Multi‐lane detection based on omnidirectional camera using anisotropic steerable filters

Abstract

Automated lane detection is a vital part of driver assistance systems in intelligent vehicles. In this study, a multilane detection method based on omnidirectional images is presented to conquer the difficulties stemming from the limited view field of the rectilinear cameras. The contributions of this study are twofold. First, to extract the features of the lane markings under various illumination and road-surface scenarios, a feature extractor based on anisotropic steerable filter is proposed. Second, a parabola lane model is used to fit the straight as well as curved lanes. According to the parabola lane model, the straight lines and curves of feature maps can be represented as straight lines in a linear space coordinate system. Then lane modelling can be treated as an optimisation question in linear space and the parameters of lanes can be estimated by minimising the objection function. The method has been tested on publicly available data sets and the real car experiments. Experimental results show that the proposed method outperforms state-of-the-arts approaches and obtains a detection accuracy of 99% in real world scenes.

Vision-based curved lane keeping control for intelligent vehicle highway system

“…, 5, are regarded as the polynomial coefficients. The steering reference input d ref is parameterized by solving equation (21). For practical applications, only solutions in the range ( À p=3), (p=3) ½ are retained.…”

“…Among the strategies, some are literally based on machine vision. The authors in Li et al 21 proposed a methodology for multi-lane detection using an omnidirectional camera, in order to overcome the difficulties encountered while using rectilinear cameras. The extracted characteristic points of road marking are ensured under different illumination conditions and various road types.…”

Precomputation‐based radix‐4 CORDIC for approximate rotations and Hough transform

“…The proposed approach uses radix-4 for i = 1, 2 and 3 alone. For these choices of i, we have (4)- (6), respectively (where angles are expressed in radians) [14]…”

“…Many applications (involving vector rotation) run on platforms with constraints on area and power consumption. Examples of this include the computation of DCT [2] for image compression applications, hardware realisation of a multi-input multi-output orthogonal frequency division multiplexing receiver [3], adaptive beamforming using QR factorisation [4] and image transformbased lane detection [5][6][7]. Keeping these applications in view, this paper presents a high-performance approximate rotation algorithm based on the coordinate rotation digital computer (CORDIC) approach [8][9][10].…”

“…An application of the proposed CORDIC for Hough transform computation is then presented. Hough transforms are valuable in the context of lane detection [5][6][7] for intelligent transportation applications. Real-time detection of the lane is a very crucial task.…”