Fast shape-based road sign detection for a driver assistance system

Abstract

Abstract-A new method is presented for detecting triangular, square and octagonal road signs efficiently and robustly. The method uses the symmetric nature of these shapes, together with the pattern of edge orientations exhibited by equiangular polygons with a known number of sides, to establish possible shape centroid locations in the image. This approach is invariant to in-plane rotation and returns the location and size of the shape detected. Results on still images show a detection rate of over 95%. The method is efficient enough for real-time applications, such as on-board-vehicle sign detection.

Road Sign Detection and Shape Reconstruction Using Gielis Curves

“…It is important to note that the algorithm accurately detects road sign of different sizes in the image. This is a clear advantage over the method developped by (Loy and Barnes, 2004) where different radii are tried in order to find the correct size of road signs.…”

“…Recently, (Loy and Zelinsky, 2002) propose the fast radial symmetry transform which is a technique similar to Hough transform. The method was extended and successfully used to detect regular polygons by (Loy and Barnes, 2004). (Gavrila, 1999) uses a template matching approach for shape detection.…”

Mobile mapping for the automated analysis of road signage and delineation

“…The principal cue used in this work for the detection of road signs is different from the conventional techniques that use either colour or shape as the main cues such as in [14][15][16][17][18]. While colour and shape are the obvious and intuitive approaches to use since road sign manufacturers conform to colour and shape restrictions, they are not without their problems.…”

A Real-Time Road Sign Detection Using Bilateral Chinese Transform

“…Indeed, two additional steps are required: a template matching recognition stage and a temporal filter to validate consistent candidates (which is a way to replace the color information). To cope with the lake of generality of the circular shape, Loy et al [2] developed three specific versions of the RST for octagonal, square and triangular shapes respectively. This proposed approach lacks in generality and was only tested on 15 images for each shape.…”