Computer vision can help spot cyber threats with startling accuracy

Machine learning, Computer security, Computer vision, image-classifiers

This article is part of our , a series of posts that explore the latest findings in artificial intelligence.

The last decade’s growing interest in was triggered by the proven capacity of neural networks in computer vision tasks. If you train a neural network with enough labeled photos of cats and dogs, it will be able to find recurring patterns in each category and classify unseen images with decent accuracy.

What else can you do with an image classifier?

In 2019, a group of cybersecurity researchers wondered if they could treat security threat detection as an image classification problem. Their intuition proved to be well-placed, and they were able to create a machine learning model that could detect malware based on images created from the content of application files. A year later, the same technique was used to develop a machine learning system that detects phishing websites.

The combination of binary visualization and machine learning is a powerful technique that can provide new solutions to old problems. It is showing promise in cybersecurity, but it could also be applied to other domains.

Detecting malware with deep learning

The traditional way to detect malware is to search files for known signatures of malicious payloads. Malware detectors maintain a database of virus definitions which include opcode sequences or code snippets, and they search new files for the presence of these signatures. Unfortunately, malware developers can easily circumvent such detection methods using different techniques such as obfuscating their code or using polymorphism techniques to mutate their code at runtime.

Dynamic analysis tools try to detect malicious behavior during runtime, but they are slow and require the setup of a sandbox environment to test suspicious programs.

In recent years, researchers have also tried a range of to detect malware. These ML models have managed to make progress on some of the challenges of malware detection, including code obfuscation. But they present new challenges, including the need to learn too many features and a virtual environment to analyze the target samples.

Binary visualization can redefine malware detection by turning it into a computer vision problem. In this methodology, files are run through algorithms that transform binary and ASCII values to color codes.

In a , researchers at the University of Plymouth and the University of Peloponnese showed that when benign and malicious files were visualized using this method, new patterns emerge that separate malicious and safe files. These differences would have gone unnoticed using classic malware detection methods.

Machine learning, Computer security, Computer vision, image-classifiers
When the contents of binary files are visualized, patterns emerge that separate malware from safe files.

According to the paper, “Malicious files have a tendency for often including ASCII characters of various categories, presenting a colorful image, while benign files have a cleaner picture and distribution of values.”

When you have such detectable patterns, you can train an to tell the difference between malicious and safe files. The researchers created a dataset of visualized binary files that included both benign and malign files. The dataset contained a variety of malicious payloads (viruses, worms, trojans, rootkits, etc.) and file types (.exe, .doc, .pdf, .txt, etc.).

The researchers then used the images to train a classifier neural network. The architecture they used is the self-organizing incremental neural network (SOINN), which is fast and is especially good at dealing with noisy data. They also used an image preprocessing technique to shrink the binary images into 1,024-dimension feature vectors, which makes it much easier and compute-efficient to learn patterns in the input data.

Machine learning, Computer security, Computer vision, image-classifiers
Architecture of deep learning system that detects malware from binary visualization.

The resulting neural network was efficient enough to compute a training dataset with 4,000 samples in 15 seconds on a personal workstation with an Intel Core i5 processor.

Experiments by the researchers showed that the deep learning model was especially good at detecting malware in .doc and .pdf files, which are the preferred medium for . The researchers suggested that the model’s performance can be improved if it is adjusted to take the filetype as one of its learning dimensions. Overall, the algorithm achieved an average detection rate of around 74 percent.

Detecting phishing websites with deep learning

are becoming a growing problem for organizations and individuals. Many phishing attacks trick the victims into clicking on a link to a malicious website that poses as a legitimate service, where they end up entering sensitive information such as credentials or financial information.

Traditional approaches for detecting phishing websites revolve around blacklisting malicious domains or whitelisting safe domains. The former method misses new phishing websites until someone falls victim, and the latter is too restrictive and requires extensive efforts to provide access to all safe domains.

Other detection methods rely on heuristics. These methods are more accurate than blacklists, but they still fall short of providing optimal detection.

In 2020, a group of researchers at the University of Plymouth and the University of Portsmouth used binary visualization and deep learning to develop a .

