Su Jianbo

Fast quality assessment of face images for face recognition

The captured face images are of all kinds of quality in a face recognition system, some poor quality images, for example blurry images, could degrade the performance of the face recognition system. It is obliged to assess the quality of the acquired images in order to eliminate the bad effects result from poor quality images. The current techniques of image quality assessment are not suitable for the quality assessment of face images. This paper analyzes the characters of face recognition and then proposes a quantitative clarity assessment method in detail. The experimental results show that the proposed method can assess the quality of the face images accurately and quickly. The method manages to satisfy the real-time requirement and can be adopted as a preprocessing step in the automatic face recognition system conveniently.

Homography-based correspondence in weakly calibrated curved surface environment and its error analysis

Homography can be computed from four or more corresponding feature pairs in the polyhedral environment. For a stereo image pair that is weakly calibrated, the homography can be estimated via three corresponding feature pairs in the image pair together with the pair of the epipoles. That makes possible to estimate homography in curved surface environment. Since the estimation procedure does not introduce redundant information, it is error-prone. This paper analyzes the origins that produce the errors, which are classified as scene-related origin and calculation-related origin, and differentiates the errors into two types, which are respectively defined as geometric error and algebraic error. The criterion to separate these two kinds of errors is also given. Algorithm to estimate the homography is then improved in its accuracy based on the error analysis. A matching-point accumulation procedure is proposed based on the prediction of the possible matching position from the estimated homography. Feasibility of the proposed algorithm is also discussed. Extensive simulation and experiments are given to illustrate the performances of the new algorithms.

Feature point matching of curved surface and robust uncertainty

Stereo matching is an important research area in stereovision and stereo matching of curved surface is especially crucial. A novel correspondence algorithm is presented and its matching uncertainty is computed robustly for feature points of curved surface. The corners are matched by using homography constraint besides epipolar constraint to solve the occlusion problem. The uncertainty sources are analyzed. A cost function is established and acts as an optimal rule to compute the matching uncertainty. An adaptive scheme Gauss weights are put forward to make the matching results robust to noises. It makes the practical application of corner matching possible. From the experimental results of an image pair of curved surface it is shown that computing uncertainty robustly can restrain the affection caused by noises to the matching precision.

A new plan and coordination strategy for robot system

This paper proposes a general plan and coordination strategy for robot system. The state space for robot system is constructed according to the task requirement and system characteristic. Reachable state of the system is figured out by the systempsilas internal and external constraints. Task plan and coordination is then transformed as trajectory solving problem in its state space, by which the realizable conditions for given task are discussed. If the task is realizable, the optimal strategy for task execution could be obtained by searching in state space. Otherwise, it could be transformed to be realizable via adjusting systempsilas configuration and/or task constraint, and the transformation condition could also be determined. This contributes to design, plan, and coordination of the robotic tasks. Experiments of manipulator path planning and multi-robot formation movement are conducted to show the validity and generalization of the proposed method.

Feature matching based on geometric constraints in weakly calibrated stereo views of curved scenes

The identification of the correspondences of points of views is an important task. A new feature matching algorithm for weakly calibrated stereo images of curved scenes is proposed, based on mere geometric constraints. After initial correspondences are built via the epipolar constraint, many point-to-point image mappings called homographies are set up to predict the matching position for feature points. To refine the predictions and reject false correspondences, four schemes are proposed. Extensive experiments on simulated data as well as on real images of scenes of variant depths show that the proposed method is effective and robust.

New coordination scheme for multi-robot systems based on state space models

Abstract A new coordination scheme for multi-robot systems is proposed. A state space model of the multi-robot system is defined and constructed in which the systems initial and goal states are included along with thetask definition and the systems internal and external constraints. Task accomplishment is considered a transitionof the system state in its state space (SS) under the systems constraints. Therefore, if there exists a connectablepath within reachable area of the SS from the initial state to the goal state, the task is realizable. The optimalstrategy for the task realization under constraints is investigated and reached by searching for the optimal statetransition trajectory of the robot system in the SS. Moreover, if there is no connectable path, which means the taskcannot be performed successfully, the task could be transformed to be realizable by making the initial state and thegoal state connectable and finding a path connecting them in the systems SS. This might be done via adjusting thesystems configuration and/or task constraints. Experiments of multi-robot formation control with obstacles in theenvironment are conducted and simulation results show the validity of the proposed method.

Performance Investigation and Improvement for the Uncalibrated Hand-Eye Coordination System

The paper studies performance of the uncalibrated hand-eye coordination system, whose image Jacobian matrix is estimated by Neural Network way. Since the training conditions of NN are far from that of the practical working conditions, there are some unmodelled dynamics to be compensated in system controller designing. The discrete model of the system is set up, in accordance with the system dynamics and visual information processing, so that the systems performance and stability are well investigated, which leads to improved control method. Simulation results validate the effectiveness of the proposed method.