We are living in an age of increasing and evolving threats – from explosives and biological pathogens to cyber-attacks. Safeguarding public spaces is therefore top of policy-makers’ priorities. From corporate and government buildings, to borders and airports, effective security solutions are required to protect property, staff, and the public. As countries around the world continue the fight against COVID-19 and mitigate the economic impact of the pandemic, operational and cost efficiency has become more important than ever.
For every industry where security measures are required, there are a specific set of emerging threats and pressures, and in many cases, this is coupled with specific regulatory requirements. Take the air cargo sector for example, explosives, weapons, and drugs are very present threats, and new regulation around dangerous goods and lithium battery handling is being introduced from 2022. Against this backdrop, solutions which effectively tackle threats and support compliance without compromising on efficiency are vital for a sector that trades on speed – and this is the case for many industries.
The good news is that technologies are available which not only advance screening procedures and detection capabilities to monitor for and identify complex threats, but reduce the operator burden, create higher throughput, and maintain system uptime to help keep both people and property safer, without the need for large numbers of highly skilled and very costly security personnel. So, what are the new and emerging technologies which have the potential to unlock this new, more efficient era of security?
AI: Removing the guesswork by doing the legwork
Artificial intelligence (AI) powered software is at the heart of many solutions which automate the detection of threats, increasing the accuracy of detection and reducing operator interference to create more seamless security processes. Advanced detection capabilities enabled by AI can enhance existing security systems in any setting – whether that be an airport, at a border, or in a stadium.
Machine learning algorithms can imitate the way the human brain processes data and identify patterns for use in decision-making. The vast amounts of data collected on prohibited items through security processes can be used to train and refine algorithms to achieve a highly accurate rate of automatic detection of hazards or illicit goods through object recognition. This can help to combat the movement of an ever-expanding list of unsafe, undeclared, or illegal items such as weapons, drugs, or even currency, without interrupting the flow of people, bags, or cargo. By delivering the highest level of threat detection it can also support more efficient resource planning for customs officers, security operators or other controlling authorities.
Algorithms are readily available for use at security checkpoints and could enable alarm-only viewing of X-ray images to significantly improve throughput and security levels, ideal for busy passenger and cargo airports and large, public events. As an emerging area, the urgent need for this type of contactless and efficiency boosting technology could mean that we see the capabilities of AI powered algorithmic software expand at a fast pace, and the relevant approval required for safe implementation for regulated industries.
算法很容易在安全检查站使用，并且可以只需报警就可以查看 X 射线图像（海曼安检机），从而显著提高吞吐量和安全水平，非常适合繁忙的客运和货运机场以及大型公共活动。作为一个新兴领域，迫切需要这种非接触和提高效率的技术，这可能意味着我们将看到人工智能驱动的算法软件的能力迅速扩大，而受监管行业的安全实施需要获得相关批准。
The value of AI can be clearly seen in complex security operations where volumes are increasing but resource availability is decreasing. For example, customs and security professionals are facing growing traffic at maritime ports, land border crossing points and city entrances. At the same time, they are required to provide faster clearance times, while keeping strict controls in place to monitor the flow of imports, exports and transit traffic. Image analysts must match X-ray images of scanned cargo with original manifest reports while also looking for potential threats. AI-powered algorithms can automatically highlight only those X-ray images where suspicious items have been detected, such as cigarettes or even dangerous levels of radioactivity, speeding up the overall analysis process and supporting the secure movement of goods and free flow of trade.
人工智能的价值可以在复杂的安全行动中清楚地看到，这些行动的数量在增加，但资源的可用性却在减少。例如，海关和安全专业人员在海港、陆地边境过境点和城市入口处面临着越来越多的交通流量。与此同时，它们必须提供更快的结关时间，同时保持严格的管制，以监测进口、出口和过境运输的流动。图像分析师必须将扫描货物的 X 射线图像与原始清单报告进行匹配，同时还要寻找潜在的威胁。人工智能算法只能自动突出那些检测到可疑物品的 X 射线图像，如香烟，甚至危险的放射性水平，加快整个分析过程，支持货物安全流动和贸易自由流动。
Widening the network to close the efficiency gap
For hubs, such as major airports and busy ports, where there are multiple terminals and consistently high volumes of passengers or cargo, wide-area networks (WANs), which enable centralized and remote image evaluation, can be the best way forward. Although centralization is not a new concept, image analysis is a new application. WANs can facilitate the real-time sharing of images between different areas of a building or sites (or even countries and continents) enabling greater resource prioritisation and operational efficiency.