The technique uses binary visualization libraries to transform website markup and source code into color values.

Machine learning, Computer security, Computer vision, image-classifiers

As is the case with benign and malign application files, when visualizing websites, unique patterns emerge that separate safe and malicious websites. The researchers write, “The legitimate site has a more detailed RGB value because it would be constructed from additional characters sourced from licenses, hyperlinks, and detailed data entry forms.

Whereas the phishing counterpart would generally contain a single or no CSS reference, multiple images rather than forms and a single login form with no security scripts. This would create a smaller data input string when scraped.”

The example below shows the visual representation of the code of the legitimate PayPal login compared to a fake phishing PayPal website.

Machine learning, Computer security, Computer vision, image-classifiers

The researchers created a dataset of images representing the code of legitimate and malicious websites and used it to train a classification machine learning model.

The architecture they used is MobileNet, a lightweight that is optimized to run on user devices instead of high-capacity cloud servers. CNNs are especially suited for tasks including image classification and object detection.

Once the model is trained, it is plugged into a phishing detection tool. When the user stumbles on a new website, it first checks whether the URL is included in its database of malicious domains. If it’s a new domain, then it is transformed through the visualization algorithm and run through the neural network to check if it has the patterns of malicious websites. This two-step architecture makes sure the system uses the speed of blacklist databases and the smart detection of the neural network–based phishing detection technique.

The researchers’ experiments showed that the technique could detect phishing websites with 94 percent accuracy. “Using visual representation techniques allows to obtain an insight into the structural differences between legitimate and phishing web pages. From our initial experimental results, the method seems promising and being able to fast detection of phishing attacker with high accuracy. Moreover, the method learns from the misclassifications and improves its efficiency,” the researchers wrote.

Machine learning, Computer security, Computer vision, image-classifiers
Architecture of deep learning system that detects phishing websites through binary visualization

, cybersecurity lecturer at the University of Portsmouth and co-author of both papers. According to Shiaeles, the researchers are now in the process of preparing the technique for adoption in real-world applications.

Shiaeles is also exploring the use of binary visualization and machine learning to detect malware traffic in .

As machine learning continues to make progress, it will provide scientists new tools to address cybersecurity challenges. Binary visualization shows that with enough creativity and rigor, we can find novel solutions to old problems.

This article was originally published by Ben Dickson on , a publication that examines trends in technology, how they affect the way we live and do business, and the problems they solve. But we also discuss the evil side of technology, the darker implications of new tech, and what we need to look out for. You can read the original article .

Internet Explorer Channel Network
News Related

OTHER NEWS

Artists launch colorful 'jellyfish' balloon to the stratosphere in inaugural test flight

© Provided by Space Beyond Earth’s “Living Light” artwork. A group of artists has launched a colorful, jellyfish-looking craft into Earth’s stratosphere as part of a unique test flight.  This ... Read more »

Inspiration4 astronaut Sian Proctor reflects on historic SpaceX spaceflight experience

© Provided by Space Sian Proctor served as the pilot for the private SpaceX mission Inspiration4. CAPE CANAVERAL, Fla. — Sian Proctor, the pilot for SpaceX’s historic Inspiration4 mission, is ... Read more »

Chiefs vs Washington live stream: how to watch NFL online from anywhere

This clash of the teams with the two worst defenses in the NFL is a must-win for both franchises, which are matched at 2-3. Expect both quarterbacks to rack up ... Read more »

Python is now no. 1. This training can make sure you're a certified expert

TLDR: The Python Programming and Git Certification Bundle includes beginner and advanced training in how to use Python to code all your biggest programming projects. Well, it finally happened. After ... Read more »

Support scientific interest with a price drop on this comprehensive robot arm kit

The pursuit of scientific and technological understanding and innovation requires consistent dedication and a desire to learn. For many eventual individual successes in the industry, that journey starts at a ... Read more »

Apple's October ‘Unleashed' event: what to expect

Apple has announced an event for October 18th, with a video containing the title “Unleashed.” If you’ve been watching the rumor mill, it won’t come as a surprise — the ... Read more »

