驻波效应的新方法研究
肖啸;杜惊雷;郭永康;杨静;谢世伟
【期刊名称】《微细加工技术》
【年(卷),期】2002(000)004
【摘要】光刻过程中,抗蚀剂内部光敏混合物(PAC)浓度受光场的影响呈驻波分布,导致抗蚀剂显影后的侧壁轮廓成锯齿状.分析了后烘(PEB)对PAC浓度分布的影响,模拟了不同后烘扩散长度下的抗蚀剂显影轮廓,从模拟结果可知利用后烘可明显减小驻波效应,得到平滑的抗蚀剂显影轮廓,提高光刻质量.
【总页数】5页(P36-39,44)
【作者】肖啸;杜惊雷;郭永康;杨静;谢世伟
【作者单位】四川大学物理科学与技术学院,成都,610064;四川大学物理科学与技术学院,成都,610064;四川大学物理科学与技术学院,成都,610064;四川大学物理科学与技术学院,成都,610064;四川大学物理科学与技术学院,成都,610064
【正文语种】中文
【中图分类】TN305.7
【相关文献】
1.微光刻中的驻波效应研究 [J], 冯伯编;张锦
2.电子束光刻中的相互邻近效应校正技术研究 [J], 赵真玉;宋会英
3.柔性隔振系统中驻波效应的研究 [J], 熊冶平;宋孔杰
4.四频差动激光陀螺减小磁效应的新方法 [J], 杨建强;廖丹;朱勇
5.光刻中驻波效应的影响分析 [J], 刘世杰;杜惊雷;肖啸;唐雄贵;彭钦军;刘建莉;郭永康
因版权原因，仅展示原文概要，查看原文内容请购买。
Fortinet高级威胁保护解决方案介绍说明书
FORTINET ADVANCED THREAT PROTECTIONINTRODUCTIONSOPHISTICATED ATTACKS YIELD BIG REWARDSSecuring today’s enterprise is more challenging than ever. Whereas in the past there was a well-defined perimeter to secure, the rise of the Internet of Things (IoT) following mobility and BYOD, as well as the continued adoption of public following widespread adoption of private cloud services, results in a much more diverse and dynamic attack surface to be protected.At the same time, the threat landscape continues to evolve in regard to both volume and sophistication as cyber crime has achieved big business status and maturity. Powered by a robust cyber crime ecosystem that includes a growing segment of “malware-as-a-service” providers, our FortiGuard Labs sawmore than 700,000 intrusion attempts by the close of 2016, with 120,000 pieces of malware and 25,000 spam messages—every minute of every day. Further, in Verizon’s 2016 Data BreachInvestigations Report, the malware at the heart of incidents lived for only 58 seconds or less and was seen only at the compromised organization in nearly all cases. Most importantly, regardless ofwhether the result of a volume-based or targeted attack, a recent Fortinet Threat Landscape publication reported that the average organization has been successfully compromised with more than six active bots communicating out to cyber criminals.That’s why Fortinet is pioneering a new approach to security, our Fortinet Security Fabric, which includes specific components recommended to address today’s advanced threats.THE FORTINET SECURITY FABRICWith so many potential ways for cyber criminals to gain entry into the dynamic enterprise, it is important to design and implement a security architecture that is broad enough to cover the entire attack surface. Further, it is critical to have security components that are powerful enough to enable all the technologies appropriate at each protection point without slowing networks or employees. And finally, it must be automated and work as a single, cohesive system to keep pace with the changing and fast-moving threat landscape.NOC/SOCCloudPartner APIAccessClient/IoTAdvanced ThreatIntelligenceApplicationBroad and Automated Security Coverage for your Expanding Attack Surfacealong with the latest emerging technologies, as they demonstrate their effectiveness, working together to stop attacks at multiple points of the organization and multiple phases of their life cycle. As such, the security components recommended as the basis for advanced threat protection span three primary phases:n n Prevention—blocking, as much as possible, typically knownthreats, often based on global intelligencen n Detection—continuing inspection, usually for unknown threatsbased on local analysis and intelligencen n Mitigation—responding with new detection to minimize incidentsand their impact, primarily by sharing intelligence throughout thesecurity architectureFurther, this approach must incorporate current security capabilities, emerging technologies, and customer-specificlearning mechanisms to create and distribute actionable security intelligence from newly detected threats in real time. And it must coordinate among security components from multiple vendors, such that the entire infrastructure can act as a single entity to protect the organization.detection components must be able to inspect traffic, objects, and user activity from the endpoint (including IoT) and access layer to the network edge and core, all the way out to the public cloud. Further, it must cover the top attack vectors including the:n n Network Layer—encompassing 3 of the top 5 vectors for cybercrime as reported by Verizonn n Application Layer—both email, which was the #1 and #3 vectorfor cyber crime as reported by Verizon, and web infrastructure, which was the top technique leading to compromise in many industriesn n Endpoint Layer—the only place to protect against off-networkattacksFortinet recommends our FortiGate next-generation firewall as the foundation of the Security Fabric, spanning the network from branch office and campus, the data center edge and core, and all the way out to workloads in public clouds. From there, coverage can be extended to endpoints with our FortiClient AdvancedEndpoint Protection software or Fabric-ready partners like Carbon Black, Ziften, and others. It can also be deepened with enterprise-class protection for email and web infrastructure with FortiMail and FortiWeb. In this manner, the top attack vectors and potential entry points can be covered with global threat intelligence. To address the fast-moving and more targeted threats, werecommend adding the local intelligence of our FortiSandbox with additional options from Fabric-ready partners like Attivo Networks and their deception infrastructure.NETWORK LAYERAPPLICATION LAYERENDPOINT LAYERvulnerability management, virtual private networking, two-factor authentication, and more to limit and control access are also available with our full FortiClient stack.POWERFUL PROCESSING TOENABLE THE SECURITY YOUNEEDNote that many of the securitytechnologies required to prevent ordetect advanced threats require deeperor more time-consuming