Classification of road sign type using mobile stereo vision

Abstract

This paper presents a portable mobile stereo vision system designed for the assessment of road signage and delineation (lines and reflective pavement markers or "cat's eyes"). This novel system allows both geometric and photometric measurements to be made on objects in a scene. Global Positioning System technology provides important location data for any measurements made. Using the system it has been shown that road signs can be classified by nature of their reflectivity. This is achieved by examining the changes in the reflected light intensity with changes in range (facilitated by stereo vision). Signs assessed include those made from retro-reflective materials, those made from diffuse reflective materials and those made from diffuse reflective materials with local illumination. Field-testing results demonstrate the systems ability to classify objects in the scene based on their reflective properties. The paper includes a discussion of a physical model that supports the experimental data.

Traffic Sign Detection and Tracking Using Robust 3D Analysis

“…fulfilling color-based and region-level analysis) is smooth, as expected, the final decision module accepts it as a TS, otherwise it is discarded. 6. Results.…”

“…Particularly, in the field of traffic sign detection, 3D analysis has been proposed in several works, although very few works have been proposed to implement pure 3D-based solutions. For instance, in [5] it is utilized to confirm detected traffic signs, and in [6] to help classification using GPS combined with the analysis of the road sign reflectance evolution. However, these approaches suffer from sensitivity to outliers in the 3D reconstruction, and do not take full advantage of the estimated 3D information.…”

Spatial video and GIS

“…Spatial video is a specific extension of these video formats where spatial attributes are applied to some or all of the images/frames within the captured sequence. Multiple examples of this type of spatial data exist in various formats and modes of capture, storage and use (Livingstone et al 1999, Rzhanov et al 2000, McLoughlin et al 2005, Foy et al 2006, Ó Riain and McCarthy 2006. In general terms, spatial attributes can include any number of different descriptors that can help define a video's image/frame location, time, altitude, orientation or other spatial attribute.…”

Efficient hybrid monocular-stereo approach to on-board video-based traffic sign detection and tracking

“…As a result, they often outdo monocular system in terms of detection performance. Particularly, in the field of road sign detection, the use 3D information from stereo systems has been proposed in several works, e.g., in [9] it is utilized to confirm detected traffic signs, and in [8] to help classification using GPS combined with the analysis of the road sign reflectance evolution. However, these approaches suffer from sensitivity to outliers in the 3D reconstruction, and do not take full advantage of the estimated 3D models.…”