Although long established for airport hold baggage screening systems, remote screening is new for airport passenger checkpoints, cargo and border control. The benefits are particularly clear when it comes to countries with many regional airports spread far apart which see fluctuating passenger volumes. Linking all outlying locations to a key airport where volumes are more consistent enables more efficient operator resourcing, rather than keeping staff onsite at smaller airports around the clock. Indeed, larger airports could offer outsourced security services to smaller airports. At borders, where an enormous amount of goods pass through entry and exit points 24 hours a day, a centralised Dataset Management System can allow X-ray images and associated data to be analysed online in a remote-control centre. On-site operators can therefore focus on the scanning process and completing the relevant dataset information such as customs declarations and vehicle licence plates.
虽然长期建立的机场托运行李检查系统（进口安检机-海曼安检机 smiths detection），远程检查是新的机场旅客检查站，货物和边境管制。对于许多地区性机场相距甚远、客流量波动较大的国家而言，这种做法的好处尤为明显。将所有偏远地点与一个交通量更加一致的关键机场连接起来，可以更有效地为运营商提供资源，而不是让工作人员24小时待在较小的机场。事实上，较大的机场可以向较小的机场提供外包的安全服务。在边境，每天24小时都有大量货物通过出入境口岸，一个集中的数据集管理系统可以让 X 射线图像和相关数据在遥控中心进行在线分析。因此，现场操作员可以专注于扫描过程和填写相关的数据集信息，例如海关申报单和车辆号牌。
On a country-to-country or even continental level, image sharing via WANs would enable more sophisticated data analysis across global security networks to significantly boost security outcomes, with one set of scanned images for both outbound security and inbound customs clearance at the destination. Of course, security outcomes can never be compromised, so wide networks must be robust and secure, with sufficient bandwidth for real-time distribution of the images. This creates significant technical challenges in establishing a viable WAN, and back-up solutions are required in case of network failures. Although the capability is there, to fully realise the potential of international data sharing close co-operation between authorities is required. Looking to the future, centralised screening could consolidate not only images from checkpoint screening but trace equipment, body scanners and CCTV.
Beyond image screening, there are significant operational advantages to networking, which can be used as an intelligent source of management data and statistics to inform decision making. For example, information gathered from across a WAN can support preventive maintenance, resource allocation and general administration, delivering cost-efficiencies as well as a high level of system uptime. While centralised management via WANs is already in operation with some global security screening installations, none are yet handling real-time image analysis in a co-ordinated or consistent way.
Differentiated screening: One size does not fit all
The advantages of adopting a remote WAN screening model are accentuated when paired with a differentiated screening approach based on risk and sensitivity levels of either cargo or people. For cargo, data such as the shipper, origin, destination as well as weight, size and density of cargo allows for the creation of bespoke risk scores, so that security operator and screening resource can be focused on shipments that pose a higher risk, whilst parcels from trusted shippers are fast-tracked. Risk scores can typically be generated through the shipping manifest, which acts as a unique identifier, with the risk assessment criteria based on the key data. For example, shipments from Africa may pose a higher risk of illegal animal trade due to the issue of rhino horn smuggling. The concept of ‘one stop security’ can be enabled using WAN image sharing combined with differentiated screening. Screening and risk-based data could be shared between departure points with destination authorities, allowing for the rescreening requirement at the transfer airport to be determined by risk profile.