analysis,yet they simply cannot be allowed to impede or even slowthe business. As a result, it is essential to accelerate, either in hardware, software, or cloud services, these functions so they can be enabled (and not turned off) and improve the organization’s security posture. Examples of these more rigorous, yet potentially hindering, security functions include anti-malware inspection on network traffic, sandbox analysis on either the network or endpoint, more advanced behavioral techniques on the endpoints, and similar technologies.Fortinet FortiGate appliances include our proprietary security processors for network traffic (our NP chips), content inspection (CP chips), and a combined system on a chip (SoC) to ensurethat all necessary FortiGate security features can be enabled on properly sized appliances to stop threats seeking entry, from the smallest remote office to the largest data center and all points in between. These features include the full next-generation firewall stack of intrusion prevention, application control, web filtering,anti-malware, SSL inspection, integrated sandboxing, and more. Further, our FortiGate virtual appliances have been optimized for cloud-scale performance in the world’s largest IaaS and PaaS environments to extend advanced threat protection out to the public cloud.AUTOMA TED TO ACT AS ASINGLE SYSTEMCovering the breadth of the organizationwith security components powerfulenough to enable the necessary securityfunctions are simply the building blocksto improve security. If these components operate independently from each other, there will be gaps between them through which cyber criminals can slip through and silos that will slow response and mitigation when that happens. Remember, in the absence of ties between the products, it falls to the security team to manually bridge gaps and coordinate responses—an inherently time-consuming and less-effective exercise. 1. prevention products sharing objects with detection productsfor the deeper analysis2. detection products distributing threat intelligence based onthat analysis for updated prevention3. response elements immediately accessing detection andprevention products for faster mitigationAnd this automation must occur across offerings from different vendors, as it is rare to find components for each protection point and attack vector that meet every organization’s requirements and budget.That’s why Fortinet has developed six primary APIs, including two specifically for local intelligence sharing from our FortiSandbox for a coordinated defense. These APIs are documented and openfor use by Fabric-ready partners with precertified integrations and indeed any third-party security product that willleverage them.INDEPENDENTL Y TESTEDThe ultimate measure of the securityinfrastructure that you have in place is itseffectiveness in preventing, detecting, andquickly responding to mitigate attacks in real-world deployment. And while most securityofferings will have dashboards showing what’s been detected, they are not going to show what’s been missed. That’s where independent, real-world testing can surely help. There are a number of credible, independent test houses that regularly test a range of security vendor offerings at various protection points and attack vectors. These include NSS Labs, Virus Bulletin, ICSA Labs, and AV-Comparatives.Fortinet security products have undergone the most independent testing, and earned the most top ratings, of any vendors’ offerings in the industry. Among them, Fortinet is the only vendor to earn:nn NSS Labs Recommendation for NGFW, DCIPS, WAF, AEP, and BDS;nn ICSA Labs ATD Certification (both standard and email) from the inception; andnn Virus Bulletin VBSpam+ rating for the full year of 2016As our co-founder and CTO Michael Xie asserted years ago and remains committed to today, “Fortinet is a strong believer in third-party testing. We are always willing to put our products on the line against our colleagues and competitors in the security sphere... because to us, regardless of where we finish, we take the results and use them to create a better product.”Copyright © 2017 Fortinet, Inc. All rights reserved. Fortinet , FortiGate, FortiCare and FortiGuard , and certain other marks are registered trademarks of Fortinet, Inc., and other Fortinet names herein may also be registered and/or common law trademarks of Fortinet. All other product or company names may be trademarks of their respective owners. Performance and other metrics contained herein were attained in internal lab tests under ideal conditions, and actual performance and other results may vary. Network variables, different network environments and other conditions may affect performance results. Nothing herein represents any binding commitment by Fortinet, and Fortinet disclaims all warranties, whether express or implied, except to the extent Fortinet enters a binding written contract, signed by Fortinet’s General Counsel, with a purchaser that expressly warrants that the identified product will perform according to certain expressly-identified performance metrics and, in such event, only the specific performance metrics expressly identified in such binding written contract shall be binding on Fortinet. For absolute clarity, any such warranty will be limited to performance in the same ideal conditions as in Fortinet’s internal lab tests. Fortinet disclaims in full any covenants, representations, and guarantees pursuant hereto, whether express or implied. Fortinet reserves the right to change, modify, transfer, or otherwise revise this publication without notice, and the most current version of the publication shall be applicable. Fortinet disclaims in full any covenants, representations, and guarantees pursuant hereto, whether express or implied. Fortinet