Ruyi Jiang

General traffic sign recognition by feature matching

Traffic sign recognition is a technology which allows us to recognize signs in real time, typically in videos, or sometimes just (off-line) in photos. It is used for Driver Assistance Systems (DAS), road surveys, or the management of road assets (to improve road safety). In this paper, we propose a method for general traffic sign recognition (tested for the New Zealand road signs) which combines previously designed steps, but with an overall adaptation towards general traffic sign recognition (i.e., not just speed or stop signs). First, color input images or frames are converted from RGB color space into HSV color space. Second, special shapes as potential signs are detected (circles, triangles, squares) using Hough transform. Third, potential signs are compared with the template signs as given in the database by using feature matching methods (SIFT or SURF features). At the end, we recognize the traffic sign in an image aiming at realtime DAS. Experiments show that the proposed method is robust for the selected test data, with over 95 percent success rate on average. On a single frame of size 1024 × 768, the system uses on average 80 ms for preprocessing, and 100 ms for matching a traffic sign candidate.

New Lane Model and Distance Transform for Lane Detection and Tracking

Particle filtering of boundary points is a robust way to estimate lanes. This paper introduces a new lane model in correspondence to this particle filter-based approach, which is flexible to detect all kinds of lanes. A modified version of an Euclidean distance transform is applied to an edge map of a road image from a birds-eye view to provide information for boundary point detection. An efficient lane tracking method is also discussed. The use of this distance transform exploits useful information in lane detection situations, and greatly facilitates the initialization of the particle filter, as well as lane tracking. Finally, the paper validates the algorithm with experimental evidence for lane detection and tracking.

Lane Detection on the iPhone

A robust and efficient lane detection system is an essential component of Lane Departure Warning Systems, which are commonly used in many vision-based Driver Assistance Systems (DAS) in intelligent transportation. Various computation platforms have been proposed in the past few years for the implementation of driver assistance systems (e.g., PC, laptop, integrated chips, PlayStation, and so on). In this paper, we propose a new platform for the implementation of lane detection, which is based on a mobile phone (the iPhone). Due to physical limitations of the iPhone w.r.t. memory and computing power, a simple and efficient lane detection algorithm using a Hough transform is developed and implemented on the iPhone, as existing algorithms developed based on the PC platform are not suitable for mobile phone devices (currently). Experiments of the lane detection algorithm are made both on PC and on iPhone.

Low-Level Image Processing for Lane Detection and Tracking

Lane detection and tracking is a significant component of vision-based driver assistance systems (DAS). Low-level image processing is the first step in such a component. This paper suggests three useful techniques for low-level image processing in lane detection situations: bird’s-eye view mapping, a specialized edge detection method, and the distance transform. The first two techniques have been widely used in DAS, while the distance transform is a method newly exploited in DAS, that can provide useful information in lane detection situations. This paper recalls two methods to generate a bird’s-eye image from the original input image, it also compares edge detectors. A modified version of the Euclidean distance transform called real orientation distance transform (RODT) is proposed. Finally, the paper discusses experiments on lane detection and tracking using these technologies.

Modeling of Unbounded Long-Range Drift in Visual Odometry

Visual odometry is a new navigation technology using video data. For long-range navigation, an intrinsic problem of visual odometry is the appearance of drift. The drift is caused by error accumulation, as visual odometry is based on relative measurements, and will grow unboundedly with time. The paper first reviews algorithms which adopt various methods to suppress this drift. However, as far as we know, no work has been done to statistically model and analyze the intrinsic properties of this drift. This paper uses an unbounded system model to represent the drift behavior of visual odometry. The model is composed of an unbounded deterministic part with unknown constant parameters, and a first-order Gauss-Markov process. A simple scheme is given to identify the unknown parameters as well as the statistics of the stochastic part from experimental data. Experiments and discussions are also provided.

CORRIDOR DETECTION AND TRACKING FOR VISION-BASED DRIVER ASSISTANCE SYSTEM

A significant component of driver assistance systems (DAS) is lane detection, and has been studied since the 1990s. However, improving and generalizing lane detection solutions proved to be a challenging task until recently. A (physical) lane is defined by road boundaries or various kinds of lane marks, and this is only partially applicable for modeling the space an ego-vehicle is able to drive in. This paper proposes a concept of (virtual) corridor for modeling this space. A corridor depends on information available about the motion of the ego-vehicle, as well as about the (physical) lane. This paper also suggests a modified version of Euclidean Distance Transform (EDT), named Row Orientation Distance Transform (RODT), to facilitate the detection of corridor boundary points. Then, boundary selection and road patch extension are applied as post-processing. Moreover, this paper also informs about the possible application of corridor for driver assistance. Finally, experiments using images from highways and urban roads with some challenging road situations are presented, illustrating the effectiveness of the proposed corridor detection algorithm. Comparison of lane and corridor on a public dataset is also provided.

Statistical modeling of long-range drift in visual odometry

An intrinsic problem of visual odometry is its drift in longrange navigation. The drift is caused by error accumulation, as visual odometry is based on relative measurements. The paper reviews algorithms that adopt various methods to minimize this drift. However, as far as we know, no work has been done to statistically model and analyze the intrinsic properties of this drift. Moreover, the quantification of drift using offset ratio has its drawbacks. This paper models the drift as a combination of wide-band noise and a first-order Gauss-Markov process, and analyzes it using Allan variance. The models parameters are identified by a statistical method. A novel drift quantification method using Monte Carlo simulation is also provided.

Discrete Driver Assistance

Applying computer technology, such as computer vision in driver assistance, implies that processes and data are modeled as being discretized rather than being continuous. The area of stereo vision provides various examples how concepts known in discrete mathematics (e.g., pixel adjacency graphs, belief propagation, dynamic programming, max-flow/min-cut, or digital straight lines) are applied when aiming for efficient and accurate pixel correspondence solutions. The paper reviews such developments for a reader in discrete mathematics who is interested in applied research (in particular, in vision-based driver assistance). As a second subject, the paper also discusses lane detection and tracking, which is a particular task in driver assistance; recently the Euclidean distance transform proved to be a very appropriate tool for obtaining a fairly robust solution.

Ego-Vehicle Corridors for Vision-Based Driver Assistance

Improving or generalizing lane detection solutions on curved roads with possibly broken lane marks is still a challenging task. This paper proposes a concept of a (virtual) corridor for modeling the space an ego-vehicle is able to drive through, using available (but often incomplete, e.g., due to occlusion, road conditions, or road intersections) information about the lane marks but also about the motion and relative position (with respect to the road) of the ego-vehicle. A corridor is defined in this paper by special features, such as two fixed starting points, a constant width, and a unique relationship with visible lane marks. Robust corridor detection is possible by hypothesis testing based on maximum a posterior (MAP) estimation, followed by boundary selection, and road patch extension. Obstacles are explicitly considered. A corridor tracking method is also discussed. Experimental results are provided.

Options in using graphics for evaluating correspondence algorithms

The paper discusses two options for evaluating the performance of stereo or motion correspondence algorithms in the context of vision-based driver assistance systems (i.e., in highly dynamic large-scale scenes). Assessing the significance of an evaluation scheme (which is always domain specific) is an issue not yet widely addressed by the research community. One option consists in modeling an existing environment and in identifying accurately the trajectory of the ego-vehicle carrying those cameras which capture sequences for correspondence analysis. Another option is given by physics-based rendering of a realistic 3D model of road scenario dynamics. The paper provides a discussion of both options.