Rapid vanishing point estimation for general road detection

Abstract

Abstract-This paper deals with fast vanishing point estimation for autonomous robot navigation. Preceding approaches showed suitable results and vanishing point estimation was used in many robotics tasks, especially in the detection of illstructured roads. The main drawback of such approaches is the computational complexity -the possibilities of hardware accelerations are mentioned in many papers, however, we believe, that the biggest benefit of a vanishing point estimation algorithm is for primarily tele-operated robots in the case of signal loss, etc., that cannot use specialized hardware just for this feature. In this paper, we investigate possibilities of an efficient implementation by the expansion of Gabor wavelets into a linear combination of Haar-like box functions to perform fast filtering via integral image trick and discuss the utilization of superpixels in the voting scheme to provide a significant speed-up (more than 40 times), while we loose only 3-5% in precision.

Fast Vanishing Point Estimation Based on Particle Swarm Optimization

“…Consequently, methods on improving the efficiency of the vanishing point estimation have been proposed. Miksik reduced the vanishing point candidates in voting by using super pixels [8]. This method can improve the efficiency of Kong's method by more than 40 times.…”

“…Different methods were proposed for estimating texture orientation. C. Russmusen, Kong, Moghadam et al [4], [5], [15] filter banks to estimate it [14]; and Miksik, Wang et al used Haar-like box to do it [8], [11]. Here we mainly follow the Gabor filter based method.…”

“…The local texture orientation θ(p) of an image at pixel p(x, y) describes the strongest local parallel structure or texture flow [8]. Different methods were proposed for estimating texture orientation.…”

Mobile robot navigation system in outdoor pedestrian environment using vision-based road recognition

“…Most systems [11], [12], [13] use the consensus direction of local textures or image edgels (edge pixels) to vote for the most likely VP. However, edgels, because of their limited scope of support, can lead to an unstable result.…”

Real-Time Road-Direction Point Detection in Complex Environment

“…[12] applied only four directions of Gabor filters to estimate dominant orientations which can speed up convolutional processing. [13] developed a model by combining binaryapproximated Gabor filters and a cascaded voting scheme to reduce computational complexity for vanishing point prediction with less time cost. [14] proposed a noise-insensitive observation model through Gabor filters with better performance in terms of accuracy and speed.…”