reserves the right to change, modify, transfer, or otherwise revise this publication without notice, and the most current version of the publication shall be applicable.GLOBAL HEADQUARTERS Fortinet Inc.899 Kifer RoadSunnyvale, CA 94086United StatesTel: +/salesEMEA SALES OFFICE 905 rue Albert Einstein 06560 Valbonne FranceTel: +33.4.8987.0500APAC SALES OFFICE 300 Beach Road 20-01The Concourse Singapore 199555Tel: +65.6513.3730LATIN AMERICA HEADQUARTERS Sawgrass Lakes Center13450 W. Sunrise Blvd., Suite 430Sunrise, FL 33323Tel: +1.954.368.9990prevention and detection over time. As a result, organizations benefit from:Comprehensive Security: FortiGuard Labs leverages real-time intelligence across 12 distinct disciplines within the threat landscape to deliver comprehensive security updates across the full range of Fortinet solutions and core technologies for synergistic protection.appliances also acting as sensors, FortiGuard Labs has established more than 200 threat information-sharing agreements with other recognized vendor research groups, ISACs, and industry groups like the Cyber Threat Alliance, with a mature process for theautomated ingestion, deduplication, and validation to turn raw data into high-value threat intelligence.For more information about Fortinet and the products that comprise the Advanced Threat Protection Solution, please visit /atp.Web SecurityFootprintWeb Filtering。
Mist Learning WLAN Mist Cloud Mist Access Points商业
The Mist Learning WLANMist CloudMist Access PointsMachine learning for optimal user experiencesMist’s patent-pending machine learning algorithms adapt in real-time to changes in user, device, and application behavior. This ensures predictable and reliable Wi-Fi and enables accurate Bluetooth® Low Energy (BLE) location-based services.Machine learning for proactive IT operationsMist’s machine learning algorithms monitor network trends inreal-time and intelligently alert IT when service levels degrade below defined thresholds. Recommendations are provided for rapid troubleshooting and/or proactive configuration changes.Built on modern cloud elements• Micro-services for SaaS agility (i.e. fast, risk free changes)• Containers ensure portability and fault tolerance • Big data collection / analysis from every wireless user• Built for speed and reliability using modern cloud elements like Storm, Spark, and KafkaSaaS agility for upgrades and patchesThe Mist cloud is built on a micro-services architecture that brings the agility of SaaS to wireless networks. On-demand network upgrades and patches take minutes instead of months.The Mist platform is the first truly new approach to wireless in nearly a decade. We started with a clean sheet of paper and deisgned a purpose-built cloud with the following unique features:Mist is leading the convergence of Wi-Fi, Bluetooth® LE and IoT. Key to this are our enterprise-grade APs, whichinclude the following unique features.Mist Access PointsAP61AP41AP21BT11DeploymentOutdoor Indoor Indoor Indoor Wi-Fi802.11ac Wave24x4:4802.11ac Wave24x4:4802.11ac Wave22x 2:2–Wi-Fi Tri-Radio –IoT Interface–––Antenna Options Internal/ExternalInternal/ExternalInternalInternalVirtualBluetooth® LEWarranty1 Year Limited Lifetime Limited Lifetime Limited LifetimeA.I. is in the AIR ™Best Wi-Fi and Bluetooth® LE performanceIn addition to delivering the best Wi-Fi 802.11ac Wave 2 range and performance, Mist APs have a dynamic vBLE 16 antenna array for the industry’s most accurate location services.Data collection, analysis and enforcementAll Mist APs have extra radios for collecting data and enforcing policies in conjunction with the Mist cloud, which is critical when doing analytics, machine learning, location services, and event correlationA single, enterprise-grade platform for Wi-Fi, Bluetooth® LE, and IoTMist APs also have an expansion port for Internet of Things (IoT). This makes Mist APs a single, enterprise-class platform for all wireless connectivity options.Mist built the first AI-driven wireless platform, designed specifically for the Smart Device Era. With the Mist Learning WLAN, you can think centrally and act locally through a combination of cloud-based intelligence and enterprise-grade Access Points. Revolutionary machine learning technology simplifies Wi-Fi operations and delivers personalized indoor location services.The Mist cloud, featuring a micro-services architectureSet, monitor and enforce service levelsMist lets you setup and track service level thresholds for key wireless criteria, such as time to connect, capacity, coverage, roaming and throughput. Drill down into users, applications, and devices and track trends over time.Simple root cause analysis and remediationMist dynamically collects over 100 states from every mobile device in near real-time and correlates this information for quick problem remediation across the wireless, wired and device domains. Predictive recommendations let you prevent issues entirely.Network rewind and dynamic packet captureWhen a user is experiencing a major network anomaly the Mist system automatically detects it and starts capturing packets in the cloud. In addition, Mist lets you rewind back in time to see what was going on exactly when the event occurred.API-firstThe entire Mist platform is 100% open and programmable using Mist APIs.Let’s get virtualPatented virtual Bluetooth® LE (vBLE) technology lets you deploy and move virtual beacons with the simple click of a mouse (or via APIs), eliminating the need for physical beacons.Machine learning eliminates manual calibrationMist continuously collects data in real-time from all mobile devices to ensure accurate location estimates without the need for Bluetooth® LE site surveys.Best location accuracy and performanceDynamic vBLE 16 antenna elements in Mist APs plus machine