{"data":{"news":{"items":[{"title":"How To Use AI To Fight Identity Fraud","date":"2019-03-29 20:49:38","id":1551,"author":"Glenn Larson, VP of Engineering, Acuant ","blurb":"

Can AI fight identity fraud?

\n","blurb_image_uid":"https://dxfpracgmlczp.cloudfront.net/2019/03/1553864006345_blurb.jpg","blurb_expanded_image_uid":null,"article":"

It’s no secret that identity fraud is a growing problem: A record 16.7 million US adults experienced identity fraud in 2017, marking an 8% increase from the year before, according to Javelin's 2018 Identity Fraud study.

\n\n

The number of fraudulent transactions, massive data breaches and instances of identity theft continues to rise as hackers and fraudsters become more sophisticated.

\n\n

AI could be the solution to identify fraud.

\n\n

ID scanning solutions have various strengths; some simply scan an ID’s barcode, whereas more robust software performs forensic and biometric tests to ensure that an ID is not forged. All of these ID scanning solutions can be strengthened considerably by applying artificial intelligence.

\n\n

Artificial intelligence and its subsets of machine learning and deep learning make it possible to accurately process, verify, and authenticate identities at scale. Here's how.

\n\n

Scaling ID Authentication With Machine Learning

\n\n

Identity documents, such as driver’s licenses and passports, are scanned to test various elements of an ID, either on the premises or remotely using mobile devices. Some examples of authentication tests include confirmation of genuine microprint text and security threads, validation of special paper and ink, comparison between OCR and barcodes and magnetic strips, data validity tests, and biometrics or facial recognition to link the individual to the ID credential.

\n\n

Using machine learning — or automation — creates a more efficient and accurate process compared to relying on an untrained human to look at the document. These solutions should contain an anonymous internal data-collection mechanism capable of storing information about the operation and performance of the software. Then, the data is automatically transmitted to the provider on a regular basis. This process — when automated — saves time and improves the quality of the result.

\n\n

Why You’re Only As Good As Your Data

\n\n

There are thousands of forms of ID — passports, DHS "Trusted Traveler" cards (Global Entry, NEXUS, SENTRI, FAST), military IDs, permanent resident cards, border-crossing cards — and obtaining significant quantities of each type to understand the variations, wear patterns, etc. is a monumental task.

\n\n

Effective ID authentication relies on the collection of metadata on the document recognition and authentication process. Instead of containing information about the document being processed, this metadata contains information about the processes that are run and the details and outcome of those processes. By analysing this information, the software trains itself to detect complex patterns and output a prediction/result. It optimises the performance of the software and library to improve the reliability of the document read and authentication processes.

\n\n

Using machine learning to differentiate between good and bad IDs is extremely efficient. However, without supervision, the logic developed by the algorithm may exclude IDs that are not fraudulent. There are many reasons why an ID may not pass even though it is valid: wear and tear, other physical damage, manufacturing errors or defects, minor design changes or even variations in production depending on where and how the card was produced (central versus instant issue).

\n\n

Most states now use central issuance but have the ability to instantly issue an ID in some situations, for VIPs, for example. Given that IDs are often not printed in the same location by the same machine, there are often anomalies based on printing quality, misprints or unclear images. It is not uncommon for an entire batch of IDs to contain a manufacturing error.

\n\n

In addition, a computer must be taught that IDs that are worn or damaged are still valid. Accessing a programme that allows the collection of operational and performance metrics for the software is useful for improving the recognition and authentication of documents supported by the document library of the provider.

\n\n

A robust document library against which to compare captured IDs is vital; access to a comprehensive and regularly updated library cuts down the time that machines must process data on their own and maximises data extraction and authentication capabilities. Semi-supervised machine learning enables adjustment of the direction of the logic without interfering with the insights that authenticate documents or slowing down data processing.

\n\n

Data Mining, Semi-Supervised Learning And Regression Analysis In The Feedback Loop

\n\n

There are multiple models of ID authentication to choose from:

\n\n

Data mining: Examining large databases to turn raw data into useful information. For more efficiency, be sure to extract clean data to save time with this process.
Semi-supervised learning: Relying on completely automated machine learning can result in “failing” documents for items that have experienced wear and tear or ones with document manufacturing errors.
Regression analysis: This method continually tests and analyses the outcomes to improve the algorithm.

\n\n

New data is fed into the algorithm to test outcomes. This process is often called the feedback loop. As new data is fed into the algorithm, the feedback loop tests that the outcomes are consistent and improving. The outcomes are then fed into the algorithm so that the software continues to learn and adjust.

