
AT&T Intellectual Property has been granted U.S. patent #12095813 for a system that detects homoglyph attacks, invented by Antoine Diffloth, Natalie Gilbert, and Sundaresan Manoharan. The system uses convolutional and siamese neural networks to identify malicious domain name variations, where visually similar characters from different alphabets (like 'O' vs. '0' or 'l' vs. 'I') are used to deceive users. This helps businesses safeguard their protected domain names from cyber threats. [Source image: Dragon Claws / iStock]
This week, newly granted patents also include:
- American Security Products' physical exchange using a safe
- Bank of America's data processing by a self-correcting, deep neural network integrated within a memory resource
- The University of Texas System's controlling prosthetic devices by gestures and other modalities
- Care View Communications' monitoring a fall state of a patient while minimizing false alarms
- Comcast Cable Communications' dynamic virtual element positioning in an augmented reality environment
- Fluid Dynamic Sciences' fast feature recognition and mesh generation in structural design
- GEMMY Industries' firework lighting device
- Intellective AI’s alert directives and focused alert directives in a behavioral recognition system
- KoiReader Technologies' edge computing device and system for vehicle, container, railcar, trailer, and driver verification
- Legacy Ventures' clot retrieval system
- Marvell Asia's system and method for bitcoin mining with reduced power
- Mary Kay’s delivery of active ingredients from plant materials to skin
- Ryder Integrated Logistics' vehicle fleet control systems and methods
- SightGlass Vision's methods and devices for reducing myopia in children
- The Nielsen Company's methods and apparatus to collect and process browsing history
- Walmart Apollo's systems for detecting and tracking humans in captured images
Get on the list.
Dallas Innovates, every day.
Sign up here to get what’s new and next in Dallas-Fort Worth.