learning in the Mist cloud ensure 1 to 3m for mobile devices, and in-room /zone level for Bluetooth® LE assets.Set, monitor and enforce service levels; Rapidly isolate and fix problems with no truck rollsMist makes Wi-Fi predictable, reliable and measurable.Automate operations to save time and money and getunprecedented visibility into the Wi-Fi user experience.Mist brings Wi-Fi, Bluetooth LE, and IoT together to deliverpersonalized, location-based experiences.Mist Wi-Fi Assurance ServiceLocate key resources, like nurses and security guards; Track high value assets, like wheelchairs and forkliftsMist Asset Visibility Service22s26s 54491215DWELL TIMEVISITSVISITS BY HOURLocation-based proximity messaging and wayfindingMist Engagement Service Detailed analyticsMonitor user visits and dwell times, with detailed drill down into specific zone traffic patterns and congestion points.Real-time visibilityMist gives you a real-time view of the RF environment from the mobile user’s perspective. This lets you detect coverage holes and other issues, even as users move around and their RF coverage patterns change.Open APIs for easy integrationMist offers a mobile SDK for iOS and Android with three types of APIs (vBLE beacon, indoor location, and zone analytics) for seamless integration with complementary products.。
维亚云原生AI网络运维解决方案说明书
BrochureCloud-Native AIOpsEnd-to-End Service Assurance3AI for Network Operations (AIOps) Portfolio and CoverageNetwork Intelligence NOC/SOC Applications Interface ProtocolsDomains Data SourcesNetwork & Service AssuranceService Quality Management Cross Domain Correlation Root Cause Analysis Deduplication & Enrichment Noise Filtering Event & Alarm Analysis Incident ClassificationCross Domain Correlation Customer Experience Analytics Probe Correlation PM, FM, CM Correlation Physical & Virtual Inventory Resource Catalog Management Auto Discovery & Reconciliation Service / Resource Provisioning Configuration Change Management Configuration Backup & Restore Configuration Delta Analysis Golden Configuration Comparison Single Pane of Glass Performance Counters Alarms KQI Events Metrics Configuration Network Probes Revenue Data Ticket Information LogsTraces Inventory License Data Customer Experience E2E Service Topology MW Route Path Analysis Traffic Pattern Analysis Forecasting & Capacity PlanningAnomaly Detection Data Science Work Bench What-If SimulationEnd-to-End Automation Zero Touch Provisioning Parameter Comparison & Analysis Closed Loop Automation Single Holistic View Rogue IP Detection IPv4 Management IPv6 Management App, Flow, & Kubernetes Monitoring Automonous SOC Log Analytics & Forensics Infra Layer Observability Auto Discovery and End-to-End Network Service Topology Performance Assurance End-to-End Topology Automation & Orchestration IP Address Management Observability/Application MonitoringAI/ML Analytics Intelligent Correlation Inventory Management Configuration ManagementFault Management & AIOps Service Assurance and Analytics (Telco, Cloud & IT)Zero Touch Operations & Closed Loop Monitoring AI-driven Capacity Forecasting and Anomaly Detection Application Performance Monitoring &Observability Field Operations Management &Tower Mobile AppRadioMicrowave Open RAN IP Mobility Firewall DWDM Core RAN Core-CS, PS FTTH IN/VAS Fixed Line IP Enterprise Environment Physical Servers IoT Ericsson VNF Nokia VNF Huawei VNF Azure Cloud Google CloudAmazon Web Services Docker + KubernetesCisco VNF SNMP v2c, v3CORBA FTP SFTP Rest API Telemetry NETCONF RESTCONF Kafka Streaming Custom Interfaces ODBC, RPC, RMI 1341 900 1222AIOps Service AssuranceEnable network operation teamsto proactively identify and resolveCX-impacting network problemsand ensure quality service delivery AIOps Service Assurance• Identify critical metrics for bottlenecks that affect Quality of Service (QoS)• Proactively plan network capacity • Forecast Network Faults • Analyze and predict network trafficTeams• Network Operations • Field Operations • Customer Care Benefit• Proactive alerting system• Collaborative space for operators• Automated process workflowtrigger at the edgeValue• Improved operationalefficiency•Faster alarm triage TopologicalIntelligenceAlarm Clusters CorrelationEngineRoot NodeRCA132****1122Product FeaturesAIOps Service Assurance is available as a cloud service on public, hybrid, and private clouds.Network Performance MonitoringMonitor your network with 1000+ KPIs along with intuitive dashboards, instant alerts, and intelligent reportingEnd to End Network TopologyAutomated end-to-end, multi-layer, multi-vendor, multi-domain topology with end-to-end visualization and managementNetwork Configuration ManagerIntegrated with the Network Auto-Discovery Framework which automatically retrieves the current status of the network Workflow AutomationOrchestrate and automate network fault troubleshooting steps and ongoing maintenance tasks with ease using a rule-based workflow engine Anomaly DetectionDetect and rectify abnormal patterns in the network automatically through Al algorithmsApplication Performance ManagementMonitor application performance across physical, virtual, and cloud environments to ensure mission-critical business apps meet end-user expectationsGIS and Network Topology Visualizer Understand the bigger picture by visualizing how the overall network infrastructure is organizedA One Stop SolutionVIAVI AIOps Service Assurance visualizes, automates, and optimizes digital experiences as well as service and network quality across hybrid telco and IT networks by integrating real-time assurance with closed-loop automation and analytics driven by artificial intelligence/machine learning.Powered by topology, assurance and analytics suite of applications, Service Assurance governs the digital experience as well as service and network