\n\n

When To Use Biometrics For Identity Management

\n\n

Biometric identity verification methods implement biometrics, such as facial or voice recognition, to strengthen the identity verification process. Not only is it more passive for consumers (since it reduces the need to remember a password or enter PII information), but it’s also a more stringent security protocol.

\n\n

Biometric security solutions utilise deep learning to mimic the way human neurons process extremely difficult information, such as faces and language. With deep-learning technology, the software provider can model large amounts of complex data, such as many images and faces.

\n\n

Facial recognition technology utilises deep learning to learn to match the image on the ID to a person’s face. Then the algorithm looks for certain patterns such as basic shapes (eyes, mouth, nose) and complex shapes (complete faces and distinctive shapes), and finally returns an output that indicates whether the image matches the ID's face or not.

\n\n

Machines + Humans = The Highest Accuracy

\n\n

Trained professionals using the software can step in to prevent bad customer experiences when the software flags a legitimate ID because it is damaged or worn. During the rare instances when the computer fails to identify what is wrong with an ID, the professional can apply an expert eye, determine what error occurred, and teach the computer how to spot the issue in the future.

\n\n

The key is to quickly understand what data points are relevant to identifying a document as fraudulent versus results that are noise due to expected variations (wear, damage, manufacturing etc.). A human eye can do this much more effectively than a computer. This creates a different method of learning, where new information is being input to the learning model, so the model can improve.

\n","article_image_uid":"https://dxfpracgmlczp.cloudfront.net/2019/03/1553864011257_main.jpg","article_expanded_image_uid":null,"article_image_caption":"","publications":"5","tags":"AI and identity fraud,Combating identity fraud,AI and fraud,ID authentication using AI","category":"Products & Technologies"},{"title":"How AI Will Change Video","date":"2019-03-29 20:44:04","id":1550,"author":"Benjamin Low, Vice President, Asia Pacific, Milestone Systems","blurb":"

AI is changing video and the video industry will never be the same again

\n","blurb_image_uid":"https://dxfpracgmlczp.cloudfront.net/2019/03/1553863453523_blurb.jpg","blurb_expanded_image_uid":null,"article":"

Video technology is evolving at a furious pace. And artificial intelligence (AI) and machine learning are the major drivers behind this.

\n\n

Both AI and machine learning have enormous potential to transform video technology and how they will be used to fundamentally change the way we live, work and even shop.

\n\n

One example will illustrate this. In 2018, Thailand’s biggest convenience store chain 7-Eleven implemented facial recognition video technology in its 11,000 stores across Thailand. The technology is used to identify the chain’s loyalty programme members, analyse in-store traffic, suggest purchases and even measure shoppers’ emotion.

\n\n

The chain is also using the technology to allow managers to single out loyalty programme members for promotions. Essentially, the chain views video as not merely a surveillance tool but also a marketing tool, a tool to better understand customers, a tool to improve staff efficiency and, potentially, a tool to customise shoppers’ in-store experience.

\n\n

In the healthcare industry, institutions are beginning to use video analytics to improve patientcare. Videos for example, are being programmed to alert staff if a patient has gone too long without being checked, or to signal that a patient has fallen and needs assistance.

\n\n

In almost every sector, AI-powered video technology is already being used to simplify everyday processes, from enabling easier security checks at the airport to allowing passengers to pay for purchases with a smile.

\n\n

From here on, the tide will only rise, with the projection of one billion video cameras connected to artificial intelligence platforms by 2020[1].

\n\n

Why AI Is So Potent Today: The Ability To Learn Deeply

\n\n

AI is the ability of a machine or a computer programme to think, act and learn like humans. Previously limitations of hardware processing power means that machine learning – an application of AI – could only deploy shallow learning of very large data sets, which looks at data in just three dimensions.

\n\n

With recent, significant advances in processing power of graphical processing units such as a new coding technique known as parallelisation, we can now utilise a deep learning approach where we can look at data in many more levels or dimensions. Hence the word “deep”.

\n\n

Software parallelisation is a coding technique for breaking a single problem into hundreds of smaller problems. The software can then run those 100 or 1,000 processes into 1,000 processing cores, instead of waiting for one core to process the data 1,000 times.

