Efficient road-lane detection is expected to be achievable by application of the Hough transform (HT) which realizes high-accuracy straight-line extraction from images. The main challenge for HT-hardware implementation in actual applications is the trade-off optimization between accuracy maximization, power-dissipation reduction and real-time requirements. We report a HT-hardware architecture for road-lane detection with parallelized voting procedure, local maximum algorithm and FPGA-prototype implementation. Parallelization of the global design is realized on the basis of θ-value discretization in the Hough space. Four major hardware modules are developed for edge detection in the original video frames, computation of the characteristic edge-pixel values (ρ,θ) in Hough-space, voting procedure for each (ρ,θ) pair with parallel local-maximum-based peak voting-point extraction in Hough space to determine the detected straight lines. Implementation of a prototype system for real-time road-lane detection on a low-cost DE1 platform with a Cyclone II FPGA device was verified to be possible. An average detection speed of 135 frames/s for VGA (640x480)-frames was achieved at 50 MHz working frequency.

Lane detection or road detection is one of the key features of autonomous driving. In computer vision area, it is still a very challenging target since there are various types of road scenarios which require a very high robustness of the algorithm. And considering the rather high speed of the vehicles, high efficiency is also a very important requirement for practicable application of autonomous driving. In this paper, we propose a deep convolution neural network based lane detection method, which consider the lane detection task as a pixel level segmentation of the lane markings. We also propose an automatic training data generating method, which can significantly reduce the effort of the training phase. Experiment proves that our method can achieve high accuracy for various road scenes in real-time.

This letter describes the development and implementation of the lane detection system accelerated by the neuromorphic hardware. Because the neuromorphic hardware has inherently parallel nature and has constant output latency regardless the size of the knowledge, the proposed lane detection system can recognize various types of lanes quickly and efficiently. Experimental results using the road images obtained in the actual driving environments showed that white and yellow lanes could be detected with an accuracy of more than 94 percent.

In this paper, a novel approach is proposed for stereo vision-based ground plane detection at superpixel-level, which is implemented by employing a Disparity Texture Map in a convolution neural network architecture. In particular, the Disparity Texture Map is calculated with a new Local Disparity Texture Descriptor (LDTD). The experimental results demonstrate our superior performance in KITTI dataset.

Lane detection plays an important role in Driver Assistance Systems and Autonomous Vehicle System. In this paper, we propose a parallel-snake model combined with balloon force for lane detection. Parallel-snake is defined as two open active contours with parallel constrain. The lane boundaries on the left and right sides are assumed as parallel curves, parallel-snake is deformed to estimate these two boundaries. As lane regions between left and right boundaries usually have low gradient, snake will lose external force on these regions. Furthermore, inspired by balloon active contour model, the balloon force is introduced into parallel-snake to expand two parallel curves from center of road to the left and right lane boundaries. Different from closed active contour, stretching force is adopted to prevent the head and tail of snake from converging together. The experimental results on three different datasets show that parallel-snake model can work well on images with shadows and handle the lane with broken boundaries as the parallel property.

This paper describes a cumulative distribution function (CDF) of edge direction for detecting road lanes. Based on the assumptions that there are no abrupt changes in the direction and location of road lanes and that the intensity of lane boundaries differs from that of the background, the CDF is formulated, which accumulates the edge magnitude for edge directions. The CDF has distinctive peak points at the vicinity of lane directions due to the directional and the positional continuities of a lane. To obtain lane-related information, we construct a scatter diagram by collecting edge pixels, of which the direction corresponds to the peak point of the CDF, then perform the principal axis-based line fitting for the scatter diagram. Because noises can cause many similar features appear or disappear in an image, to prevent false alarms or miss detection, a recursive estimator of the CDF was introduced, and also a scene understanding index (SUI) was formulated by the statistical parameters of the CDF. The proposed algorithm has been implemented in real time on video data obtained from a test vehicle driven on a typical highway.