quality with intelligence that monitors, detects and heals by leveraging local orchestrators, driving digital transformation initiatives towards autonomic network management.4 Cloud-native AIOps: End-to-End Service Assurance© 2023 VIAVI Solutions, Inc. Product specifications and descriptions in this document are subject to change without notice. cloud-native-aiops-br-xpf-nse-ae 30193679 900 0223Contact Us +1 844 GO VIAVI (+1 844 468 4284)To reach the VIAVI office nearest you, visit /contact Proven Success Cases - Focus: CSPsEmpowering Fault ManagementThrough automated service qualitymanagement, customer experienceimpacting failures and the probableroot cause of network failures canbe identified and rectified, therebyreducing resources required tosupport the network leading tooverall OPEX reduction Improved Capacity Planning Planning the network capacity is very crucial, especially on special occasions. AIOps Service Assurance forecasting prediction algorithms give an idea to capacity planning teams as to the reduction or addition needed based on observed data Automated Network Orchestration Repeatable activities can be automated using the root cause analysis (RCA) workflow feature which, in turn, increases the efficiency of the network avoiding manual intervention and potentialassociated human errors。
海康威视DS-2DF6336V-AEL智能PTZ全景摄像头说明书
DS-2DF6336V-AEL3MP High Frame Rate Smart PTZ Camera Hikvision DS-2DF6336V-AEL smart PTZ Dome Cameras areable to capture high quality images of fast moving objectswith its high frame rate technology (up to 60fps@3MP).Embedded with 1/3’’ progressive scan CMOS chip makes true WDR (120dB) and 2MP real-time resolution possible. With the 36X optical zoom Day/Night lens, the camera offers more details over expansive areas.DS-2DF6336V-AEL smart PTZ Dome Cameras also features a wide range of smart functions, including face detection, intrusion detection, line crossing detection and audio exception, benefitting users with great improvement on security efficiency, more importantly, with key events / objects being recorded for further forensic needs. These features, combined with smart tracking, which enables the camera to detect any progressively moving object and follow it within the camera’s area of coverage without fault. Smart Defog and HLC/BLC are further supported to improve image quality in challenging conditions. Key Features•1/3” HD CMOS sensor•3MP(2048*1536) resolution •High frame rate, up to 60fps@3MP •36X Optical Zoom•120dB True WDR•Smart Tracking•Smart Detection•Defog•Hi-PoE / 24VAC power supplyFunction DescriptionBasic function:•High performance CMOS, up to 2048x1536 resolution •±0.1° Preset Accuracy•ONVIF(Open Network Video Interface Forum),CGI(Common Gateway Interface), PSIA(Physical Security Interoperability Alliance), to ensure greater interoperability between different platforms and compatibility•3D intelligent positioning function•High frame rate at 3MP•3D DNR•120dB WDR•IP67 standard•7 alarm inputs and 2 alarm outputs•Scheduled PTZ movementSmart function:•Smart tracking modes: Manual/ Panorama/ Intrusion trigger/ Line crossing trigger/ Region entrance trigger/ Region exiting trigger, support smart tracking when patrol between multiple scenarios•Smart detection: support Face detection, Intrusion detection (object or people stopped in ROI), Line crossing detection (Flow control and Virtual line), Audio exception detection, Region entrance detection, Region exiting detection•Smart recording: support edge recording, support dual-VCA for smart search in smart NVR•Smart image processing: support defog, HLC/BLC •Smart codec: low bit rate, ROICamera function:•Auto iris, auto focus, auto white balance, backlight compensation and auto day & night switch•Min. Illumination: 0.05Lux@(F1.6,AGC ON)(Color),0.01Lux@(F1.6,AGC ON)(B/W)•Support 24 privacy masks•Support configuration of video stream with adjust of: resolution, frame rate, bit rate type (CBR or VBR), Max bit rate;•Multi-language (including Portuguese) PTZ function:•360° endless pan range and -15°~ 90° tilt range •540°/s Pan Preset Speed and 400°/s Tilt Preset Speed •0.1°~ 300°/s Manual Pan Speed and 0.1°~ 240°/s Manual Tilt Speed•300 presets programmable; preset image freezing capability•8 patrols, up to 32 presets per patrol•4 patterns, with the recording time not less than 10 minutes per pattern•Proportional zoom function•Park action: auto call up of PTZ movement, after a defined time of inactivity•Power-off memory: restore PTZ & Lens status after reboot Network function:•H.264/MJPEG video compression• H.264 encoding with Baseline/Main/High profile•ROI(Region of Interest) encoding(support 6 areas with adjustable levels)•Built-in Web server•Onboard storage, up to 128GB•Support up to 8 NAS storage; Edge recording(transmit the videos from SD card to the NAS after network resumed) •HTTPS encryption and IEEE 802.1X port-based network access control•Support trip-streams•Multiple network protocols supported: IPv4/IPv6, HTTP, HTTPS, ARP, 802.1X, QoS, FTP, SMTP, UPnP, SNMP, DNS, DDNS, NTP, RTSP, RTP, TCP, RTCP, UDP, IGMP, ICMP, DHCP, PPPoE•1 audio input and 1 audio outputModel DS-2DF6336V-AELCamera ModuleImage Sensor 1/3’’ Progressive Scan CMOSMin. Illumination F1.6, 50IRE, AGC On: Color: 0.05 lux, B/W : 0.01 luxMax. Image Resolution 2048×1536Focal Length 4.5-162mm, 36xDigital Zoom16XZoom Speed Approx.4.5s(Optical Wide-Tele)Angle of View60.6-3.68 degree (Wide-Tele)Min. Working Distance10-1500mm(Wide-Tele)Aperture Range F1.6~F4.4Focus Mode Auto / Semiautomatic / ManualWDR120dBS / N Ratio≥ 55dBShutter Time50Hz: 1~1/30,000s; 60Hz: 1~1/30,000sAGC Auto / ManualWhite Balance Auto / Manual /ATW/Indoor/Outdoor/Daylight lamp/Sodium lamp Day & Night IR Cut FilterPrivacy Mask24 privacy masks programmableEnhancement3D DNR, Defog, HLC/BLC, SVCPan and