Qingzhou Mao

CrackTree: Automatic crack detection from pavement images

Pavement cracks are important information for evaluating the road condition and conducting the necessary road maintenance. In this paper, we develop CrackTree, a fully-automatic method to detect cracks from pavement images. In practice, crack detection is a very challenging problem because of (1) low contrast between cracks and the surrounding pavement, (2) intensity inhomogeneity along the cracks, and (3) possible shadows with similar intensity to the cracks. To address these problems, the proposed method consists of three steps. First, we develop a geodesic shadow-removal algorithm to remove the pavement shadows while preserving the cracks. Second, we build a crack probability map using tensor voting, which enhances the connection of the crack fragments with good proximity and curve continuity. Finally, we sample a set of crack seeds from the crack probability map, represent these seeds by a graph model, derive minimum spanning trees from this graph, and conduct recursive tree-edge pruning to identify desirable cracks. We evaluate the proposed method on a collection of 206 real pavement images and the experimental results show that the proposed method achieves a better performance than several existing methods.

Mining time-dependent attractive areas and movement patterns from taxi trajectory data

Mining attractive areas that people interested in and their related movement patterns can lead to instructive insight to transport management, urban planning and location-based services (LBS). The number of visiting that it attracts is used in this paper to measure an areas level of attractiveness (LoA). As one of the most widely used mode of transport, taxi can tell a lot of stories about not only road network traffic condition, but also areas people interested in crossing a day and their related travel patterns, such as travel destination and average travel distance. Conventional taxi trajectory analysis, or more generally, probe vehicle and floating car trajectory analysis, more focuses on road network travel time and average speed estimation. This study from another angle, uses taxi trajectory data to discover attractive areas that people often visit, for instance, hot shopping and leisure places or living and working areas based on their LoA which hereby is represented as the frequency and density of passenger pick-up and drop-off points, because each point represents a certain scope where attractiveness generates. To obtain meaningful patterns, clustering approach is used to group spatiotemporally similar pick-up and drop-off points, because peoples interests to these areas varies through time of the day, day of the week, even season of the year. Moreover, a time-dependent travel flow interaction matrix is established, which is a variation of O-D (Origin-Destination) matrix used in transport domain, and can be used to better understand movement patterns by quantizing the attractiveness among clusters. Background geographic information is used to facilitate the understanding of the movement. This study represents a novel application of taxi trajectory data, reveals peoples travel demand and movement patterns in a more deep sense to serve transport management, urban planning, as well as spatiotemporal-tailored location search and services.

FoSA: F* Seed-growing Approach for crack-line detection from pavement images

Most existing approaches for pavement crack line detection implicitly assume that pavement cracks in images are with high contrast and good continuity. This assumption does not hold in pavement distress detection practice, where pavement cracks are often blurry and discontinuous due to particle materials of road surface, crack degradation, and unreliable crack shadows. To this end, we propose in this paper FoSA - F* Seed-growing Approach for automatic crack-line detection, which extends the F* algorithm in two aspects. It exploits a seed-growing strategy to remove the requirement that the start and end points should be set in advance. Moreover, it narrows the global searching space to the interested local space to improve its efficiency. Empirical study demonstrates the correctness, completeness and efficiency of FoSA.

ALIMC: Activity Landmark-Based Indoor Mapping via Crowdsourcing

Indoor maps are integral to pedestrian navigation systems, an essential element of intelligent transportation systems (ITS). In this paper, we propose ALIMC, i.e., Activity Landmark-based Indoor Mapping system via Crowdsourcing. ALIMC can automatically construct indoor maps for anonymous buildings without any prior knowledge using crowdsourcing data collected by smartphones. ALIMC abstracts the indoor map using a link-node model in which the pathways are the links and the intersections of the pathways are the nodes, such as corners, elevators, and stairs. When passing through the nodes, pedestrians do the corresponding activities, which are detected by smartphones. After activity detection, ALIMC extracts the activity landmarks from the crowdsourcing data and clusters the activity landmarks into different clusters, each of which is treated as a node of the indoor map. ALIMC then estimates the relative distances between all the nodes and obtains a distance matrix. Based on the distance matrix, ALIMC generates a relative indoor map using the multidimensional scaling technique. Finally, ALIMC converts the relative indoor map into an absolute one based on several reference points. To evaluate ALIMC, we implement ALIMC in an office building. Experiment results show that the 80th percentile error of the mapping accuracy is about 0.8-1.5 m.

A Closed-Loop Speed Advisory Model With Driver's Behavior Adaptability for Eco-Driving

Providing drivers with speed advisories is an effective eco-driving method at signalized intersections. However, all current research on speed advisory models has excluded the drivers behavior factor. In this paper, we focus on developing a speed advisory model that is able to adapt to the drivers behavior for eco-driving. First, we propose a closed-loop speed advisory framework, with simulation results to show that the current model could not fit in the closed-loop implementation. Next, the continuous acceleration with explicit high velocity boundary (CAEHV) model is established to address the issues when the existing model is used. However, the simulation results for the CAEHV model are not fully satisfactory due to the existence of oscillations in actual speed trajectories. Third, the CAEHV with coasting (CAEHV-C) model is established, in which the vehicle coasting is applied to supplement cruising to avoid oscillations. Simulation results show that the fuel economy performance of the CAEHV-C model is improved by 4% when compared with the CAEHV model. It also shows that CAEHV-C performs the best in terms of the drivers behavior adaptability.

