Machine vision software developer Robotic Vision Technologies Inc. (RVT) and its partners have been awarded four new patents for its vision-guided robotic automation software by the U.S. Patent & Trademark Office. “We believe that vision is the main enabler for global machine automation,” said Rick Weidinger, founder and CEO of RVT. “Vision guided robotics is advancing at an extraordinary pace, and these patents recognize the innovation in machine vision robotics technology that we are bringing to the marketplace.” The first patent is the automatic feature detection and tracking of moving parts for a program to learn the shape and features of an object. This allows a machine to handle an object even when it is moving, adjusting for changes in location and orientation the way people automatically do. The second patent includes safety features using light-based sensors for places identified where humans might enter or exit. A model of an industrial robot’s location is created using low-cost sensors and then monitored so that the robot will note when something could be in its way. The safety system instructs the robot to slow or stop depending on the extent of the safety space breach to avoid endangering a person. This vision safety system eliminates the need for costly and traditional safety cages or laser screens. The third patent features distance determination between radio frequency identification device (RFID) tags. Using two or more RFID tags allows for additional position information to be provided to vision guidance systems. A relative distance between the tags is calculated and distance determination may be implemented for robotic sensing, assembly systems and wearable technology to track motion and positions of objects. The last patent includes foreground motion detection in compressed video data with software that can tell the difference between background and foreground features in compressed video streams. Compressed video is often used in high-speed vision systems, but traditional systems need to totally decompress to analyze what it is seeing. This technology allows for extraction of foreground content of a compressed video stream directly from the compressed video allowing the background to be ignored as desired. These new patents bring RVT’s total granted patent portfolio to 10, with another 10 pending. RVT’s software and technologies enable image recognition, machine vision, machine learning and robot guidance processes in industrial and nonindustrial markets. “These patents are a testament to the extraordinary work that has been accomplished and the growing recognition of vision guided robotics as an integral part of cutting-edge machines,” Weidinger said. “They underscore how, as active participants in vision guided robotics technology, we work hard to see what’s coming next, understand the demand and build it.”