TiltRange Pan:360° endless; Tilt: -15°~90°(Auto Flip)Speed Pan Manual Speed: 0.1°~300°/s, Pan Preset Speed: 540°/s Tilt Manual Speed: 0.1°~240°/s, Tilt Preset Speed: 400°/sNumber of Preset300Patrol8 patrols, up to 32 presets per patrolPattern10 patterns, with the recording time not less than 10 minutes per patternPark Action Preset / Patrol / Pattern / Auto scan / Tilt scan / Random scan / Frame scan / Panorama scanScheduled Task Preset / Patrol / Pattern / Auto scan / Tilt scan / Random scan / Frame scan / Panorama scan/Dome reboot/Dome adjust/Aux output Smart FeaturesSmart tracking Manual/ Panorama/ Intrusion trigger/ Line crossing trigger/ Region entrance trigger/ Region exiting trigger Smart tracking when patrol between multiple scenariosSmart detection Face detection, Intrusion detection, Line crossing detection, Audio exception detection, Region entrance detection, Region exiting detectionROI encoding Support 6 areas with adjustable levels AlarmAlarm I/O7/2Alarm Trigger Face detection, Intrusion detection, Line crossing detection, Region entrance, Region exiting, Audio exception detection, Motion detection, Dynamic analysis, Tampering alarm, Network disconnect, IP address conflict, Storage exceptionAlarm Action Preset, Patrol, Pattern, Micro SD/SDHC card recording, Relay output, Notification on Client, Send Email, Upload to FTP, Trigger ChannelInput/OutputMonitor Output 1.0V[p-p] / 75Ω, NTSC (or PAL) composite, BNCAudio Input 1 Mic in/Line in interface, line input: 2-2.4V[p-p]; output impedance: 1KΩ, ±10% Audio Output 1 Audio output interface, line level, impedance: 600ΩNetworkEthernet10Base-T / 100Base-TX, RJ45 connectorMain Stream 50Hz: 25fps(2048×1536), 25fps(1920×1080) / 50fps(2048×1536), 50fps(1920×1080) 60Hz: 30fps(2048×1536), 30fps(1920×1080) / 60fps(2048×1536), 60fps(1920×1080)Sub Stream50Hz: 25fps(704×576, 352×288, 176×144); 60Hz: 30fps(704×480, 352×240, 176×120)Third Stream 50Hz: 25fps(2048×1536, 1920×1080, 1280×960, 1280×720, 704×576, 352×288, 176×144) 60Hz: 30fps(2048×1536, 1920×1080, 1280×960, 1280×720, 704×480, 352×240, 176×120)Image Compression H.264/MJPEG, H.264 encoding with Baseline/Main/High profileAccessoriesDS-1602ZJ Wall Mount DS-1602ZJ-Conner Conner Mount DS-1602ZJ-Pole Pole Mount DS-1602ZJ-BoxBox MountDimensions Audio Compression G.711ulaw/G.711alaw/G.726/MP2L2/G.722ProtocolsIPv4/IPv6, HTTP, HTTPS, ARP, 802.1X, QoS, FTP, SMTP, UPnP, SNMP, DNS, DDNS, NTP, RTSP, RTP, TCP, RTCP, UDP, IGMP, ICMP, DHCP, PPPoESimultaneous Live View Up to 20 usersMini SD Memory Card Support up to 128GB Micro SD/SDHC/SDXC card. Support Edge recording User/Host Level Up to 32 users,3 Levels: Administrator, Operator and UserSecurity MeasuresUser authentication (ID and PW); Host authentication (MAC address); IP address filtering System IntegrationApplication programming Open-ended API, support ONVIF, PSIA , CGI and GenetecWeb Browser IE 7+, Chrome 18 +, Firefox 5.0 +, Safari 5.02 +, support multi-language RS-485 Protocols HIKVISION, Pelco-P, Pelco-D, self-adaptive PowerHi-PoE&24 VAC, Max.60W Working Temperature -40°C ~ 65°C (-40°F ~ 149°F) Humidity 90% or lessProtection Level IP67, IK10, TVS 4,000V lightning protection, surge protection and voltage transient protection Certification FCC, CE, UL, RoHS, IEC/EN 61000, IEC/EN 55022, IEC/EN 55024, IEC /EN60950-1 Dimensions Φ178.3(mm)×326.0(mm ) (Φ7.02”×12.8”) Weight (approx.) 8kg (approx.)Mount OptionLong-arm wall mount: DS-1602ZJ; Corner mount: DS-1602ZJ-corner;Pole Mount: DS-1602ZJ-pole; Power box mount:DS-1602ZJ-box; Swan-neck mount: DS-1619ZJDS-1619ZJ Swan-neck MountDS-1662ZJ Pendant mount DS-1661ZJ Pendant mount DS-1663ZJ Celling mount DS-1005KI USB Joy-stickDS-1100KI Network Keyboard。
Nozomi Networks Central Management Console数据表说明书
DATA SHEETCentral Management ConsoleCentralized OT and IoT Security and Visibility for Distributed SitesNozomi Networks Central Management Console™ (CMC) delivers centralized OT and IoT security management – no matter how large or distributed your business is.Whether you’re consolidating visibility and risk management at the edge or in the cloud, the CMC is fast and simple to deploy.SeeAll OT and IoT devices and behavior onyour networks for unmatched awareness DetectCyber threats, vulnerabilities, risks and anomalies for faster response UnifySecurity, visibility and monitoring across all your assets for improved resiliencyDelivers fast insight into key metrics, alerts, incidents, vulnerabilities and more Manages Nozomi Networks appliances and servicesUnified Security Monitoring Integrates quickly with asset, ticketing &identity management systems, and SIEMs Streamlines security processes across IT/OT and harmonizes security data /integrationsEnterprise-level Best PracticesLeverages enterprise single sign-on (SSO) credentials for fast access to OT and IoT informationProvides deep role-based access control (RBAC) “least privilege” permission options for maximum securityEasily Streamline SOC/IT Workflows The CMC showing a geographic map of your Guardian appliances.Answers questions fast with powerful queries about any and all OT/IoT data Deploys in the cloud (AWS or Azure), or at the edge, for anytime, anywhere accessCaptures key data such as traffic throughput, TCP connections, protocols used between zones and more Accelerates incident response and troubleshooting effortsSummarized Asset InformationAggregates asset inventories that are automatically generated by Guardian at each siteProvides key information such as:• OT asset inventory • IoT asset inventory• Vulnerabilities by asset type, vendor or locationon nodes and connectionsFilters by subnet, type, role, zone and topologyGroups assets visually, in lists and detailed single asset viewsDrilldown to Individual AssetsEnables access to local Guardian appliances and individual asset