ViPR - Visual Search
 |
|||||||||||||
 |
|||||||||||||
| Download this as a PDFEvolution Robotics focuses much of its R&D in the area of computer vision because of the potential it offers for current and future product developments. Vision is a versatile capability useful for many tasks including, but not limited to autonomous navigation, object recognition and human-robot interaction. In addition, camera technologies have become a commodity thanks to the widespread application of webcams, digital cameras, and cell-phone cameras. The main challenge with vision is to develop algorithms that reliably and efficiently solve a problem in realistic settings, with limited computing power.
ViPR Evolution Robotics ViPR (visual pattern recognition) technology provides a reliable and robust vision solution that truly gives electronic devices the ability to detect and recognize complex visual patterns - in effect, to see.
How It Works The combination of several key elements allows the ViPR technology to achieve a high level of performance.First, is the choice of descriptors it uses to encode unique visual patterns such as the corner of an object or the print on a label. As the most distinct regions (called features) are localized in the image, unique descriptors are computed for each of them. Several hundred such features are automatically extracted and stored in a database to describe the unique patterns in each image.
Second, is the ability to analyze a new image to collect sufficient evidence to reliably find a match within an extremely large set of possible candidates. The algorithm that ViPR uses to select the correct candidate is similar to a voting mechanism: each feature votes for the candidate which includes a similar feature (e.g., a corner feature in the new image that matches a corner feature in a trained image). The correct candidate will receive the largest number of votes since most of the features will be in agreement; however, a single or a few votes might be incorrectly cast on wrong candidates. The likelihood that a large number of votes are cast on the wrong candidate is small, demonstrating that the algorithm is very reliable in selecting the correct match.
Third, is the ability to do all this computation in an extremely efficient manner: recognition happens in a fraction of a second when searching a database of several hundred patterns. Performance/Features
Using ViPR Technology A ViPR-enabled device can automatically detect and recognize visual patterns using low- or high-end camera sensors. The algorithms that make up the technology are particularly robust and provide an unprecedented level of reliability even with heavy distortions that can be introduced by the imaging device, a wide range of lighting conditions, and pattern occlusions.LaneHawk™, a new loss-prevention product for retailers, also uses ViPR technology to recognize grocery items by analyzing the printed patterns on their box, instead of using the barcode. To find out more about LaneHawk, visit www.evoretail.com.
ViPR technology can be purchased for development as part of the ERSP® 3.1 SDK. The ERSP 3.1 SDK is available for Linux and the ERSP 3.0 SDK is available for Windows. It can also be licensed as a standalone technology to be included in your products. |
Main RSS Feed of All Posts