Packers vs Bears live stream: how to watch NFL online from anywhere

The Packers’ seemingly reluctant win over the Bengals last weekend was one of the strangest games in recent memory, so much so that a Bears triumph can’t be ruled out… ... Read more »

Chargers vs Ravens live stream: how to watch NFL online from anywhere

Two of the oustanding figures from the early weeks of the season, Justin Herbert and Lamar Jackson, go head-to-head in a game that’s got FUN written all over it. The ... Read more »

Russian crew returns from shooting the first feature film on the ISS

Shooting for the first feature-length movie in space has wrapped. Space.comreports Russian actress Yulia Pereslid, producer Klim Shipenko and cosmonaut Oleg Novitskiy have returned to Earth after the first two ... Read more »

How to watch the Apple October 2021 event: will new MacBooks get unleashed?

Apple has announced it is holding it’s ‘Unleashed’ event, where it’s expected to show off new MacBook Pros running on upgraded hardware, and we’ll show you how to watch the ... Read more »

Jabra's huge Epic Daily Deals sale includes $50 off the Elite 85t earbuds

One of Amazon’s Epic Daily Deals today is a gigantic sale on Jabra true wireless earbuds. The items available include multiple generations of Jabra’s true wireless earbuds going as far ... Read more »

Samsung Galaxy S22 leak reveals more about the smartphone's design

We’ve already heard plenty about the Samsung Galaxy S22 as its expected launch date gets closer and closer, but the leaks keep on coming – and the latest one gives ... Read more »

Daniel Craig's James Bond movies show where the Star Wars sequels went wrong

Big No Time To Die spoilers follow. Watch the film before reading this. They actually went and did it… For a franchise that’s spent most of its six decades adhering ... Read more »

7 reasons why programmers burn out — and 7 ways to prevent it

When Jerry started their job as a junior developer, they couldn’t wait to get to work each morning. And each evening, even after their 10-hour workdays, they’d still feel energized. ... Read more »

Best Buy's Black Friday plans: Talking Tech podcast

Hit play on the player above to hear the podcast and follow along with the transcript below. This transcript was automatically generated, and then edited for clarity in its current ... Read more »

What do 10 million fewer iPhone 13 shipments mean?

While the iPhone 13 range seemingly launched without a hitch in September, a recent Bloomberg report claimed that Apple was expecting to ship 10 million fewer units of its newest ... Read more »

Russian spacecraft returns to Earth with movie crew aboard

A Russian actress and director-producer who spent 12 days aboard the International Space Station returned to Earth early Sunday along with cosmonaut Oleg Novitskiy who had been in space for ... Read more »

Three ways to add quicker access to the Windows 11 Task Manager

© Provided by Windows Central Here are three ways to create quick access to the Task Manager on Windows 11. On Windows 11, you can no longer right-click the Taskbar ... Read more »

HP Omen 15 (2021)

© Provided by TechRadar The HP Omen 15 from the front, displaying a vivd, colorful desktop background HP Omen 15 (2021): Two minute review If you’re looking for a powerful ... Read more »

The pandemic laid bare the deep problems in America - from healthcare to housing. Now the question is: will we do anything about it?

Sophia Garcia, center, with Legacy LA in Boyle Heights joins housing advocates and tenants gathered against eviction of tenants from the 50 unit Tokio Hotel apartments n downtown Los Angeles. ... Read more »

The dark side of wellness: the overlap between spiritual thinking and far-right conspiracies

It was the afternoon of 4 July 2020, and Melissa Rein Lively’s video was about to go viral. A PR executive in Arizona, she already had the appearance of a ... Read more »

Britain's Got Talent Finalist Airlifted To Hospital With 'Severe Injuries' After Stunt Goes Wrong On US Version Of Show

A British escape artist has been airlifted to hospital after his act went drastically wrong during rehearsals for a spin-off of America’s Got Talent. Jonathan Goodwin, who is from Wales, ... Read more »

Should you buy a Lego calendar now or wait for the Black Friday Lego deals?