details, such as:Quickly KnowYour Assets and Their Risks• Operating system • MAC vendor • MAC address • Installed software • Vulnerabilities • Captured URLs/files• IP • Subnet • Zone • Role • AlertsPortion of interactive Network Visualization Graph .reliability threatsReports attacks in process, reducing the mean-time-to-detection (MTTD)Consolidates vulnerability assessment across sitesPowerful Tools for Fast AnalysisDecodes incidents with Time Machine ™ before and after system snapshots Provides fast answers with a powerful ad hoc query tooldetection with signature-based threat detection for complete coverageIntegrates quickly with ticketing systems and SIEMs for streamlined security processesSmart Incidents Speed ForensicssDecreases response time with Smart Incidents ™ that:• Correlate and consolidate alerts• Provide operational and security context • Supply automatic packet capturesOptimizeTroubleshooting and Forensic EffortsSmart Incident showing related alerts and security context.Centralized Monitoring of OT RisksConsolidates information for thousands of sites and devicesQuickly scales for enterprise-wide deployment with optimum performance Adapts to all sites, with multiple appliance models and flexible deployment optionsSwift DeploymentInstalls as a proven, plug-and-play, ISO 9001: 2015 certified productDeploys in the cloud on AWS or Azure, and at the edge on virtual and physical appliancesRolls out to numerous sites within weeksHigh Availability, High SecurityImmediately ValuableImproves visibility, cybersecurity and reliabilityIntegrates with existing tools and workflows for fast adoption and high productivityAccelerates IT/OT collaborationRealize Rapid Time to ValueAttain HighPerformance for Multinational DeploymentsSample deployment map for centrally monitoring and securing many facilities.Ensures continuous OT and IoT monitoring with high availability and multitenant CMC configurations Connects with field appliances using encrypted, bandwidth-optimized data transfersOT and IoTSecurity and VisibilityYou can deploy the Nozomi Networks solution in a wide variety of mixed environments for rapid asset discovery, network visualization and accelerated security.Nozomi Networks offers a range of licensing options to fit your organization's preferences, including:• Subscription licenses• Enterprise license agreements (ELAs)• Perpetual licensesPlease contact your Nozomi Networks Sales Director for more information.Multiple Licensing OptionsIndustrialStrength Cyber and Operational ResiliencyThreat Intelligence GuardianIntelligenceIntegrated SystemsOT Virtual ContainerPhysicalCentral ManagementConsoleSample Deployment ArchitectureNozomi Networks partners deeply with the IT/OT services and technology companies you trust. These include:• Strategic alliances with enterprise IT and managed security providers • Technology integrations with leading IT/OT solutions • Global Network of SI, VAR and Distribution Partners Visit /partners for more information.World Class PartnersYou can tailor the Nozomi Networks solution to meet your needs by utilizing its flexible architecture, extensive range of appliances, and integrations with other systems.Purdue Model ExampleSITE #2SITE #NLevel 0Field NetworkLevel 1Control NetworkLevel 2Process NetworkLevel 3Operations (ICT/DMZ)HMIENGINEERING WORKSTATIONDNSHISTORIANLevel 4Enterprise ITPLCs/RTUsCentral Management ConsoleHIGH AVAILABILITY CONFIGURATIONThreat IntelligenceSUBSCRIPTIONGuardian+Smart PollingO P T I O N A LAsset IntelligenceSUBSCRIPTIONSITE #1REMOTE ACCESSINTEGRATED SYSTEMS SOC / CONTROL CENTERBUSINESSCOMPUTERCAMERA TABLETCMC Appliancesin the Cloud or at the EdgeVirtual appliance for centralized management of multi-tier, distributed deployments.Cloud or VirtualUnlimited Max. Managed Appliances** Based on the infrastructureRack-mounted appliance for centralized management of multi-tier, distributed deployments.NCMC-100 Series50 Max. Managed AppliancesFor complete and current tech specs, visit: /techspecs , or contact us.Here are several options for deployment and support assistance:• Nozomi Networks Global Strategic Alliance Partners • Nozomi Networks SI/VARs• Nozomi Networks Professional Services • Nozomi Networks Global Customer SupportMultipleDeployment and Support OptionsSUBSCRIPTIONThreat IntelligenceThe Threat Intelligence service delivers ongoing OT and IoT threat and vulnerability intelligence. It helps you stay on top of the dynamic threat landscape and reduce your mean-time-to-detection (MTTD).GUARDIAN ADD-ONRemote CollectorsRemote Collectors are low-resource appliances that capture data from your distributed locations and send it to Guardian for analysis. They improve visibility while reducing deployment costs.(MTTR).GUARDIAN ADD-ONSmart PollingSmart Polling adds low-volume active polling to Guardian’s passive asset discovery, enhancing your asset tracking, vulnerability assessment and security monitoring.© 2020 Nozomi Networks, Inc. All Rights Reserved.DS-CMC-8.5x11-012Nozomi NetworksThe Leading Solution for OT and IoT Security and VisibilityNozomi Networks is the leader in OT and IoT security and visibility. We accelerate digital transformation by unifying cybersecurity visibility for the largest critical infrastructure, energy, manufacturing, mining, transportation, building automation and other OT sites around the world. Our innovation and research make it possible to tackle escalating cyber risks through exceptional network visibility, threat detection and operational insight.。
基于最优Gabor滤波器的无纺布缺陷检测
基于最优Gabor滤波器的无纺布缺陷检测
撒雨昕;蔡硕
【期刊名称】《信息通信》
【年(卷),期】2018(000)010
【摘要】缺陷检测是无纺布生产过程中至关重要的一个环节,针对无纺布的常见缺陷,设计了一种基于最优Gabor滤波器的无纺布缺陷检测算法.首先针对5尺度8方向上的40个Gabor滤波器,对每个滤波子图像的熵进行计算,确定了一个最佳的滤波方向;其次,根据损失评价函数值从最佳方向上的5个滤波器中确定了一个最佳的滤波尺度;最后,对滤波结果进行阈值分割及伪缺陷剔除,实现了缺陷检测.从实验结果来看,所设计的算法可以有效地检测出无纺布缺陷.
