Robust real-time detection of multi-color markers on a cell phone

Abstract

We describe a fast algorithm to detect special multi-color markers with a camera cell phone. These color markers can be used for environmental labeling, for example, as a wayfinding aid for persons with visual impairment. Using a cascade of elemental detectors, robust detection is achieved at an extremely low computational cost. We also introduce a strategy to select surfaces for the marker that ensure very low specular reflection, thus facilitating color-based recognition.

An integrated artificial vision framework for assisting visually impaired users

“…signs in Braille) or by ad hoc devices. Meaningful from the Computer Vision standpoint are the passive signs [13,14] and the reflective signs for infrared illuminators [15] . Navigation in traffic intersections has been addressed in [16,17] .…”

The last meter

“…2). The detection software uses the algorithm described in [4], which returns the position (in the image) of four equi-spaced keypoints on the marker’s circumference as well as on the marker’s center. Given the known size of the marker (16 cm in diameter) and the optical/imaging characteristics of the camera, the camera’s pose (position and orientation) can be estimated from these five keypoints.…”

“…Indeed, in our previous work on our smartphone based color marker detection system [7,4,13], we explored and tested a variety of user interface (UI) options before arriving at the UI used in our current system (see the “Apparatus” Section). Given such a mechanism (which is fixed in our current study), we explore the effects that fundamental constraints imposed by the object recognition technology itself have on the user’s performance in acquiring well-framed pictures.…”

Fast Tracking System for Multi-colored Pie-shaped Markers

“…The shape properties of the black-and-white patterns in a marker are used for marker detection in many applications (Claus and Fitzgibbon 2004;Fiala 2004;Kato and Billinghurst 1999;Naimark and Foxlin 2002;Rohs and Gfeller 2004;Wagner and Schmalstieg 2007). Several marker tracking systems have been developed for augmented reality and human-machine interfaces based on distinctive colors, rather than the shape analysis of black-and-white patterns (Lee and Woo 2009;Bagherinia and Manduchi 2013;Cho and Neumann 1998;State et al 1996;Johansson and Balkenius 2007;Coughlan and Manduchi 2007;Sýkora et al 2008). State et al (1996) developed a ring-type colored marker tracking system that improved the accuracy of augmented reality registration.…”

“…Sýkora et al (2008) introduced a lightweight and robust tracking technique based on colored balls, which can estimate the 3-D positions of several visible colored balls in real-time using a standard color camera. Bagherinia and Manduchi (2013) developed a robust real-time detection system for multi-colored markers on a cell phone, where pie-shaped color markers with four colored sectors were introduced as environmental labels for blind route-finding. This type of pie-shaped color marker is easily extended by adding new colored sectors with no change in the overall marker size.…”