Algorithm Provides Enhanced Recognition Capabilities for Use in a Wide Range of Robotics, Artificial Intelligence and Computer Vision Applications
VILNIUS, Lithuania, Sept. 10 /PRNewswire/ -- Neurotechnology (http://www.neurotechnology.com), a provider of high-precision biometric and object identification technologies, today announced the availability of SentiSight 2.0 SDK universal object recognition technology for the development of robotics, artificial intelligence and computer-based vision applications. The SentiSight 2.0 algorithm provides significantly enhanced 2D and 3D object recognition quality using still or video images from most digital cameras, including Webcams. Among the new features in SentiSight 2.0 are the ability to find and count the number of objects in a scene and the ability to compare and identify pictures even when the perspective has changed. The algorithm is tolerant to scale, rotation and pose under a variety of image conditions, providing the versatility required for a wide range of computer and machine vision applications, from manufacturing and security systems to Web-based image search engines. Because SentiSight can process video streams in real time, it is suitable for use in autonomous robot navigation, assembly line parts recognition and other applications that require fast and accurate real-time identification.
"Our AI and Robotics Division focused a significant amount of effort over the past year to develop an even more versatile and capable object recognition algorithm for our SentiSight 2.0 SDK," said Denis Kochetkov, SentiSight project leader for Neurotechnology. "SentiSight 2.0 continues the tradition we have built with our biometric identification technologies over the past 18 years: providing products of high quality and reliability with cost-effective pricing."
SentiSight 2.0 enables fully automatic and manual object learning as well as simultaneous multiple object detection and recognition. Used in conjunction with a digital camera or other visual input device, the algorithm in SentiSight 2.0 enables a computer or robot to "learn" 2D and 3D objects or a series of objects in a scene by extracting specific features and object descriptors from different sides, distances and angles. SentiSight then develops an object model that can be stored in a database. Later, when presented with an image or scene from a live camera, still image or video stream, the SentiSight algorithm can detect whether a particular object is in the scene, identify where the object is located and even count the number of objects in the scene.
While SentiSight 1.0 identified the point where a recognized object was located, it provided little additional information about the object. SentiSight 2.0 features the ability to identify and approximately outline the region an object occupies in a scene, providing additional information about the size, orientation and scale of the recognized object.
SentiSight 2.0 supports both Microsoft Windows and Linux operating systems and gives developers complete control over SDK data input and output, enabling the functions to be used with most cameras.
A new SentiSight 2.0 algorithm demo application is also available. This completely new demo application provides a convenient user interface and enables offline working mode in both the learning and recognition stages.
SentiSight 2.0 SDK is available now with highly competitive licensing options from Neurotechnology and from partners and distributors worldwide. A 30-day trial version with full functionality is also available for download. For more information, please go to http://www.neurotechnology.com.
Neurotechnology is a provider of high-precision biometric fingerprint, face and iris identification algorithms, object recognition technology and software development products. More than 1900 system integrators, security companies and hardware providers integrate Neurotechnology's algorithms and software development technologies into their own products, with millions of customer installations worldwide.
Neurotechnology's identification algorithms have consistently earned the highest honors in some of the industry's most rigorous competitions, including the National Institute of Standards & Technology (NIST)'s Fingerprint Vendor Technology Evaluation (FpVTE) and the International Fingerprint Verification Competitions (FVC).
Drawing from years of academic research in the fields of
neuroinformatics, image processing and pattern recognition, Neurotechnology
was founded in 1990 in Vilnius, Lithuania under the name Neurotechnologija
and released its first fingerprint identification system in 1991. Since
that time the company has released more than 40 products and version
upgrades for identification and verification of objects and personal
identity. On April 14, 2008 the company announced an official name change
Jennifer Allen Newton
Bluehouse Consulting Group, Inc.
Jennifer (at) bluehousecg.com
Copyright©2008 PR Newswire.
All rights reserved