【总页数】3页(P18-20)
【作者】撒雨昕;蔡硕
【作者单位】武汉外国语学校,湖北武汉430022;武汉理工大学信息工程学院,湖北武汉 430070
【正文语种】中文
【中图分类】TP311.1
【相关文献】
1.基于最优Gabor滤波器的织物缺陷检测 [J], 袁端磊;宋寅卯
2.基于匹配GabOr滤波器的规则纹理缺陷检测方法 [J], 贡玉南;华建兴;黄秀宝
3.基于优化Gabor滤波器的牙膏管肩缺陷检测 [J], 高梦阳;王晗;曾耀斌;张芳健;蔡
念;董永超;陈新度
4.基于灰度均衡模型联合Gabor滤波器的钢轨表面缺陷检测方法 [J], 刘琴琴;周慧云;王兴洲
5.基于优化Gabor滤波器的铸坏表面缺陷检测应用研究 [J], 徐建亮;毛建辉;方晓汾
因版权原因，仅展示原文概要，查看原文内容请购买。
- 1、下载文档前请自行甄别文档内容的完整性，平台不提供额外的编辑、内容补充、找答案等附加服务。
- 2、"仅部分预览"的文档,不可在线预览部分如存在完整性等问题,可反馈申请退款(可完整预览的文档不适用该条件!)。
- 3、如文档侵犯您的权益，请联系客服反馈,我们会尽快为您处理(人工客服工作时间：9:00-18:30)。
978-1-4673-5939-9/13/$31.00 ©2013 IEEE
1488
nodes. Figure 1 shows the Voronoi Diagram of a set of nodes and the Voronoi cell of node s (expressed in dashed lines). Considering principles of Voronoi Diagram [6], Voronoi cell of each node is the convex area nearest to itself. Any point in the area is closer to the node than any other nodes. Hence, in certain Voronoi cell, if an area cannot be covered by node of the cell, it cannot be covered by other nodes. The detecting algorithms in [3, 4] are based on the principle.
2013 IEEE Wireless Communications and Networking Conference (WCNC): NETWORKS
Virtual Edge Based Coverage Hole Detection Algorithm in Wireless Sensor Networks
the detection of coverage holes is based on the Voronoi Diagram [3, 4], which is a fast and easy way to detect holes. But it can only express the holes’ locations with vertexes of Voronoi polygons, rather than describe locations and shapes of holes exactly. Yao [5] developed a hole detection algorithm based on the Simplicial Complex method. By establishing a maximum simpicial complex sub-network, it can describe holes with polygons formed by boundary edges and get more accurate result than Voronoi Diagram algorithm. However, as shapes of coverage holes are mostly not regular, there exists error by representing holes with polygons. In this paper, we assume that each node has the same sensing radius in the Boolean sensing model [7]. The nodes are aware of their geographical locations. Since methods on localizing sensor nodes are mature in both indoor and outdoor conditions [8, 9, 10], we assume that at least one of them is feasible in our research. Our algorithm screens out boundary nodes by Voronoi Diagram. Then it further calculates boundary nodes with Virtual-Coverage-Edge method to achieve exact locations of coverage holes. II. COVERAGE HOLE DETECTION ALGORITHM If a node is adjacent to any coverage hole, we call the node as hole-boundary node or boundary node. Our algorithm has two phases: the first phase is to get hole-boundary-nodes by Voronoi Diagram; the second phase is to gain locations of holes by calculation of boundary nodes with the method of Virtual-Coverage-Edge. A. Nodes Screening Strategy First, our algorithm decides whether a node is a boundary node and how to screen out boundary nodes among large numbers of randomly deployed nodes. 1) Voronoi Diagram Voronoi Diagram is an important structure of computational geometry. With the knowledge of coordinates of a set of points in a limited plane space, the space can be divided into several cells by Voronoi Diagram. Furthermore, each cell contains only one point. Therefore, when applied to wireless sensor networks, Voronoi cells can represent neighboring information of sensor
Yunzhou Zhang1,2, Xiaohua Zhang1, Zeyu Wang1, Honglei Liu1 1. College of Information Science and Engineering 2. Shenyang SIASUNG Robot & Automation Northeastern University Company, Ltd. Shenyang, P.R.China zhangyunzhou@
Hale Waihona Puke I. INTRODUCTION In practical applications such as environmental monitoring, forest fire detection, security and military applications, wireless sensor networks (WSNs) are always required to cover the area of interest with certain degree. For WSNs deployed deterministically, coverage requirement can be met by appropriate deployment strategy. However, most WSNs have to be deployed randomly with the help of aircrafts, like helicopters, which can hardly guarantee coverage quality of networks. Meanwhile, for WSNs that have already deployed, nodes deployed over certain area may be destroyed owing to environmental factors as winds, obstacles, vibration, exogenic actions or failure of electronic components. Furthermore, the energy of each node is limited, thus the failure of nodes can also be caused by power depletion, impacting coverage quality of network. Therefore, coverage holes problem of WSN is unavoidable. The coverage holes could have a negative effect on the performance of WSN in many aspects [1]. Recently, coverage holes of WSN have emerged as one of the hottest study fields of WSN. There are many researches on coverage-hole detection, hole boundary search and repair of coverage holes. Lots of algorithms on holes detection have been proposed. Zhao et al. [2] introduced a computational geometry method to detect coverage holes. The method discusses whether any hole exists by calculating the angles among each node and its neighbor nodes. The method is useful for getting the number of holes in certain area, but it does not provide the exact location of holes. Another classic method on