Detecting and Tracking Moving Targets on Omnidirectional Vision

A method was presented to implement the detecting and tracking of moving targets through omnidirec-tional vision. The method combined optical flow with particle filter arithmetic, in which optical flow field was used to detect and locate moving targets and particle filter was used to track the detected moving objects. According to the circular image character of omnidirectional vision, the calculation equation of optical flow field and the tracking arithmetic of particle filter were improved based on the polar coordinates at the omnidirectional center. The edge of a randomly moving object could be detected by optical flow field and was surrounded by a reference region in the particle filter. A dynamic motion model was established to predict particle state. Histograms were used as the fea-tures in the reference region and candidate regions. The mutual information (MI) and Gaussian function were com-bined to calculate particle weights. Finally, the state of tracked object was computed by the total particle states with weights. Experiment results show that the proposed method could detect and track moving objects with better real-time performance and accuracy.

Helicopter blades running elevation measurement using omnidirectional vision

Omnidirectional dynamic space parameters of high-speed rotating helicopter blades are precise 3 D vector description of the blades. In particular, the elevation difference is directly related to the aerodynamic performance and maneuverability of the helicopter. The state of the art detection techniques based on optics and common vision have several drawbacks, such as high demands on devices but poor extensibility, limited measurement range and fixed measurement position. In this paper, a novel approach of helicopter blades running elevation measurement is proposed based on omnidirectional vision. With the advantages of panoramic visual imaging integration, 360° field of view and rotation in-variance, high-resolution images of all rotating blades positions are obtained at one time. By studying the non-linear calibration and calculation model of omnidirectional vision system, aiming at solving the problem of inaccurate visual space mapping model,the omnidirectional and full-scale measurement of the elevation difference are finalized. Experiments are carried out on our multifunctional simulation blades test system and the practical blades test tower, respectively. The experimental results demonstrate the effectiveness of the proposed method and show that the proposed method can considerably reduce the complexity of measurement.

An object tracking and global localization method using the cylindrical projection of omnidirectional image

We present an omnidirectional vision system we have implemented to provide our mobile robot with a fast tracking and robust localization capability. An algorithm is proposed to do reconstruction of the environment from the omnidirectional image and global localization of the robot in the context of the Middle Size League RoboCup field. This is accomplished by learning a set of visual landmarks such as the goals and the corner posts. Due to the dynamic changing environment and the partially observable landmarks, four localization cases are discussed in order to get robust localization performance. Localization is performed using a method that matches the observed landmarks, i.e. color blobs, which are extracted from the environment. The advantages of the cylindrical projection are discussed giving special consideration to the characteristics of the visual landmark and the meaning of the blob extraction. The analysis is established based on real time experiments with our omnidirectional vision system and the actual mobile robot. The comparative studies are presented and the feasibility of the method is shown.

Unwrapping and stereo rectification for omnidirectional images

Omnidirectional imaging sensors have been used in more and more applications when a very large field of view is required.In this paper,we investigate the unwrapping,epipolar geometry and stereo rectification issues for omnidirectional vision when the particular mirror model and the camera parameters are unknown in priori.First,the omnidirectional camera is calibrated under the Taylor model,and the parameters related to this model are obtained.In order to make the classical computer vision algorithms of conventional perspective cameras applicable,the ring omnidirectional image is unwrapped into two kinds of panoramas:cylinder and cuboid.Then the epipolar geometry of arbitrary camera configuration is analyzed and the essential matrix is deduced with its properties being indicated for ring images.After that,a simple stereo rectification method based on the essential matrix and the conformal mapping is proposed.Simulations and real data experimental results illustrate that our methods are effective for the omnidirectional camera under the constraint of a single view point.