Showing papers by "Shahriar Negahdaripour published in 2006"

An ROV Stereovision System for Ship-Hull Inspection

Abstract: Ship hulls, as well as bridges, port dock pilings, dams, and various underwater structures need to be inspected for periodic maintenance. Additionally, there is a critical need to provide protection against sabotage activities, and to establish effective countermeasures against illegal smuggling activities. Unmanned underwater vehicles are suitable platforms for the development of automated inspection systems, but require integration with appropriate sensor technologies. This paper describes a vision system for automated ship-hull inspection, based on computing the necessary information for positioning, navigation, and mapping of the hull from stereo images. Binocular cues are critical in resolving a number of complex visual artifacts that hamper monocular vision in shallow-water conditions. Furthermore, they simplify the estimation of vehicle pose and motion, which is fundamental for successful automatic operation. The system has been implemented on a commercial remotely operated vehicle (ROV), and tested in pool and dock tests. Results from various trials are presented to demonstrate the system capabilities

Integrated System for Robust 6-DOF Positioning Utilizing New Closed-Form Visual Motion Estimation Methods in Planar Terrains

Abstract: Estimating the relative positions and (or) trajectory of a camera from video images is a fundamental problem in motion vision. Of special relevance is the closed-form solution for planar scenes, for processing fly-over imagery from airborne and underwater robotics platforms, automated airplane landing utilizing runway landmarks, photomosaicing, etc. However, the method's robustness can break down in certain scenarios, e.g., due to inherent translation-rotation ambiguity of visual motion with short baselines and narrow field of view. The robustness can be improved by devising methods that compute a smaller set of motion parameters, utilizing other sensors to measure the remaining components. This paper addressed key issues in six degrees of freedom positioning from fly-over imagery by integrating vision with rotational angle sensors. First, we propose and utilize robust closed-form solutions for estimating the motion and orientation of a planar surface from the image flow variations up to first order, given measurements of pitch and roll motions. We also describe a calibration technique to enable the integration of angle sensor and visual measurements. Next, an error analysis enables us to evaluate the impact of inaccurate pitch and roll measurements on the estimates from the new closed-form solutions. Finally, the performance of our new methods and the integrated positioning system are evaluated in various experiments with synthetic and real data

Direct and Indirect 3-D Reconstruction from Opti-Acoustic Stereo Imaging

Abstract: Utilization of an acoustic camera for range measurements is a significant advantage for 3-D shape recovery of underwater targets by opti-acoustic stereo imaging, where the associated epipolar geometry of visual and acoustic image correspondences is described in terms of conic sections and trigonometric functions. In this paper, we propose and analyze a number of methods based on direct and indirect approaches that provide insight on the merits of the new imaging and 3-D object reconstruction paradigm. We have devised certain indirect methods, built on a regularization formulation, to first compute from noisy correspondences maximum likelihood estimates that satisfy the epipolar geometry. The 3-D target points can then be determined from a number of closed-form solutions applied to these ML estimates. An alternative direct approach is also presented for 3-D reocnstruction directly from noisy correspondences. Computer simulations verify consistency between the analytical and experimental reconstruction SNRs -- the criterion applied in performance assessment of these various solutions.

EKF-Based Recursive Dual Estimation of Structure & Motion from Stereo Data

Abstract: Extended Kalman filters (EKF) have been proposed to estimate ego-motion and to recursively update scene structure in the form of 3-D positions of selected prominent features from motion and stereo sequences. Previous methods typically accommodate no more than a few dozen features for real-time processing. To maintain motion estimation accuracy, this calls for high contrast images to compute image feature locations with precision. Within manmade environments, various prominent corner points exist that can be extracted and tracked with required accuracy. However, prominent features are more difficult to localize precisely in natural scenes. Statistically, more feature points become necessary to maintain the same level of motion estimation accuracy and robustness. However, this imposes a computational burden beyond the capability of EKF-based techniques for real-time processing. A sequential dual EKF estimator utilizing stereo data is proposed for improved computation efficiency. Two important issues, unbiased estimation and stochastic stability are addressed. Furthermore, the dynamic feature set is handled in a more effective, efficient and robust way. Experimental results to demonstrate the merits of the new theoretical and algorithmic developments are presented.

Integrating BC & GC Models In Computing Stereo Disparity As Markov Random Field

Abstract: Belief propagation and graph cuts have emerged as powerful tools for computing efficient approximate so- lution to stereo disparity field modelled as the Markov random field (MRF). These algorithms have provided the best performance based on results on a standard data set (1). However, employment of the brightness constancy (BC) assumption severely limits the range of their applications. Previously, augmenting the BC with gradient constancy (GC) assumption has shown to produce a more robust optical flow algorithm (2), (3). In this paper, these constraints are integrated within the MRF framework to devise an en- hanced global method that broadens the application domains for stereo computation. Results of experiments with both semi-synthetic data and more challenging ocean images are presented to illustrate that the proposed method generally outperforms earlier dense optical flow and stereo algorithms.