Specialists from St. Petersburg State University (SPbSU) and the St. Petersburg Federal Research Center of the Russian Academy of Sciences (SPC RAS) have trained a neural network to identify vulnerabilities in computer tests (CAPTCHA). These texts allow you to determine who is trying to enter the site — a user or a computer bot. This was reported by the press service of St. Petersburg State University.
The neural network was trained on a special dataset — a set of data for machine learning.
The main difference between our approach and analogues is that it allows you to recognize much more complex text tests, which contain more than 20 distortions at once.
The developers emphasized that their method will help improve the security of Internet resources.
Information security specialists can refine the methods of recognition and bypassing the captcha, and on this basis develop improved algorithms for protection against automated attacks.
The new method allowed to recognize 63% of distorted captcha images. This indicates the insecurity of sites using this type of computer texts, the St. Petersburg specialists summarized.
Earlier www1.ru reported that Yandex Cloud presented a new neural network-empath that recognizes emotions in human speech.
Read materials on the topic:
Russian AI will start making voice comments to scooter riders and smokers
A neural network has been developed in St. Petersburg to detect brain hematomas on CT scans
Now on home
The service contains data on 45,000 fraudulent sites
The state has accelerated the introduction of unmanned solutions, integrating them into the real sector of the economy
Modernized engines may equip the Lada Azimut crossover
The price is 132 billion 265.8 million rubles
The manufacturer plans to strengthen its lineup of light commercial vehicles
The production of carbon fiber was organized in the shortest possible time
Electric vans will speed up the repair of urban transport infrastructure
Countries are working to synchronize regulations in the field of AI
The service's average daily audience is 55 million people
Stable Isomaterial Based on Metakaolin Has a Density Below 300 kg/m³
Re-identification quality improved twofold with new DynaMix method
Russians will be able to find out about debts online