NEW YORK, NY and PARIS–(Marketwire – July 7, 2010) – LTU technologies (www.ltutech.com), a pioneer of image recognition and search technologies, today announced the release of its LTU engine 4.0 and enhanced LTU engine/ON demand platform featuring next generation image matching algorithm and new functionalities tailored specifically to media intelligence and mobile solutions. The new features include: detection of very fine differences between two images (‘fine comparison’) and significantly expanded capacity for searchable images. The new image matching algorithm offers improved robustness against glare, noise, cropping, image compression, artifacts, and outperforms any image matching technologies seen on the market thus far.
With over 96% accuracy when matching many types of images that have been flipped, closely cropped, cut out and used in collages or placed among text-heavy pages, the newly released LTU engine provides an automated fast and easy way to identify and analyze ad placements in a fraction of a second. The new ‘fine comparison’ feature is even capable of pin-pointing subtle differences between matching images that are difficult for the human eye to catch. These features enable media intelligence and ad tracking services to timely and accurately report on media placements, detect differences in advertisements between regions and as a result, increase the accuracy and speed of their services.
“One year ago, we decided to integrate LTU’s technology in our printed ad monitoring process for one reason. It’s just the best known solution on the market,” said Philippe Chéron, President of Yacast, a Paris-based leader of monitoring the media distribution of music and advertising.
“Algorithms, scalability and integration are remarkable; LTU’s 10 years of experience cannot be improvised. LTU is one of our main partners and will deeply contribute to our new technological projects and business development.”
For developers looking to integrate powerful visual search capabilities into their mobile applications, the new image matching algorithm has been carefully developed to handle extreme glare, rotation, noise, compression and other conditions commonly seen with images taken with a mobile device. As a result, the image matching functionality has improved by 10 – 20 points in accuracy and boasts near 100 percent positive recognition on many content categories such as paintings, cover art, movie posters and more.
“As a pioneer in image recognition and visual search, we are continually improving our capabilities and uncovering new markets that will benefit from the technology,” said Alex Winter, LTU technologies co-founder and CEO. “Also, thanks to the amazing work of our world-class team of image recognition PhDs, systems architects and software engineers, we are again reinforcing our commitment to provide our customers with the best, most accurate and fastest technology on the market.”
About LTU technologies
Founded in 1999 by scientists from MIT, Oxford University and INRIA, LTU technologies provides cutting-edge image recognition technologies to leading commercial and government organizations worldwide.
The technology has successfully been integrated into both commercial and government solutions for brand and copyright protection, police investigations (stolen art, child abuse), media intelligence and advertisement tracking, site search for eCommerce and stock photography, mobile applications and many more. For more information, visit http://www.ltutech.com.
Related posts:
- Third Generation Aptina™ A-Pix™ Technology Enhances Mobile Imaging Solutions
- Duggal Visual Solutions to Offer New KODAK PROFESSIONAL ENDURA EP Media and KODAK PROFESSIONAL UV-Curable Display Film-Plus as Part of its Media Portfolio
- OmniVision Launches 12.6-Megapixel Mobile Sensor with 1080p/60 HD Video and Electronic Image Stabilization