Some of the most popular Black Friday Lego deals are on the company’s popular advent calendar sets – these are your traditional December day-counters, except instead of chocolate, cheese or ... Read more »

How to enable TPM 2.0 for Windows 11

One of the more controversial things about Windows 11 has been the system requirements that have been put in place, especially with TPM 2.0. Although the basic requirements are not ... Read more »

Sir David Amess: Anti-terror police search London homes

Anti-terror police continued to search properties in London on Sunday after the Met was given more time to question Sir David Amess’ suspected killer. Officers stood guard outside Ali Harbi ... Read more »

How to make an authentic cup of matcha

You’ve no doubt seen matcha on menus in restaurants and coffee shops. Maybe the latte form has even become your go-to drink order. And yet, you’ve probably given little thought ... Read more »

10 leggings celebrities can't stop wearing

— Recommendations are independently chosen by Reviewed’s editors. Purchases you make through our links may earn us a commission. Leggings are a staple in many people’s wardrobes. Whether you’re wearing ... Read more »

Macron condemns 'inexcusable' crackdown on 1961 Algerian protests in Paris

President Emmanuel Macron has described the deadly crackdown by Paris police on a peaceful protest by Algerians as “inexcusable for the Republic”, going further than his predecessors in admitting crimes ... Read more »

Strictly Come Dancing in ‘homophobia and sexism' row over ‘smutty' comments made to male dancers

Strictly Come Dancing has been accused of sexism for allegedly “predatory” comments made by judges towards male dancers. The BBC’s primetime dancing show has reportedly received a higher than usual ... Read more »

David Amess: Father ‘traumatised' after son held over MP's killing

The father of the suspect in the fatal stabbing of Tory MP Sir David Amess has said he feels “traumatised” by his son’s arrest. Police have yet to officially name ... Read more »

China to keep up scrutiny of internet sector

This illustration file picture taken in Paris shows the logo of the US social network application Linkedin on the screen of a tablet. Microsoft on Oct. 14 said it will ... Read more »

Why 3D Pinball needs to come back in Windows 11

The release of Windows 11 has come and gone, but already there’s feature wishes for future releases of the newest Windows upgrade. One in particular from me is for 3D ... Read more »

Everton vs West Ham live stream: how to watch Premier League online from anywhere now

Former Everton boss David Moyes faces his old Merseyside sparring partner Rafa Benitez today in a match that could have significant implications for both managers top six aspirations. Read on ... Read more »

Dolphins vs Jaguars live stream: how to watch NFL online from anywhere

The Jaguars may be the unofficial London team, but the Dolphins are (apparently) the best supported franchise in Blighty, thanks in no small part to the exploits of Ace Ventura. ... Read more »

Susan Sarandon Recalls Final Conversation With Ex-Lover David Bowie: ‘We Said Things That Needed To Be Said'

Susan Sarandon has spoken of how she reconnected with her former lover David Bowie before his death in 2016. The Hollywood star said they both had “things that needed to ... Read more »

Dua Lipa wears a glittering safety pin mini skirt at the Versace & Frieze party

Dua Lipa wore a sequin mini skirt with gold safety-pin detailing and matching neon safety pin earrings as she attended the Versace & Frieze party in London last night (October ... Read more »

Microsoft Windows 11 released: Check if your PC is compatible with the Windows 10 successor

Microsoft released Windows 11 to all eligible devices earlier this month. The new Windows 11 update comes with an all-new design and some key upgrades under the hood. For a ... Read more »

Valorant: How To Play With a Controller Complete Guide

Valorant is one of the fastest-growing FPS shooters on PC that focuses on using a mouse and keyboard, but can you use a controller to play the game? © GiveMeSport ... Read more »

French fishermen threaten Channel blockade as Macron snubbed by EU countries

Fishermen gut freshly caught fish aboard fishing boat ‘About Time’ while trawling in the English Channel from the Port of Newhaven, East Sussex, U.K – Jason Alden /Bloomberg France’s failure ... Read more »

Robert Webb Labels Strictly Come Dancing ‘Brutal' And ‘Weird': ‘I Was Almost Weeping As It Is So Intense'

Robert Webb labelled Strictly Come Dancing “brutal” and “weird” just a week before he quit the BBC dance show due to health concerns. On Wednesday, the Peep Show star announced ... Read more »