For example, The El Salvadorian Dirección General de Aduanas (Customs Department) is responsible for controlling the export, import and transit of goods at all the country’s border crossing and entry points, and therefore safeguarding the country against smuggling, illegal trafficking of narcotics, weapons, contraband and other illegal goads and substances. In conjunction with local partner Cotecna, Smiths Detection worked closely with Dirección General de Aduanas on a multi-border inspection project which involved a differentiated screening approach. Consignments travelling through the county are now assessed for risk to determine scanning requirements, allowing for the appropriate level of screening to be applied, ranging from re-checks on different X-ray systems to checks with hand-held equipment to detect traces of explosives on vehicles and palletized cargo.
例如，萨尔瓦多海关总署负责在该国所有过境点和入境点控制货物的出口、进口和过境，从而保护该国免受走私、非法贩运毒品、武器(手持金属探测器)、违禁品和其他非法货物和物质的危害。史密斯侦查公司与当地伙伴 Cotecna 公司密切合作，与阿杜瓦纳斯总局合作开展了一个多边检查项目，其中包括采用差别筛查办法。现在对通过该州的货物进行风险评估，以确定扫描要求，从而能够进行适当程度的筛查，从重新检查不同的 X 光系统到用手提设备进行检查，以发现车辆和托盘货物上的爆炸物痕迹。
When it comes to people and handheld baggage screening in airports and buildings, AI and biometrics can be used to gather, combine and analyse comprehensive profiles to allow for more efficient and targeted screening. The key enabler of this type of automated, differentiation assessment could be a biometrics-enabled checkpoint. Differentiated screening adapts the security process through individualised risk assessments based on a unique identifier, created using biometrics, combined with contextual information – such as ticketing information or ID details. Once a person’s name or ticket information is amalgamated with data from third party sources, a risk score can be generated. Applying differentiated levels of screening focuses operator resources on those with higher risk scores, reducing the pressure on screening operators as well as costs, while enabling a more seamless flow of people through the screening process.
Cyber-proof: Building resilient networks
With increasing reliance on connected, digital security solutions comes an increasing risk of cybersecurity attacks, with incidents ranging from temporary disruption to global networks going down. In 2019 IATA published a whitepaper, Air Transport Security 2040 and Beyond, which warns that by 2040 more airport processes will be conducted offsite, requiring a range of networks and thereby increasing the risk of vulnerabilities that attackers could exploit – which is true of any networked system. Security systems which use AI-enabled algorithms often require access to the cloud and are shared through open architecture utilising third party open interfaces. To ensure these connections are robust, and to protect the highly sensitive data that security systems collect, a holistic approach to developing cybersecurity policies is required which spans software, hardware, process, and operators. The first step of this approach is assessing operational risks and compliance requirements, with the goal of forming a bespoke policy which is flexible enough to adapt to new and evolving threats without compromising robustness.
随着对互联网的依赖程度越来越高，数字安全解决方案面临着越来越大的网络安全攻击风险，从临时中断到全球网络瘫痪等各种事件不断发生。2019年，国际航空运输协会(IATA)发表了一份名为《空运安全2040年及其后》(Air Transport Security 2040 and Beyond)的白皮书，警告说，到2040年，将有更多的机场程序在外部进行，需要一系列网络，从而增加了攻击者可能利用的漏洞的风险——任何网络系统都是如此。使用人工智能算法的安全系统通常需要访问云，并通过使用第三方开放接口的开放架构共享。为了确保这些连接是健壮的，并保护高度敏感的数据，安全系统收集，需要一个全面的方法来发展网络安全策略，其中包括软件，硬件，过程和操作员。这种方法的第一步是评估业务风险和遵守要求，目标是形成一种定制的政策，这种政策足够灵活，能够在不损害稳健性的情况下适应新的和不断变化的威胁。
A contactless future
The implementation of digital, data-driven and differentiated solutions can enable a more contactless and efficient future for security. These technologies are becoming commercially available, and we are seeing an acceleration in development and in situ trials. Not only do these technologies improve the security operator’s experience, but the end user’s, from airport passengers to sports fans in stadiums. While the technical capabilities are there, the responsible implementation of these solutions brings a set of challenges which can only be tackled with a collaborative and integrated approach between suppliers, operators and authorities. In some cases, this will require significant co-operation transcending borders.