Multiple Vehicle-Like Target Tracking Based on the Velodyne Lidar

Abstract This paper proposes a novel multiple vehicle-like target tracking method based on a Velodyne HDL64E light detection and ranging (LiDAR) system. The proposed method combines multiple hypothesis tracking (MHT) algorithm with dynamic point cloud registration (DPCR), which is able to solve the multiple vehicle-like target tracking in highly dynamic urban environments without any auxiliary information from GPS or IMU. Specifically, to track targets consistently, the DPCR is developed to calculate accurately the pose of the ego-vehicle for the transformation of raw measurements taken in the moving coordinate systems into a static absolute coordinate system; while in turn, MHT helps to improve the performance of DPCR by discriminating and removing the dynamic points from the scene. Furthermore, the proposed MHT method is also able to solve the occlusion problem existing in the point cloud. Experiments on sets of urban environments prove that the presented method is effective and robust, even in highly dynamic environments.

A Robust Crowdsourcing-Based Indoor Localization System

WiFi fingerprinting-based indoor localization has been widely used due to its simplicity and can be implemented on the smartphones. The major drawback of WiFi fingerprinting is that the radio map construction is very labor-intensive and time-consuming. Another drawback of WiFi fingerprinting is the Received Signal Strength (RSS) variance problem, caused by environmental changes and device diversity. RSS variance severely degrades the localization accuracy. In this paper, we propose a robust crowdsourcing-based indoor localization system (RCILS). RCILS can automatically construct the radio map using crowdsourcing data collected by smartphones. RCILS abstracts the indoor map as the semantics graph in which the edges are the possible user paths and the vertexes are the location where users may take special activities. RCILS extracts the activity sequence contained in the trajectories by activity detection and pedestrian dead-reckoning. Based on the semantics graph and activity sequence, crowdsourcing trajectories can be located and a radio map is constructed based on the localization results. For the RSS variance problem, RCILS uses the trajectory fingerprint model for indoor localization. During online localization, RCILS obtains an RSS sequence and realizes localization by matching the RSS sequence with the radio map. To evaluate RCILS, we apply RCILS in an office building. Experiment results demonstrate the efficiency and robustness of RCILS.

Efficient and lossless compression of raster maps

In this paper, a block-line-separated encoding approach (BLiSE) is proposed for efficient and lossless compression of raster maps. Firstly, BLiSE separates a raster map into blocks and lines through a raster-map-decomposition algorithm. For each block, BLiSE uses a flag encoding to record the location and color information, and a 4-direction Freeman coding to encode the block boundary. For each line, BLiSE directly applies the 8-direction Freeman coding. With this separate-encoding strategy, BLiSE is highly suitable for compression of navigation maps. We evaluate BLiSE on a navigation-map dataset containing 100 raster maps and compare it with three traditional approaches, which are GIF, PNG and JP2000-LS. The results indicate an average compression ratio of 82.83 of the proposed BLiSE, which is much higher than that of several competing approaches.

Automatic inpainting by removing fence-like structures in RGBD images

Recent inpainting techniques usually require human interactions which are labor intensive and dependent on the user experiences. In this paper, we introduce an automatic inpainting technique to remove undesired fence-like structures from images. Specifically, the proposed technique works on the RGBD images which have recently become cheaper and easier to obtain using the Microsoft Kinect. The basic idea is to segment and remove the undesired fence-like structures by using both depth and color information, and then adapt an existing inpainting algorithm to fill the holes resulting from the structure removal. We found that it is difficult to achieve a satisfactory segmentation of such structures by only using the depth channel. In this paper, we use the depth information to help identify a set of foreground and background strokes, with which we apply a graph-cut algorithm on the color channels to obtain a more accurate segmentation for inpainting. We demonstrate the effectiveness of the proposed technique by experiments on a set of Kinect images.

Block-constraint line scanning method for lane detection

Considering the plentiful road markings in China, we present a Block-Constraint Line scanning (BCLS) method for lane detection in this paper. In this method, images are firstly pre-processed by a morphological top-hat transform, and then an imaging model is created for building relationship between lane parameters of the image coordinate and the WGS coordinate, from which target points on lane lines could be retained by a block-constraint line scanning algorithm. Finally, lanes could be extracted by a Progressive Probabilistic Hough Transform (PPHT) and the number of lanes is figured out through clustering. Our method is fast enough to meet real-time requirement. Experiments were carried out on the intelligent vehicle SmartV (Fig.1) on the Wuhan urban roads in China and the results show that this method can efficiently and accurately extract lanes in complex environments, even with the presence of non-lane road markings.