\n\n

With parallelisation, there is a quantum leap forward in how fast we can solve a problem. Having the ability to solve problems faster allows us to go deeper with a problem and process larger, more complex data sets. As the world’s data is set to grow tenfold by 2020[2], for businesses, being able to process data faster and deeper will become a defining factor in whether they can stay ahead of the business curve.

\n\n

Bringing The Potential Of AI Augmentation To Life

\n\n

One of the greatest application of AI and machine learning is in performing low-cognitive functions. AI-enabled devices and machines are able to master and perform tasks that humans cannot do very well. With proper aggregation of information, machines can be better at low-cognitive tasks and often deliver a better quality of service than humans.

\n\n

For example, humans cannot sit and watch two or more video cameras simultaneously. Our attention span simply does not work that way. However, machines are extremely good and detailed at this. While we see objects, the machine sees the smallest finite details – that is, each and every single pixel. Within the pixel, the machine can see even more details: the shade of colours of that image.

\n\n

By aggregating data and by allowing machines to automate responses and solutions, we can boost humans’ interaction with their environment.

\n\n

Here is one example. Imagine a scenario where a law enforcement officer is viewing a large surveillance screen. Between the officer and the large screen is an additional clear screen which can also have video and data cast onto it. Finally, the officer is wearing smart glasses that can project information onto its lenses.

\n\n

In a surveillance situation, a video feed can be shown on the main screen, while the screen in the medium distance will augment the main screen by appearing to layer extra visuals on top, for instance the face of a suspect. The smart glasses will then show detailed text data, for instance car licence plates or descriptions of suspects. The live video, augmented visuals and text data will work in concert. Data on the main viewing screen can even change to show information according to what the officer is seeing through his smart glasses’ lens. While this may sound rather futuristic, it is actually possible today.

\n\n

With AI, there will be massive advancements in how we review and utilise video and data.

\n\n

The City of Hartford in Connecticut, United States of America, is a great case study of how technology can be turned into a force multiplier. Working in tandem with local law enforcement and partners BriefCam and Axis, Milestone Systems was able to enhance the City’s C4 Crime Center and significantly upgrade the Hartford Police department’s ability to prevent and effectively respond to incidents throughout the city.

\n\n

The Crime Center features the Milestone XProtect Smart Wall, which has thirty 55-inch, 4K video monitors connected to a high-powered workstation, which runs the Milestone XProtect Smart Client on every screen. The Center is staffed almost round-the-clock by civilian crime analysts who monitor the 450 PTZ video network cameras from Axis Communications that are located throughout the city.

\n\n

With this centre, instead of spending 30 hours doing low-cognitive, manual tasks such as freezing on a rooftop monitoring a drug house all day and night, the City of Hartford’s officers now sit at their desk and within just a few minutes, know exactly where a drug house is by seeing an augmented reality of foot traffic over time.

\n\n

With the enhanced system, officers can simply go into the data and extract the problem with precision and efficiency. Not only are many crimes now solvable, the video technology is changing how police work will be done in the future.

\n\n

An Intelligent Industrial Revolution

\n\n

Having machines take over low-cognitive tasks will be a significant game changer for years to come.

\n\n

Take Amazon. The online retail giant is applying this to its retail stores where the concept of a checkout is being replaced by customers simply walking out. By using data from smartphones, cameras, sensors, purchase histories and other data points, Amazon is making it possible for its customers to walk into a store, pick up what they need and walk out. Everything else is taken care of by machines. This type of thinking and tool creation is in its earliest infancy but will continue to address areas where value can be added to our lives.

\n\n

In the book The Inevitable, author Kevin Keely says the next 10,000 startups will be based on finding applications for AI, similar to what happened with electrification during the second industrial revolution. The intelligent industrial revolution is beginning to happen all around us. It will be very disruptive within the security and surveillance industry. But it will also be insightful and liberating as it will free humans to perform higher cognitive processes.

\n\n

[1] Cloud-To-Edge Inference For Smarter Ai Cities, NVIDIA, 2018

\n\n

[2] Data Age 2025: The Evolution of Data to Life-Critical, SeaGate, 2017

\n","article_image_uid":"https://dxfpracgmlczp.cloudfront.net/2019/03/1553863459460_main pic.jpg","article_expanded_image_uid":null,"article_image_caption":"","publications":"5","tags":"video,AI,artificial intelligence,video technology,video and AI,revolution of video technology,How AI will change video technology","category":"Products & Technologies"}],"pageInfo":{"hasNextPage":false,"hasPreviousPage":true,"totalCount":42,"totalPages":4,"currentIndex":40}}}}