近日，微軟和英特爾開發了防病毒軟體，可以將惡意軟體轉換為2D圖像，並可以通過神經網絡對其進行檢查。
- 名為STAMINA的軟體可將惡意軟體位轉換為2D圖像
- 然後使用計算機視覺軟體檢查這些圖像
- 該方法可以減少軟體檢查的數據點數量
- 在對惡意軟體進行分類方面顯示出99％的成功率
微軟和英特爾已經合作開發一種新型的惡意軟體檢測工具。
該項目稱為靜態惡意軟體-圖像網絡分析（STAMINA），是科技巨頭的共同努力，旨在開發一種軟體，該軟體可通過將惡意代碼轉換為可通過深度學習進行評估的灰度圖像來嗅探出惡意代碼。
英特爾和微軟表示，一種名為STAMINA的新型病毒檢測軟體將惡意軟體轉換為2D圖像，可以通過計算機視覺算法對其進行掃描。
具體來說，STAMINA使用設計用於分析圖像的計算機視覺軟體將一維惡意軟體位轉換為二維灰度圖像，然後「查看」圖像中可能指示特定類型惡意代碼的模式。
組裝圖像後，STAMINA然後將其調整為較小的尺寸，以使其更易於查看。
據研究人員稱，這種壓縮有助於避免該軟體評估數十億像素（這可能會減慢該過程），並且不會對其識別惡意軟體的能力產生負面影響。
據ZDNet稱，對STAMINA的培訓使用了從Windows Defender（該公司生產的防病毒軟體）中提取的數百萬個惡意軟體示例，並且在其發現計算機病毒的任務中顯示了早期的希望。
該系統對惡意軟體進行分類的準確率略高於99％，誤報率低於2.6％。
The approach could help reduce the amount of data that needs to be scanned by algorithms and make malware detection more efficient (stock)
顯然，該AI在較小的文件大小方面已顯示出更多的成功，但是據微軟稱，STAMINA最終可以部署為僅專注於較小的文件。
無論哪種方式，該工具都可以是對當前掃描惡意軟體的方法的一種改進，該方法可以創建非常大的數據點，並增加了惡意軟體掉入裂縫的機會。
Microsoft and Intel develop antivirus software that turns malware into 2D images that can be examined by a neural network
- The software, called STAMINA, converts malware bits into 2D images
- It then examines those images using computer visions software
- The approach could reduce the number of data points examined by software
- It has shown a 99 percent success rate in classifying malware
Microsoft and Intel have partnered up in an effort to develop a new kind of malware detection
The project, called Static Malware-as-Image Network Analysis (STAMINA), is a joint effort by the tech giants to develop a software that sniffs out malicious code by converting it into greyscale images that can be assessed by utilizing deep-learning.
Intel and Microsoft say a new kind of virus-detecting software called STAMINA converts malware into 2D images that can be scanned by a computer vision algorithm (stock)
Specifically, STAMINA converts one-dimensional malware bits into two-dimensional greyscale images and then 'looks' at the images for patterns that may indicate specific types of malicious code using computer vision software designed to analyze images.
One the image is assembled, STAMINA then resizes it into a smaller dimension to make it easier to view.
This compressions, according to researchers helps avoid needing the software to assess billions of pixels - which would likely slow the process - and does not negatively affect its ability to identify malware.
According to ZDNet, STAMINA is trained using millions of examples of malware pulled from Windows Defender - an antivirus software made by the company - and has shown early promise in its missions to spot computer viruses.
The system has a little more than 99 percent accuracy with classifying malware and a false positive rate of below 2.6 percent.
The approach could help reduce the amount of data that needs to be scanned by algorithms and make malware detection more efficient (stock)
The AI apparently has apparently shown more success with smaller file sizes but according to Microsoft, STAMINA could eventually be deployed to focus solely on smaller files.
Either way the tool could be an improvement over current methods of scanning for malware that create very large data points and increase the chances of malware falling through the cracks.