Yongjian He

  • Citations Per Year
Learn More
Prevent collision with other vehicles is crucial for developing advanced driver assistance systems. Vision-based approaches for vehicle detection attract more attention than those using other sensors. In this study, we address the problem of detecting front vehicles in still images. Unlike traditional methods which mainly based on the holistic appearance of(More)
This paper presents an central difference Kalman filter (CDKF) based Simultaneous Localization and Mapping (SLAM) algorithm, which is an alternative to the classical extended Kalman filter based SLAM solution (EKF-SLAM). EKF-SLAM suffers from two important problems, which are the calculation of Jacobians and the linear approximations to the nonlinear(More)
In this paper, we consider the problem of estimating the pose of a driver from video data. We propose to use isophote features to improve classification performance when illumination varies. In particular, we both use the direction and curvature features. The features are encoded into direction histogram and curvature histogram in the blocks of the image.(More)
In this paper, we propose a novel definition of opposite path. Its core feature is that the sequence of candidate paths and the distances between adjacent nodes in the tour are considered simultaneously. In a sense, the candidate path and its corresponding opposite path have the same (or similar at least) distance to the optimal path in the current(More)
An novel method using local convexity and octree is proposed to segment irregular 3D point cloud into surfaces. An modified octree method is used to produce the init facet set which usually contains too many facets and is meaningless, and then the connection of the leaf nodes is found. Finally, the adjacent facets are combined if they satisfied the local(More)
Image identification is a new technology in these two decades. This main content of this technology is to study the classification and description of some objects or processes (image in general), while the reasonable application of enhancement algorithm and feature location technology of the image target can effectively increase the accuracy and rate of the(More)
This paper addresses the problem of detecting on-road vehicles in still images captured by the on-board cameras. We model this as a labelling inference procedure and incorporate the part-based representation of the rear-ends of vehicle within a hidden random field based probabilistic model. Representing objects with parts inherently good for dealing with(More)